sha
stringlengths 40
40
| text
stringlengths 1
13.4M
| id
stringlengths 2
117
| tags
listlengths 1
7.91k
| created_at
stringlengths 25
25
| metadata
stringlengths 2
875k
| last_modified
stringlengths 25
25
| arxiv
listlengths 0
25
| languages
listlengths 0
7.91k
| tags_str
stringlengths 17
159k
| text_str
stringlengths 1
447k
| text_lists
listlengths 0
352
| processed_texts
listlengths 1
353
| tokens_length
listlengths 1
353
| input_texts
listlengths 1
40
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
bfd47830889a356e108cb50210f03ce9fbbd48a6
|
# Dataset Card for "med_alpaca_standardized_cluster_96_std"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
AdapterOcean/med_alpaca_standardized_cluster_96_std
|
[
"region:us"
] |
2023-10-24T02:20:02+00:00
|
{"dataset_info": {"features": [{"name": "message", "dtype": "string"}, {"name": "message_type", "dtype": "string"}, {"name": "message_id", "dtype": "int64"}, {"name": "conversation_id", "dtype": "int64"}, {"name": "cluster", "dtype": "float64"}, {"name": "__index_level_0__", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 20049249, "num_examples": 41543}], "download_size": 9724843, "dataset_size": 20049249}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T02:20:06+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "med_alpaca_standardized_cluster_96_std"
More Information needed
|
[
"# Dataset Card for \"med_alpaca_standardized_cluster_96_std\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"med_alpaca_standardized_cluster_96_std\"\n\nMore Information needed"
] |
[
6,
25
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"med_alpaca_standardized_cluster_96_std\"\n\nMore Information needed"
] |
f6d3acb460ca56422af936229d8078432d67998e
|
# Dataset Card for "med_alpaca_standardized_cluster_96_alpaca"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
AdapterOcean/med_alpaca_standardized_cluster_96_alpaca
|
[
"region:us"
] |
2023-10-24T02:20:08+00:00
|
{"dataset_info": {"features": [{"name": "input", "dtype": "string"}, {"name": "output", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 18857563, "num_examples": 13847}], "download_size": 9295048, "dataset_size": 18857563}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T02:20:12+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "med_alpaca_standardized_cluster_96_alpaca"
More Information needed
|
[
"# Dataset Card for \"med_alpaca_standardized_cluster_96_alpaca\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"med_alpaca_standardized_cluster_96_alpaca\"\n\nMore Information needed"
] |
[
6,
25
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"med_alpaca_standardized_cluster_96_alpaca\"\n\nMore Information needed"
] |
63a10bee15e4524fc50224dc190d82206f153900
|
# Dataset Card for "med_alpaca_standardized_cluster_96"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
AdapterOcean/med_alpaca_standardized_cluster_96
|
[
"region:us"
] |
2023-10-24T02:20:12+00:00
|
{"dataset_info": {"features": [{"name": "text", "dtype": "string"}, {"name": "conversation_id", "dtype": "int64"}, {"name": "embedding", "sequence": "float64"}, {"name": "cluster", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 131899589, "num_examples": 13848}], "download_size": 37266810, "dataset_size": 131899589}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T02:20:17+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "med_alpaca_standardized_cluster_96"
More Information needed
|
[
"# Dataset Card for \"med_alpaca_standardized_cluster_96\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"med_alpaca_standardized_cluster_96\"\n\nMore Information needed"
] |
[
6,
22
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"med_alpaca_standardized_cluster_96\"\n\nMore Information needed"
] |
78fbd64741d2b37e67d3101593c6a8634610d0d7
|
# Dataset Card for "med_alpaca_standardized_cluster_97_std"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
AdapterOcean/med_alpaca_standardized_cluster_97_std
|
[
"region:us"
] |
2023-10-24T02:25:21+00:00
|
{"dataset_info": {"features": [{"name": "message", "dtype": "string"}, {"name": "message_type", "dtype": "string"}, {"name": "message_id", "dtype": "int64"}, {"name": "conversation_id", "dtype": "int64"}, {"name": "cluster", "dtype": "float64"}, {"name": "__index_level_0__", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 9037858, "num_examples": 20939}], "download_size": 3142297, "dataset_size": 9037858}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T02:25:24+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "med_alpaca_standardized_cluster_97_std"
More Information needed
|
[
"# Dataset Card for \"med_alpaca_standardized_cluster_97_std\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"med_alpaca_standardized_cluster_97_std\"\n\nMore Information needed"
] |
[
6,
25
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"med_alpaca_standardized_cluster_97_std\"\n\nMore Information needed"
] |
2528732843ffcf027ddadb6202b26e31ed896047
|
# Dataset Card for "med_alpaca_standardized_cluster_97_alpaca"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
AdapterOcean/med_alpaca_standardized_cluster_97_alpaca
|
[
"region:us"
] |
2023-10-24T02:25:26+00:00
|
{"dataset_info": {"features": [{"name": "input", "dtype": "string"}, {"name": "output", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 8390574, "num_examples": 6979}], "download_size": 2842335, "dataset_size": 8390574}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T02:25:29+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "med_alpaca_standardized_cluster_97_alpaca"
More Information needed
|
[
"# Dataset Card for \"med_alpaca_standardized_cluster_97_alpaca\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"med_alpaca_standardized_cluster_97_alpaca\"\n\nMore Information needed"
] |
[
6,
25
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"med_alpaca_standardized_cluster_97_alpaca\"\n\nMore Information needed"
] |
fa703ce0427e979058634c410872d15f1bad8c69
|
# Dataset Card for "med_alpaca_standardized_cluster_97"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
AdapterOcean/med_alpaca_standardized_cluster_97
|
[
"region:us"
] |
2023-10-24T02:25:29+00:00
|
{"dataset_info": {"features": [{"name": "text", "dtype": "string"}, {"name": "conversation_id", "dtype": "int64"}, {"name": "embedding", "sequence": "float64"}, {"name": "cluster", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 65415362, "num_examples": 6980}], "download_size": 17640548, "dataset_size": 65415362}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T02:25:33+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "med_alpaca_standardized_cluster_97"
More Information needed
|
[
"# Dataset Card for \"med_alpaca_standardized_cluster_97\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"med_alpaca_standardized_cluster_97\"\n\nMore Information needed"
] |
[
6,
22
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"med_alpaca_standardized_cluster_97\"\n\nMore Information needed"
] |
3c150cc249cc82b1484d9519fbc78ae948d3cae8
|
# Dataset Card for "med_alpaca_standardized_cluster_98_std"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
AdapterOcean/med_alpaca_standardized_cluster_98_std
|
[
"region:us"
] |
2023-10-24T02:30:33+00:00
|
{"dataset_info": {"features": [{"name": "message", "dtype": "string"}, {"name": "message_type", "dtype": "string"}, {"name": "message_id", "dtype": "int64"}, {"name": "conversation_id", "dtype": "int64"}, {"name": "cluster", "dtype": "float64"}, {"name": "__index_level_0__", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 14102021, "num_examples": 21354}], "download_size": 7281566, "dataset_size": 14102021}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T02:30:37+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "med_alpaca_standardized_cluster_98_std"
More Information needed
|
[
"# Dataset Card for \"med_alpaca_standardized_cluster_98_std\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"med_alpaca_standardized_cluster_98_std\"\n\nMore Information needed"
] |
[
6,
25
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"med_alpaca_standardized_cluster_98_std\"\n\nMore Information needed"
] |
198cb37e1a7e0e84aa4f8938553825d30e81563d
|
# Dataset Card for "med_alpaca_standardized_cluster_98_alpaca"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
AdapterOcean/med_alpaca_standardized_cluster_98_alpaca
|
[
"region:us"
] |
2023-10-24T02:30:38+00:00
|
{"dataset_info": {"features": [{"name": "input", "dtype": "string"}, {"name": "output", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 13488443, "num_examples": 7117}], "download_size": 7232687, "dataset_size": 13488443}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T02:30:42+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "med_alpaca_standardized_cluster_98_alpaca"
More Information needed
|
[
"# Dataset Card for \"med_alpaca_standardized_cluster_98_alpaca\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"med_alpaca_standardized_cluster_98_alpaca\"\n\nMore Information needed"
] |
[
6,
25
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"med_alpaca_standardized_cluster_98_alpaca\"\n\nMore Information needed"
] |
f0291232ae7183c35ab235888916b017e7458c6b
|
# Dataset Card for "med_alpaca_standardized_cluster_98"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
AdapterOcean/med_alpaca_standardized_cluster_98
|
[
"region:us"
] |
2023-10-24T02:30:42+00:00
|
{"dataset_info": {"features": [{"name": "text", "dtype": "string"}, {"name": "conversation_id", "dtype": "int64"}, {"name": "embedding", "sequence": "float64"}, {"name": "cluster", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 71594107, "num_examples": 7118}], "download_size": 21529040, "dataset_size": 71594107}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T02:30:47+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "med_alpaca_standardized_cluster_98"
More Information needed
|
[
"# Dataset Card for \"med_alpaca_standardized_cluster_98\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"med_alpaca_standardized_cluster_98\"\n\nMore Information needed"
] |
[
6,
22
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"med_alpaca_standardized_cluster_98\"\n\nMore Information needed"
] |
2d096f6c50a3a43463be3e4d336a663306c1fb24
|
# Dataset Card for "med_alpaca_standardized_cluster_99_std"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
AdapterOcean/med_alpaca_standardized_cluster_99_std
|
[
"region:us"
] |
2023-10-24T02:35:44+00:00
|
{"dataset_info": {"features": [{"name": "message", "dtype": "string"}, {"name": "message_type", "dtype": "string"}, {"name": "message_id", "dtype": "int64"}, {"name": "conversation_id", "dtype": "int64"}, {"name": "cluster", "dtype": "float64"}, {"name": "__index_level_0__", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 4631811, "num_examples": 14028}], "download_size": 1757658, "dataset_size": 4631811}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T02:35:48+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "med_alpaca_standardized_cluster_99_std"
More Information needed
|
[
"# Dataset Card for \"med_alpaca_standardized_cluster_99_std\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"med_alpaca_standardized_cluster_99_std\"\n\nMore Information needed"
] |
[
6,
25
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"med_alpaca_standardized_cluster_99_std\"\n\nMore Information needed"
] |
0787feb21b41c32e578f7fee4cff88e4a84836c2
|
# Dataset Card for "med_alpaca_standardized_cluster_99_alpaca"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
AdapterOcean/med_alpaca_standardized_cluster_99_alpaca
|
[
"region:us"
] |
2023-10-24T02:35:49+00:00
|
{"dataset_info": {"features": [{"name": "input", "dtype": "string"}, {"name": "output", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 4228056, "num_examples": 4675}], "download_size": 1687387, "dataset_size": 4228056}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T02:35:51+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "med_alpaca_standardized_cluster_99_alpaca"
More Information needed
|
[
"# Dataset Card for \"med_alpaca_standardized_cluster_99_alpaca\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"med_alpaca_standardized_cluster_99_alpaca\"\n\nMore Information needed"
] |
[
6,
25
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"med_alpaca_standardized_cluster_99_alpaca\"\n\nMore Information needed"
] |
373cb74de5a2a70554fd603c39d6bc99fe70a321
|
# Dataset Card for "med_alpaca_standardized_cluster_99"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
AdapterOcean/med_alpaca_standardized_cluster_99
|
[
"region:us"
] |
2023-10-24T02:35:51+00:00
|
{"dataset_info": {"features": [{"name": "text", "dtype": "string"}, {"name": "conversation_id", "dtype": "int64"}, {"name": "embedding", "sequence": "float64"}, {"name": "cluster", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 42399863, "num_examples": 4676}], "download_size": 11298955, "dataset_size": 42399863}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T02:35:54+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "med_alpaca_standardized_cluster_99"
More Information needed
|
[
"# Dataset Card for \"med_alpaca_standardized_cluster_99\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"med_alpaca_standardized_cluster_99\"\n\nMore Information needed"
] |
[
6,
22
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"med_alpaca_standardized_cluster_99\"\n\nMore Information needed"
] |
6906fbb8b9879caa1c3c9fc938e1fa189b06cc4c
|
# Dataset Card for "commonvoice_train_gender_accent_16k"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
NathanRoll/commonvoice_train_gender_accent_16k
|
[
"region:us"
] |
2023-10-24T02:47:30+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}], "dataset_info": {"features": [{"name": "client_id", "dtype": "string"}, {"name": "path", "dtype": "string"}, {"name": "audio", "dtype": {"audio": {"sampling_rate": 48000}}}, {"name": "sentence", "dtype": "string"}, {"name": "up_votes", "dtype": "int64"}, {"name": "down_votes", "dtype": "int64"}, {"name": "age", "dtype": "string"}, {"name": "gender", "dtype": "string"}, {"name": "accent", "dtype": "string"}, {"name": "locale", "dtype": "string"}, {"name": "segment", "dtype": "string"}, {"name": "variant", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 22821449138.692142, "num_examples": 562872}], "download_size": 22694772617, "dataset_size": 22821449138.692142}}
|
2023-10-24T03:05:05+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "commonvoice_train_gender_accent_16k"
More Information needed
|
[
"# Dataset Card for \"commonvoice_train_gender_accent_16k\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"commonvoice_train_gender_accent_16k\"\n\nMore Information needed"
] |
[
6,
26
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"commonvoice_train_gender_accent_16k\"\n\nMore Information needed"
] |
6dabe3b36af85ac4ceb0deb408ecb324ef247cfb
|
# Dataset Card for "mult_1023"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
tingchih/mult_1023
|
[
"region:us"
] |
2023-10-24T03:01:42+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "test", "path": "data/test-*"}]}], "dataset_info": {"features": [{"name": "text", "dtype": "string"}, {"name": "label", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 47982484, "num_examples": 277071}, {"name": "test", "num_bytes": 20569135, "num_examples": 118745}], "download_size": 44901294, "dataset_size": 68551619}}
|
2023-10-24T03:01:46+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "mult_1023"
More Information needed
|
[
"# Dataset Card for \"mult_1023\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"mult_1023\"\n\nMore Information needed"
] |
[
6,
14
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"mult_1023\"\n\nMore Information needed"
] |
b12ec6eea25b3b68bd2aa3d4e47ca540fb836220
|
Compiled from several different sets of songs:
- (ITG) In the Groove
- (ITG) In the Groove 2
Songs were downloaded from https://search.stepmaniaonline.net/packs/in+the+groove and are stored here for persistence.
In The Groove/ITG typically refers to DDR beatmaps done with an eye towards pad play.
Dataset info: https://paperswithcode.com/dataset/itg
|
origami-digital/in-the-groove
|
[
"task_categories:audio-classification",
"license:unknown",
"stepmania",
"music",
"region:us"
] |
2023-10-24T03:30:55+00:00
|
{"license": "unknown", "task_categories": ["audio-classification"], "pretty_name": "In The Groove Stepmania Songpack", "tags": ["stepmania", "music"]}
|
2023-10-24T04:06:55+00:00
|
[] |
[] |
TAGS
#task_categories-audio-classification #license-unknown #stepmania #music #region-us
|
Compiled from several different sets of songs:
- (ITG) In the Groove
- (ITG) In the Groove 2
Songs were downloaded from URL and are stored here for persistence.
In The Groove/ITG typically refers to DDR beatmaps done with an eye towards pad play.
Dataset info: URL
|
[] |
[
"TAGS\n#task_categories-audio-classification #license-unknown #stepmania #music #region-us \n"
] |
[
30
] |
[
"passage: TAGS\n#task_categories-audio-classification #license-unknown #stepmania #music #region-us \n"
] |
7a3a9b83dbb42a3b482735fd0a244be343a1057b
|
# Dataset Card for Evaluation run of health360/Healix-3B
## Dataset Description
- **Homepage:**
- **Repository:** https://huggingface.co/health360/Healix-3B
- **Paper:**
- **Leaderboard:** https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard
- **Point of Contact:** [email protected]
### Dataset Summary
Dataset automatically created during the evaluation run of model [health360/Healix-3B](https://huggingface.co/health360/Healix-3B) on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).
The dataset is composed of 3 configuration, each one coresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.
An additional configuration "results" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)).
To load the details from a run, you can for instance do the following:
```python
from datasets import load_dataset
data = load_dataset("open-llm-leaderboard/details_health360__Healix-3B",
"harness_winogrande_5",
split="train")
```
## Latest results
These are the [latest results from run 2023-10-24T04:48:25.880438](https://huggingface.co/datasets/open-llm-leaderboard/details_health360__Healix-3B/blob/main/results_2023-10-24T04-48-25.880438.json)(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the "latest" split for each eval):
```python
{
"all": {
"em": 0.001572986577181208,
"em_stderr": 0.00040584511324177333,
"f1": 0.05958368288590625,
"f1_stderr": 0.0014378927693332686,
"acc": 0.3325200323365684,
"acc_stderr": 0.007863382545576577
},
"harness|drop|3": {
"em": 0.001572986577181208,
"em_stderr": 0.00040584511324177333,
"f1": 0.05958368288590625,
"f1_stderr": 0.0014378927693332686
},
"harness|gsm8k|5": {
"acc": 0.0075815011372251705,
"acc_stderr": 0.0023892815120772266
},
"harness|winogrande|5": {
"acc": 0.6574585635359116,
"acc_stderr": 0.013337483579075927
}
}
```
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
[More Information Needed]
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
[More Information Needed]
|
open-llm-leaderboard/details_health360__Healix-3B
|
[
"region:us"
] |
2023-10-24T03:48:30+00:00
|
{"pretty_name": "Evaluation run of health360/Healix-3B", "dataset_summary": "Dataset automatically created during the evaluation run of model [health360/Healix-3B](https://huggingface.co/health360/Healix-3B) on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).\n\nThe dataset is composed of 3 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)).\n\nTo load the details from a run, you can for instance do the following:\n```python\nfrom datasets import load_dataset\ndata = load_dataset(\"open-llm-leaderboard/details_health360__Healix-3B\",\n\t\"harness_winogrande_5\",\n\tsplit=\"train\")\n```\n\n## Latest results\n\nThese are the [latest results from run 2023-10-24T04:48:25.880438](https://huggingface.co/datasets/open-llm-leaderboard/details_health360__Healix-3B/blob/main/results_2023-10-24T04-48-25.880438.json)(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):\n\n```python\n{\n \"all\": {\n \"em\": 0.001572986577181208,\n \"em_stderr\": 0.00040584511324177333,\n \"f1\": 0.05958368288590625,\n \"f1_stderr\": 0.0014378927693332686,\n \"acc\": 0.3325200323365684,\n \"acc_stderr\": 0.007863382545576577\n },\n \"harness|drop|3\": {\n \"em\": 0.001572986577181208,\n \"em_stderr\": 0.00040584511324177333,\n \"f1\": 0.05958368288590625,\n \"f1_stderr\": 0.0014378927693332686\n },\n \"harness|gsm8k|5\": {\n \"acc\": 0.0075815011372251705,\n \"acc_stderr\": 0.0023892815120772266\n },\n \"harness|winogrande|5\": {\n \"acc\": 0.6574585635359116,\n \"acc_stderr\": 0.013337483579075927\n }\n}\n```", "repo_url": "https://huggingface.co/health360/Healix-3B", "leaderboard_url": "https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard", "point_of_contact": "[email protected]", "configs": [{"config_name": "harness_drop_3", "data_files": [{"split": "2023_10_24T04_48_25.880438", "path": ["**/details_harness|drop|3_2023-10-24T04-48-25.880438.parquet"]}, {"split": "latest", "path": ["**/details_harness|drop|3_2023-10-24T04-48-25.880438.parquet"]}]}, {"config_name": "harness_gsm8k_5", "data_files": [{"split": "2023_10_24T04_48_25.880438", "path": ["**/details_harness|gsm8k|5_2023-10-24T04-48-25.880438.parquet"]}, {"split": "latest", "path": ["**/details_harness|gsm8k|5_2023-10-24T04-48-25.880438.parquet"]}]}, {"config_name": "harness_winogrande_5", "data_files": [{"split": "2023_10_24T04_48_25.880438", "path": ["**/details_harness|winogrande|5_2023-10-24T04-48-25.880438.parquet"]}, {"split": "latest", "path": ["**/details_harness|winogrande|5_2023-10-24T04-48-25.880438.parquet"]}]}, {"config_name": "results", "data_files": [{"split": "2023_10_24T04_48_25.880438", "path": ["results_2023-10-24T04-48-25.880438.parquet"]}, {"split": "latest", "path": ["results_2023-10-24T04-48-25.880438.parquet"]}]}]}
|
2023-10-24T03:48:36+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for Evaluation run of health360/Healix-3B
## Dataset Description
- Homepage:
- Repository: URL
- Paper:
- Leaderboard: URL
- Point of Contact: clementine@URL
### Dataset Summary
Dataset automatically created during the evaluation run of model health360/Healix-3B on the Open LLM Leaderboard.
The dataset is composed of 3 configuration, each one coresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.
An additional configuration "results" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the Open LLM Leaderboard).
To load the details from a run, you can for instance do the following:
## Latest results
These are the latest results from run 2023-10-24T04:48:25.880438(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the "latest" split for each eval):
### Supported Tasks and Leaderboards
### Languages
## Dataset Structure
### Data Instances
### Data Fields
### Data Splits
## Dataset Creation
### Curation Rationale
### Source Data
#### Initial Data Collection and Normalization
#### Who are the source language producers?
### Annotations
#### Annotation process
#### Who are the annotators?
### Personal and Sensitive Information
## Considerations for Using the Data
### Social Impact of Dataset
### Discussion of Biases
### Other Known Limitations
## Additional Information
### Dataset Curators
### Licensing Information
### Contributions
|
[
"# Dataset Card for Evaluation run of health360/Healix-3B",
"## Dataset Description\n\n- Homepage: \n- Repository: URL\n- Paper: \n- Leaderboard: URL\n- Point of Contact: clementine@URL",
"### Dataset Summary\n\nDataset automatically created during the evaluation run of model health360/Healix-3B on the Open LLM Leaderboard.\n\nThe dataset is composed of 3 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:",
"## Latest results\n\nThese are the latest results from run 2023-10-24T04:48:25.880438(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):",
"### Supported Tasks and Leaderboards",
"### Languages",
"## Dataset Structure",
"### Data Instances",
"### Data Fields",
"### Data Splits",
"## Dataset Creation",
"### Curation Rationale",
"### Source Data",
"#### Initial Data Collection and Normalization",
"#### Who are the source language producers?",
"### Annotations",
"#### Annotation process",
"#### Who are the annotators?",
"### Personal and Sensitive Information",
"## Considerations for Using the Data",
"### Social Impact of Dataset",
"### Discussion of Biases",
"### Other Known Limitations",
"## Additional Information",
"### Dataset Curators",
"### Licensing Information",
"### Contributions"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for Evaluation run of health360/Healix-3B",
"## Dataset Description\n\n- Homepage: \n- Repository: URL\n- Paper: \n- Leaderboard: URL\n- Point of Contact: clementine@URL",
"### Dataset Summary\n\nDataset automatically created during the evaluation run of model health360/Healix-3B on the Open LLM Leaderboard.\n\nThe dataset is composed of 3 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:",
"## Latest results\n\nThese are the latest results from run 2023-10-24T04:48:25.880438(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):",
"### Supported Tasks and Leaderboards",
"### Languages",
"## Dataset Structure",
"### Data Instances",
"### Data Fields",
"### Data Splits",
"## Dataset Creation",
"### Curation Rationale",
"### Source Data",
"#### Initial Data Collection and Normalization",
"#### Who are the source language producers?",
"### Annotations",
"#### Annotation process",
"#### Who are the annotators?",
"### Personal and Sensitive Information",
"## Considerations for Using the Data",
"### Social Impact of Dataset",
"### Discussion of Biases",
"### Other Known Limitations",
"## Additional Information",
"### Dataset Curators",
"### Licensing Information",
"### Contributions"
] |
[
6,
17,
31,
165,
66,
10,
4,
6,
6,
5,
5,
5,
7,
4,
10,
10,
5,
5,
9,
8,
8,
7,
8,
7,
5,
6,
6,
5
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for Evaluation run of health360/Healix-3B## Dataset Description\n\n- Homepage: \n- Repository: URL\n- Paper: \n- Leaderboard: URL\n- Point of Contact: clementine@URL### Dataset Summary\n\nDataset automatically created during the evaluation run of model health360/Healix-3B on the Open LLM Leaderboard.\n\nThe dataset is composed of 3 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:## Latest results\n\nThese are the latest results from run 2023-10-24T04:48:25.880438(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):### Supported Tasks and Leaderboards### Languages## Dataset Structure### Data Instances### Data Fields### Data Splits## Dataset Creation### Curation Rationale### Source Data#### Initial Data Collection and Normalization#### Who are the source language producers?### Annotations#### Annotation process#### Who are the annotators?### Personal and Sensitive Information## Considerations for Using the Data### Social Impact of Dataset### Discussion of Biases### Other Known Limitations## Additional Information### Dataset Curators### Licensing Information### Contributions"
] |
dc0ca55b99422a6c4acadce4d124bab74320df51
|
# Dataset Card for "closed-QA-vi"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
nlplabtdtu/closed-QA-vi
|
[
"region:us"
] |
2023-10-24T04:08:44+00:00
|
{"dataset_info": {"features": [{"name": "id", "dtype": "string"}, {"name": "question", "dtype": "string"}, {"name": "answer", "dtype": "string"}, {"name": "context", "dtype": "string"}, {"name": "hint", "dtype": "string"}, {"name": "ok", "dtype": "bool"}, {"name": "__index_level_0__", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 11217015, "num_examples": 6380}], "download_size": 5360083, "dataset_size": 11217015}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T07:54:13+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "closed-QA-vi"
More Information needed
|
[
"# Dataset Card for \"closed-QA-vi\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"closed-QA-vi\"\n\nMore Information needed"
] |
[
6,
16
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"closed-QA-vi\"\n\nMore Information needed"
] |
b8bf9c11e87cb2e46ede2fbcadcf5ade1f7928ae
|
# Dataset Card for "so_tay_sv_train"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
Back-up/so_tay_sv_train
|
[
"region:us"
] |
2023-10-24T04:27:14+00:00
|
{"dataset_info": {"features": [{"name": "Questions", "dtype": "string"}, {"name": "Answers", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 103025.68181818182, "num_examples": 140}], "download_size": 48253, "dataset_size": 103025.68181818182}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T04:27:16+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "so_tay_sv_train"
More Information needed
|
[
"# Dataset Card for \"so_tay_sv_train\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"so_tay_sv_train\"\n\nMore Information needed"
] |
[
6,
18
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"so_tay_sv_train\"\n\nMore Information needed"
] |
7d38a2ca78f07fa31bd6074acea735d3cb921fe0
|
# Dataset Card for "so_tay_sv_test"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
Back-up/so_tay_sv_test
|
[
"region:us"
] |
2023-10-24T04:27:16+00:00
|
{"dataset_info": {"features": [{"name": "Questions", "dtype": "string"}, {"name": "Answers", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 26492.31818181818, "num_examples": 36}], "download_size": 20842, "dataset_size": 26492.31818181818}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T04:27:17+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "so_tay_sv_test"
More Information needed
|
[
"# Dataset Card for \"so_tay_sv_test\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"so_tay_sv_test\"\n\nMore Information needed"
] |
[
6,
17
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"so_tay_sv_test\"\n\nMore Information needed"
] |
308a6b6a55fa65ac73f2e634bad2301db3e9af3c
|
# VL Illusion Dataset
Paper:
Homepage: [vl-illusion.github.io](https://vl-illusion.github.io/)
|
sled-umich/VL-Illusion
|
[
"region:us"
] |
2023-10-24T04:32:08+00:00
|
{}
|
2023-11-04T18:11:21+00:00
|
[] |
[] |
TAGS
#region-us
|
# VL Illusion Dataset
Paper:
Homepage: URL
|
[
"# VL Illusion Dataset\n\nPaper:\nHomepage: URL"
] |
[
"TAGS\n#region-us \n",
"# VL Illusion Dataset\n\nPaper:\nHomepage: URL"
] |
[
6,
13
] |
[
"passage: TAGS\n#region-us \n# VL Illusion Dataset\n\nPaper:\nHomepage: URL"
] |
1f3b9b9d0e76d047f0f0fc0a5571cda7ec978317
|
scraped from https://tamil.goodreturns.in/topic/malaysia
|
syafie-nzm/crawl-tamil.goodreturns.in
|
[
"region:us"
] |
2023-10-24T04:32:44+00:00
|
{}
|
2023-10-24T04:36:22+00:00
|
[] |
[] |
TAGS
#region-us
|
scraped from URL
|
[] |
[
"TAGS\n#region-us \n"
] |
[
6
] |
[
"passage: TAGS\n#region-us \n"
] |
5c56f82a6a4c7ed0d7efb0d055c64c1bcde8403b
|
# Dataset Card for "fix-subdomain-train"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
Back-up/fix-subdomain-train
|
[
"region:us"
] |
2023-10-24T04:35:22+00:00
|
{"dataset_info": {"features": [{"name": "Answers", "dtype": "string"}, {"name": "Questions", "dtype": "string"}], "splits": [{"name": "test", "num_bytes": 771141.6, "num_examples": 4396}], "download_size": 188088, "dataset_size": 771141.6}, "configs": [{"config_name": "default", "data_files": [{"split": "test", "path": "data/test-*"}]}]}
|
2023-10-24T04:35:24+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "fix-subdomain-train"
More Information needed
|
[
"# Dataset Card for \"fix-subdomain-train\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"fix-subdomain-train\"\n\nMore Information needed"
] |
[
6,
18
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"fix-subdomain-train\"\n\nMore Information needed"
] |
8cb1a6c427cc128483b3b6218c1e723356eeee77
|
# Dataset Card for "fix-subdomain-test"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
Back-up/fix-subdomain-test
|
[
"region:us"
] |
2023-10-24T04:35:24+00:00
|
{"dataset_info": {"features": [{"name": "Answers", "dtype": "string"}, {"name": "Questions", "dtype": "string"}], "splits": [{"name": "test", "num_bytes": 192785.4, "num_examples": 1099}], "download_size": 48044, "dataset_size": 192785.4}, "configs": [{"config_name": "default", "data_files": [{"split": "test", "path": "data/test-*"}]}]}
|
2023-10-24T04:35:25+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "fix-subdomain-test"
More Information needed
|
[
"# Dataset Card for \"fix-subdomain-test\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"fix-subdomain-test\"\n\nMore Information needed"
] |
[
6,
17
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"fix-subdomain-test\"\n\nMore Information needed"
] |
8fc673fd553378e3b34211820c3ebefdb892c67f
|
# DiPlomat
<!-- Provide a quick summary of the dataset. -->
Pragmatic reasoning plays a pivotal role in deciphering implicit meanings that frequently arise in real-life
conversations and is essential for the development of communicative social agents.
In this paper, we introduce a novel challenge, **DiPlomat**, aiming at benchmarking machines’ capabilities
on pragmatic reasoning and situated conversational understanding.
Compared with previous works that treat different figurative expressions
(e.g. metaphor, sarcasm) as individual tasks, **DiPlomat** provides a cohesive framework
towards general pragmatic understanding.
## Dataset Details
The **DiPlomat** dataset owns 4,177 data and covers a vocabulary of 48,900 words.
More than that, human-annotated answers reach an amount of 6,494,
hold a vocabulary size of 20,000, and cover 5 types of reasoning.
Along with the dataset, we propose two tasks:
**Pragmatic Identification and Reasoning (PIR)** and **Conversational Question Answering (CQA)**. Furthermore, we provide the
data that we use for **zero-NLI**.
- **Language(s) (NLP):** [English]
- **License:** [CC BY-NC-SA (Attribution-NonCommercial-ShareAlike)](https://creativecommons.org/licenses/by-nc-sa/4.0/)
### Dataset Sources
<!-- Provide the basic links for the dataset. -->
- **Repository:** [link](https://github.com/diplomat-dataset/diplomat)
- **Paper:** [DiPlomat: A Dialogue Dataset for Situated Pragmatic Reasoning](https://arxiv.org/abs/2306.09030)
- **Website:** [link](https://diplomat-dataset.github.io)
## Dataset Structure
<!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. -->
| Field | Task|
| --- | --- |
| ``PIR_first`` | Pragmatic Identification and Reasoning Subtask1|
| ``PIR_second`` | Pragmatic Identification and Reasoning Subtask2|
| ``CQA`` | Conversational Question Answering|
| ``NLI_with_context`` | Zero-Shot NLI with context|
| ``NLI_without_context`` | Zero-Shot NLI without context|
**NOTE:** If you'd like to test on the whole PIR task, please don't change the order of PIR Subtask 1's and Subtask 2's test file's data, as both of them are deliberately arranged as the same order.
## Dataset Creation
### Source Data
We leverage the data of [INTERVIEW dataset](https://www.kaggle.com/datasets/shuyangli94/interview-npr-media-dialog-transcripts) collected by
Majumder et al as our source.
<!-- This section describes the source data (e.g. news text and headlines, social media posts, translated sentences, ...). -->
### Annotating Process
<!-- This section describes the data collection and processing process such as data selection criteria, filtering and normalization methods, tools and libraries used, etc. -->
#### Step I. Automatic Selection:
The extensive size of the source dataset introduces redundancy,
and thus requires automatic measures to alleviate the burden of human annotation.
Therefore, we employ algorithms and models to perform an initial filtering process.
#### Step II. Fine-grained Annotation:
We leverage Amazon Mechanical Turk to conduct detailed annotations of pragmatic turns within our dialogues.
Workers participating in the annotation task are instructed to select
all turns that exhibit a divergence between their literal meaning and their intended meaning.
Due to the subjective nature of pragmatic reasoning, we request the workers to provide confidence scores
along with reasons for their choices.
#### Step III. Human Refinement:
In this process, tasks for workers are formulated as multiple-choice questions.
Previously collected human-annotated reasons are transformed into choices, utilizing a template format:
[turn {turn_id}: {reason}]. In addition, to mitigate the impact of careless workers,
we introduce a distractor choice for each gold choice.
## Citation
<!-- If there is a paper or blog post introducing the dataset, the APA and Bibtex information for that should go in this section. -->
```
@inproceedings{li2023diplomat,
title={DiPlomat: A Dialogue Dataset for Situated Pragmatic Reasoning},
author={Hengli Li and Song-Chun Zhu and Zilong Zheng},
booktitle={Thirty-seventh Conference on Neural Information Processing Systems Datasets and Benchmarks Track},
year={2023}
}
```
## Dataset Card Contact
If there is any problem with the dataset, please email [[email protected]](mailto: [email protected]).
|
bigai/DiPlomat
|
[
"language:en",
"license:cc-by-nc-sa-4.0",
"arxiv:2306.09030",
"region:us"
] |
2023-10-24T04:46:58+00:00
|
{"language": ["en"], "license": "cc-by-nc-sa-4.0", "pretty_name": "DiPlomat", "dataset_info": [{"config_name": "CQA", "features": [{"name": "text", "sequence": "string"}, {"name": "speaker", "sequence": "string"}, {"name": "gold_statement", "dtype": "string"}, {"name": "questions", "dtype": "string"}, {"name": "answer", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 19805284, "num_examples": 15585}, {"name": "validation", "num_bytes": 1959148, "num_examples": 1559}, {"name": "test", "num_bytes": 2967746, "num_examples": 2338}], "download_size": 25566918, "dataset_size": 24732178}, {"config_name": "NLI_with_context", "features": [{"name": "dialogue", "sequence": "string"}, {"name": "speaker", "sequence": "string"}, {"name": "human answer", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 2977929, "num_examples": 2551}], "download_size": 3042193, "dataset_size": 2977929}, {"config_name": "NLI_without_context", "features": [{"name": "text", "dtype": "string"}, {"name": "hypothesis", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 1095335, "num_examples": 2551}], "download_size": 1146864, "dataset_size": 1095335}, {"config_name": "PIR_first", "features": [{"name": "text", "sequence": "string"}, {"name": "speaker", "sequence": "string"}, {"name": "correct_turn_number", "sequence": "int64"}], "splits": [{"name": "train", "num_bytes": 3442927, "num_examples": 3341}, {"name": "validation", "num_bytes": 161433, "num_examples": 168}, {"name": "test", "num_bytes": 687605, "num_examples": 668}], "download_size": 4366468, "dataset_size": 4291965}, {"config_name": "PIR_second", "features": [{"name": "text", "sequence": "string"}, {"name": "speaker", "sequence": "string"}, {"name": "correct_turn_number", "dtype": "int64"}, {"name": "label", "dtype": "int64"}, {"name": "choice", "sequence": "string"}], "splits": [{"name": "train", "num_bytes": 9263111, "num_examples": 5188}, {"name": "validation", "num_bytes": 399924, "num_examples": 244}, {"name": "test", "num_bytes": 1890798, "num_examples": 1062}], "download_size": 11740508, "dataset_size": 11553833}]}
|
2023-11-13T04:46:54+00:00
|
[
"2306.09030"
] |
[
"en"
] |
TAGS
#language-English #license-cc-by-nc-sa-4.0 #arxiv-2306.09030 #region-us
|
DiPlomat
========
Pragmatic reasoning plays a pivotal role in deciphering implicit meanings that frequently arise in real-life
conversations and is essential for the development of communicative social agents.
In this paper, we introduce a novel challenge, DiPlomat, aiming at benchmarking machines’ capabilities
on pragmatic reasoning and situated conversational understanding.
Compared with previous works that treat different figurative expressions
(e.g. metaphor, sarcasm) as individual tasks, DiPlomat provides a cohesive framework
towards general pragmatic understanding.
Dataset Details
---------------
The DiPlomat dataset owns 4,177 data and covers a vocabulary of 48,900 words.
More than that, human-annotated answers reach an amount of 6,494,
hold a vocabulary size of 20,000, and cover 5 types of reasoning.
Along with the dataset, we propose two tasks:
Pragmatic Identification and Reasoning (PIR) and Conversational Question Answering (CQA). Furthermore, we provide the
data that we use for zero-NLI.
* Language(s) (NLP): [English]
* License: CC BY-NC-SA (Attribution-NonCommercial-ShareAlike)
### Dataset Sources
* Repository: link
* Paper: DiPlomat: A Dialogue Dataset for Situated Pragmatic Reasoning
* Website: link
Dataset Structure
-----------------
NOTE: If you'd like to test on the whole PIR task, please don't change the order of PIR Subtask 1's and Subtask 2's test file's data, as both of them are deliberately arranged as the same order.
Dataset Creation
----------------
### Source Data
We leverage the data of INTERVIEW dataset collected by
Majumder et al as our source.
### Annotating Process
#### Step I. Automatic Selection:
The extensive size of the source dataset introduces redundancy,
and thus requires automatic measures to alleviate the burden of human annotation.
Therefore, we employ algorithms and models to perform an initial filtering process.
#### Step II. Fine-grained Annotation:
We leverage Amazon Mechanical Turk to conduct detailed annotations of pragmatic turns within our dialogues.
Workers participating in the annotation task are instructed to select
all turns that exhibit a divergence between their literal meaning and their intended meaning.
Due to the subjective nature of pragmatic reasoning, we request the workers to provide confidence scores
along with reasons for their choices.
#### Step III. Human Refinement:
In this process, tasks for workers are formulated as multiple-choice questions.
Previously collected human-annotated reasons are transformed into choices, utilizing a template format:
[turn {turn\_id}: {reason}]. In addition, to mitigate the impact of careless workers,
we introduce a distractor choice for each gold choice.
Dataset Card Contact
--------------------
If there is any problem with the dataset, please email lihengli@URL.
|
[
"### Dataset Sources\n\n\n* Repository: link\n* Paper: DiPlomat: A Dialogue Dataset for Situated Pragmatic Reasoning\n* Website: link\n\n\nDataset Structure\n-----------------\n\n\n\nNOTE: If you'd like to test on the whole PIR task, please don't change the order of PIR Subtask 1's and Subtask 2's test file's data, as both of them are deliberately arranged as the same order.\n\n\nDataset Creation\n----------------",
"### Source Data\n\n\nWe leverage the data of INTERVIEW dataset collected by\nMajumder et al as our source.",
"### Annotating Process",
"#### Step I. Automatic Selection:\n\n\nThe extensive size of the source dataset introduces redundancy,\nand thus requires automatic measures to alleviate the burden of human annotation.\nTherefore, we employ algorithms and models to perform an initial filtering process.",
"#### Step II. Fine-grained Annotation:\n\n\nWe leverage Amazon Mechanical Turk to conduct detailed annotations of pragmatic turns within our dialogues.\nWorkers participating in the annotation task are instructed to select\nall turns that exhibit a divergence between their literal meaning and their intended meaning.\nDue to the subjective nature of pragmatic reasoning, we request the workers to provide confidence scores\nalong with reasons for their choices.",
"#### Step III. Human Refinement:\n\n\nIn this process, tasks for workers are formulated as multiple-choice questions.\nPreviously collected human-annotated reasons are transformed into choices, utilizing a template format:\n[turn {turn\\_id}: {reason}]. In addition, to mitigate the impact of careless workers,\nwe introduce a distractor choice for each gold choice.\n\n\nDataset Card Contact\n--------------------\n\n\nIf there is any problem with the dataset, please email lihengli@URL."
] |
[
"TAGS\n#language-English #license-cc-by-nc-sa-4.0 #arxiv-2306.09030 #region-us \n",
"### Dataset Sources\n\n\n* Repository: link\n* Paper: DiPlomat: A Dialogue Dataset for Situated Pragmatic Reasoning\n* Website: link\n\n\nDataset Structure\n-----------------\n\n\n\nNOTE: If you'd like to test on the whole PIR task, please don't change the order of PIR Subtask 1's and Subtask 2's test file's data, as both of them are deliberately arranged as the same order.\n\n\nDataset Creation\n----------------",
"### Source Data\n\n\nWe leverage the data of INTERVIEW dataset collected by\nMajumder et al as our source.",
"### Annotating Process",
"#### Step I. Automatic Selection:\n\n\nThe extensive size of the source dataset introduces redundancy,\nand thus requires automatic measures to alleviate the burden of human annotation.\nTherefore, we employ algorithms and models to perform an initial filtering process.",
"#### Step II. Fine-grained Annotation:\n\n\nWe leverage Amazon Mechanical Turk to conduct detailed annotations of pragmatic turns within our dialogues.\nWorkers participating in the annotation task are instructed to select\nall turns that exhibit a divergence between their literal meaning and their intended meaning.\nDue to the subjective nature of pragmatic reasoning, we request the workers to provide confidence scores\nalong with reasons for their choices.",
"#### Step III. Human Refinement:\n\n\nIn this process, tasks for workers are formulated as multiple-choice questions.\nPreviously collected human-annotated reasons are transformed into choices, utilizing a template format:\n[turn {turn\\_id}: {reason}]. In addition, to mitigate the impact of careless workers,\nwe introduce a distractor choice for each gold choice.\n\n\nDataset Card Contact\n--------------------\n\n\nIf there is any problem with the dataset, please email lihengli@URL."
] |
[
31,
106,
26,
6,
61,
99,
117
] |
[
"passage: TAGS\n#language-English #license-cc-by-nc-sa-4.0 #arxiv-2306.09030 #region-us \n### Dataset Sources\n\n\n* Repository: link\n* Paper: DiPlomat: A Dialogue Dataset for Situated Pragmatic Reasoning\n* Website: link\n\n\nDataset Structure\n-----------------\n\n\n\nNOTE: If you'd like to test on the whole PIR task, please don't change the order of PIR Subtask 1's and Subtask 2's test file's data, as both of them are deliberately arranged as the same order.\n\n\nDataset Creation\n----------------### Source Data\n\n\nWe leverage the data of INTERVIEW dataset collected by\nMajumder et al as our source.### Annotating Process#### Step I. Automatic Selection:\n\n\nThe extensive size of the source dataset introduces redundancy,\nand thus requires automatic measures to alleviate the burden of human annotation.\nTherefore, we employ algorithms and models to perform an initial filtering process.#### Step II. Fine-grained Annotation:\n\n\nWe leverage Amazon Mechanical Turk to conduct detailed annotations of pragmatic turns within our dialogues.\nWorkers participating in the annotation task are instructed to select\nall turns that exhibit a divergence between their literal meaning and their intended meaning.\nDue to the subjective nature of pragmatic reasoning, we request the workers to provide confidence scores\nalong with reasons for their choices.#### Step III. Human Refinement:\n\n\nIn this process, tasks for workers are formulated as multiple-choice questions.\nPreviously collected human-annotated reasons are transformed into choices, utilizing a template format:\n[turn {turn\\_id}: {reason}]. In addition, to mitigate the impact of careless workers,\nwe introduce a distractor choice for each gold choice.\n\n\nDataset Card Contact\n--------------------\n\n\nIf there is any problem with the dataset, please email lihengli@URL."
] |
5c010ae05b5479c8609975972e0b8e6644b96b3b
|
# Dataset Card for "github-commits"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
akazad/github-commits
|
[
"region:us"
] |
2023-10-24T04:58:40+00:00
|
{"dataset_info": {"features": [{"name": "hash", "dtype": "string"}, {"name": "msg", "dtype": "string"}, {"name": "author", "dtype": "string"}, {"name": "email", "dtype": "string"}, {"name": "date", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 42673325, "num_examples": 155401}], "download_size": 24367297, "dataset_size": 42673325}}
|
2023-10-24T06:14:14+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "github-commits"
More Information needed
|
[
"# Dataset Card for \"github-commits\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"github-commits\"\n\nMore Information needed"
] |
[
6,
16
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"github-commits\"\n\nMore Information needed"
] |
7704b61b8acf9353dc915d495c0be4cd081646cb
|
# Dataset Card for "tokenzied_news_2gb_data"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
leeseeun/tokenzied_news_2gb_data
|
[
"region:us"
] |
2023-10-24T05:03:53+00:00
|
{"dataset_info": {"features": [{"name": "input_ids", "sequence": "int32"}], "splits": [{"name": "train", "num_bytes": 2230572200, "num_examples": 544042}], "download_size": 989285251, "dataset_size": 2230572200}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T05:05:04+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "tokenzied_news_2gb_data"
More Information needed
|
[
"# Dataset Card for \"tokenzied_news_2gb_data\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"tokenzied_news_2gb_data\"\n\nMore Information needed"
] |
[
6,
20
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"tokenzied_news_2gb_data\"\n\nMore Information needed"
] |
bc081acce60e5d178f808e37923e39a77ea13ac3
|
# Dataset Card for "sft_test_custom_dataset_RLHF_updated"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
sayan1101/sft_test_custom_dataset_RLHF_updated
|
[
"region:us"
] |
2023-10-24T06:12:20+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "test", "path": "data/test-*"}, {"split": "valid", "path": "data/valid-*"}]}], "dataset_info": {"features": [{"name": "label", "dtype": "string"}, {"name": "prompt", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 35042, "num_examples": 51}, {"name": "test", "num_bytes": 35042, "num_examples": 51}, {"name": "valid", "num_bytes": 35042, "num_examples": 51}], "download_size": 87354, "dataset_size": 105126}}
|
2023-10-24T06:24:08+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "sft_test_custom_dataset_RLHF_updated"
More Information needed
|
[
"# Dataset Card for \"sft_test_custom_dataset_RLHF_updated\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"sft_test_custom_dataset_RLHF_updated\"\n\nMore Information needed"
] |
[
6,
26
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"sft_test_custom_dataset_RLHF_updated\"\n\nMore Information needed"
] |
58c366f4a8ab05ad920a1b35c57130d2bb29fceb
|
# Dataset Card for "so-llama2-500"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
RayLy/so-llama2-500
|
[
"region:us"
] |
2023-10-24T06:19:45+00:00
|
{"dataset_info": {"features": [{"name": "text", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 709086, "num_examples": 265}], "download_size": 177937, "dataset_size": 709086}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T06:21:57+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "so-llama2-500"
More Information needed
|
[
"# Dataset Card for \"so-llama2-500\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"so-llama2-500\"\n\nMore Information needed"
] |
[
6,
16
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"so-llama2-500\"\n\nMore Information needed"
] |
1ea2147e1948da28f37cb9e9a9148ac6adb7bd12
|
{"inputs": "###Instruction: Managing stress through creativity###Response: When you're feeling stressed or on the edge of anxiety, consider turning to creative outlets like drawing, painting, or even adult coloring books. It can provide a sense of relief and help you focus your thoughts."},
{"inputs": "###Instruction: Coping with post-breakup solitude###Response: After a breakup, loneliness can be overwhelming. If you have free evenings, try finding activities that keep you engaged. This might include exploring new hobbies or spending more time with friends."},
{"inputs": "###Instruction: Discovering personal purpose in life###Response: Finding your life's purpose can be challenging. Instead of dwelling on the idea that life is without meaning, take small steps towards self-discovery. Set goals, learn new things, and explore various interests."},
{"inputs": "###Instruction: Navigating fear and anxiety as a person of color###Response: Fear and anxiety in the face of adversity can be daunting. Your experiences as a person of color are valuable, and it's essential to share them with supportive communities or seek guidance from those who understand your perspective."},
{"inputs": "###Instruction: Dealing with uncertainty and insecurity###Response: When you're uncertain about your path in life and struggling with self-doubt, it's important to remember that many people go through similar phases. Explore different interests, take small steps, and don't hesitate to seek guidance and support."},
{"inputs": "###Instruction: Finding meaning in life's challenges###Response: Life's challenges can make it difficult to see a clear purpose. However, they also offer opportunities for growth. Embrace difficulties as chances to learn and discover your strengths. Your journey is unique, and there is meaning in every experience."},
{"inputs": "###Instruction: Coping with the unknown###Response: When you're faced with the unknown, it can be unsettling. Remember that uncertainty is a part of life, and it's a chance to explore and grow. Take one step at a time, seek advice, and maintain a positive outlook."},
{"inputs": "###Instruction: Enhancing mental well-being through mindfulness###Response: Incorporating mindfulness practices into your daily routine can help improve your mental well-being. Techniques like meditation, deep breathing, and focusing on the present moment can reduce stress and increase self-awareness."},
{"inputs": "###Instruction: Coping with anxiety in social situations###Response: Social anxiety can be challenging, but it's possible to overcome it. Gradual exposure to social situations, self-compassion, and cognitive-behavioral therapy can be effective tools for managing anxiety in social settings."},
{"inputs": "###Instruction: Building resilience in the face of adversity###Response: Resilience is a valuable trait for mental health. You can enhance it by developing problem-solving skills, seeking support from friends and family, and maintaining a positive outlook even in challenging times."},
{"inputs": "###Instruction: Nurturing self-esteem and self-acceptance###Response: Developing a healthy self-esteem is crucial for mental well-being. Focus on self-acceptance, practice self-compassion, and challenge negative self-talk to build a more positive self-image."},
{"inputs": "###Instruction: Coping with depression and seeking help###Response: If you're dealing with depression, remember that seeking professional help is a critical step. Therapists, counselors, and support groups can provide the assistance and guidance needed to overcome depression."},
{"inputs": "###Instruction: Building strong relationships and emotional connections###Response: Positive relationships are essential for mental health. Invest in building strong connections with friends and loved ones by actively listening, showing empathy, and communicating openly."},
{"inputs": "###Instruction: Handling stress at work and maintaining work-life balance###Response: Balancing work and personal life is vital for mental well-being. Set boundaries, prioritize self-care, and practice time management to reduce work-related stress."},
{"inputs": "###Instruction: Recognizing signs of suicidal thoughts###Response: It's crucial to be aware of the signs of suicidal thoughts in yourself or others. These signs may include talking about death or dying, expressing hopelessness, or giving away belongings. If you notice these signs, seek help immediately."},
{"inputs": "###Instruction: Providing support to someone in crisis###Response: If someone you know is in crisis, it's essential to offer your support. Listen without judgment, encourage them to talk about their feelings, and help them connect with professional resources, such as a crisis helpline or a therapist."},
{"inputs": "###Instruction: Promoting open conversations about mental health###Response: Open and honest conversations about mental health can reduce the stigma surrounding it. Encourage dialogue with friends and loved ones about their emotional well-being. It's okay not to be okay, and seeking help is a sign of strength."},
{"inputs": "###Instruction: Preventing access to lethal means###Response: If you or someone you know is at risk of suicide, it's crucial to restrict access to lethal means such as firearms, medications, or sharp objects. This can save lives by creating a safer environment."},
{"inputs": "###Instruction: Encouraging professional help for suicidal thoughts###Response: Suicidal thoughts are a serious concern that requires professional intervention. Encourage anyone struggling with these thoughts to seek help from a therapist, counselor, or a crisis hotline. You are not alone in this journey."},
{"inputs": "###Instruction: Supporting mental health initiatives###Response: Advocate for mental health initiatives and organizations that raise awareness and provide resources for those at risk of suicide. Your support can make a significant difference in saving lives."},
{"inputs": "###Instruction: Fostering hope and resilience###Response: In times of despair, it's crucial to foster hope and resilience. Connect with supportive communities, engage in self-care, and remember that healing is possible. Your life is valuable, and there is hope for a better future."},
{"inputs": "###Instruction: Recognizing the impact of bullying on mental health###Response: Bullying can have a severe impact on mental health and may contribute to suicidal thoughts. Take a stand against bullying, offer support to those affected, and create a safe environment for everyone."},
{"inputs": "###Instruction: Seeking help for self-harm and suicidal tendencies###Response: If you're struggling with self-harm or suicidal tendencies, please seek immediate professional help. Therapists, counselors, and crisis hotlines are available to provide the support and guidance you need."},
{"inputs": "###Instruction: Promoting self-care and emotional well-being###Response: Prioritize self-care and emotional well-being. Engage in activities that bring you joy, connect with positive influences, and remember that seeking help is a sign of strength. Your mental health matters."},
{"inputs": "###Instruction: Overcoming depression through self-care###Response: Self-care plays a significant role in overcoming depression. Prioritize activities that bring you joy, practice self-compassion, and engage in regular exercise to boost your mood and mental well-being."},
{"inputs": "###Instruction: Recognizing the signs of depression in yourself###Response: Being aware of the signs of depression in yourself is essential. These signs may include persistent sadness, loss of interest in activities, changes in appetite, and trouble sleeping. Seeking help early can make a difference."},
{"inputs": "###Instruction: Providing support to a loved one with depression###Response: If someone you care about is dealing with depression, offer your support and understanding. Listen without judgment, encourage them to seek professional help, and be patient as they navigate their journey to recovery."},
{"inputs": "###Instruction: Reducing the stigma around mental health###Response: It's important to reduce the stigma surrounding mental health. Open conversations about depression and mental well-being can create a more supportive environment for those who are struggling."},
{"inputs": "###Instruction: Coping with depression and isolation###Response: Coping with depression and isolation can be challenging. Connect with friends and loved ones, participate in group activities, and seek professional therapy to combat feelings of loneliness and despair."},
{"inputs": "###Instruction: Setting achievable goals for managing depression###Response: Setting achievable goals is a practical way to manage depression. Start with small steps, celebrate your successes, and gradually work your way toward larger objectives. This can provide a sense of accomplishment and hope."},
{"inputs": "###Instruction: Strategies for improving sleep and managing depression###Response: Adequate sleep is crucial for managing depression. Create a bedtime routine, limit screen time before bed, and consider relaxation techniques to improve your sleep quality and overall mental health."},
{"inputs": "###Instruction: Seeking professional help for depression###Response: Depression is a serious condition that often requires professional intervention. Don't hesitate to reach out to therapists, counselors, or support groups to get the assistance and guidance you need for recovery."},
{"inputs": "###Instruction: Overcoming loneliness through self-connection###Response: Overcoming loneliness often starts with connecting with yourself. Engage in self-reflection, discover your interests, and invest time in self-care. Building a strong connection with yourself can help alleviate feelings of loneliness."},
{"inputs": "###Instruction: Cultivating social connections to combat loneliness###Response: Cultivating social connections is essential to combat loneliness. Seek opportunities to meet new people, join clubs or communities aligned with your interests, and engage in social activities to build meaningful relationships."},
{"inputs": "###Instruction: Coping with loneliness in old age###Response: Loneliness can be particularly challenging in old age. It's essential to stay socially active, reach out to family and friends, and consider volunteering or participating in senior programs to combat loneliness."},
{"inputs": "###Instruction: Using technology to reduce loneliness###Response: Technology can be a valuable tool to reduce loneliness. Stay connected with loved ones through video calls and social media, join online communities related to your interests, and explore virtual events to stay engaged and combat loneliness."},
{"inputs": "###Instruction: Managing loneliness during the holidays###Response: Loneliness during the holidays can be especially challenging. Reach out to friends and family, create your own holiday traditions, and consider volunteering or attending local events to make the season more enjoyable."},
{"inputs": "###Instruction: Seeking professional help for loneliness###Response: If loneliness is overwhelming and persistent, seeking professional help is a valid option. Therapists and counselors can provide strategies and support to address loneliness and improve your mental well-being."},
{"inputs": "###Instruction: Finding meaning and purpose to combat loneliness###Response: Finding meaning and purpose in your life can help combat loneliness. Pursue activities that align with your values, set goals, and seek personal growth to feel more fulfilled and connected to the world around you."},
{"inputs": "###Instruction: Fostering a pet to alleviate loneliness###Response: Fostering a pet can be a rewarding way to alleviate loneliness. Pets offer companionship and can help reduce feelings of isolation. Consider adopting a furry friend to improve your emotional well-being."},
{"inputs": "###Instruction: Managing anxiety through relaxation techniques###Response: Managing anxiety often involves relaxation techniques. Consider practices like deep breathing, meditation, or progressive muscle relaxation to calm your mind and reduce anxiety."},
{"inputs": "###Instruction: Overcoming social anxiety and building confidence###Response: Overcoming social anxiety requires building confidence. Gradual exposure to social situations, positive self-talk, and seeking support from friends or a therapist can help you become more socially comfortable."},
{"inputs": "###Instruction: Coping with test anxiety and improving performance###Response: Coping with test anxiety is essential for better performance. Effective study strategies, time management, and relaxation techniques can help reduce anxiety and enhance your test-taking abilities."},
{"inputs": "###Instruction: Reducing anxiety through physical activity###Response: Physical activity can help reduce anxiety. Regular exercise releases endorphins, which can improve your mood and alleviate anxiety. Incorporate exercise into your routine for better mental well-being."},
{"inputs": "###Instruction: Seeking therapy for anxiety and panic attacks###Response: If you experience frequent anxiety or panic attacks, seeking therapy is a valuable option. Cognitive-behavioral therapy and counseling can provide you with strategies to manage anxiety and regain control."},
{"inputs": "###Instruction: Managing work-related anxiety and achieving work-life balance###Response: Managing work-related anxiety is crucial for maintaining a work-life balance. Set clear boundaries, prioritize self-care, and practice stress-reduction techniques to reduce anxiety at work."},
{"inputs": "###Instruction: Overcoming driving anxiety and regaining independence###Response: Overcoming driving anxiety can help you regain independence. Gradual exposure to driving, relaxation techniques, and support from friends or a therapist can assist you in managing driving-related anxiety."},
{"inputs": "###Instruction: Using technology for anxiety management###Response: Technology can be a helpful tool for managing anxiety. Apps for mindfulness, stress reduction, and virtual therapy can assist you in tracking and alleviating anxiety symptoms."},
{"inputs": "###Instruction: Managing anger through self-awareness###Response: Managing anger begins with self-awareness. Recognizing your triggers and learning to control your reactions can help you respond to anger in a more constructive way."},
{"inputs": "###Instruction: Coping with anger in relationships###Response: Coping with anger in relationships is crucial. Effective communication, empathy, and conflict resolution skills can help you navigate and resolve anger-related issues with your loved ones."},
{"inputs": "###Instruction: Handling workplace anger and maintaining professionalism###Response: Handling workplace anger while maintaining professionalism is essential. Deep breathing, time management, and addressing issues through appropriate channels can help you manage workplace anger without compromising your job."},
{"inputs": "###Instruction: Overcoming road rage and promoting safe driving###Response: Overcoming road rage is important for safe driving. Implement strategies like practicing patience, listening to soothing music, and avoiding aggressive behavior on the road to ensure road safety."},
{"inputs": "###Instruction: Dealing with anger issues and seeking therapy###Response: If you're struggling with anger issues, seeking therapy can be beneficial. Anger management therapy and counseling can provide you with tools to control your anger and improve your mental well-being."},
{"inputs": "###Instruction: Nurturing emotional intelligence to manage anger###Response: Nurturing emotional intelligence is key to managing anger. Developing self-awareness, empathy, and effective emotional regulation can help you respond to anger in a healthier and more productive way."},
{"inputs": "###Instruction: Using anger as a motivator for positive change###Response: Anger can be channeled as a motivator for positive change. Instead of letting anger consume you, use it as a driving force to address issues, advocate for change, and make a difference in your life or the world."},
{"inputs": "###Instruction: Self-Harm Prevention Strategies###Response: Preventing self-harm is essential for your well-being. Reach out to a trusted friend or family member when you're struggling, create a self-care plan, and seek professional help if needed."},
{"inputs": "###Instruction: Coping with self-harm urges###Response: Coping with self-harm urges can be challenging. Distract yourself with a healthy activity, like drawing, journaling, or exercise, and remember that seeking support from a therapist or counselor is a positive step."},
{"inputs": "###Instruction: Building a support network for self-harm recovery###Response: Building a strong support network is vital for self-harm recovery. Open up to friends and family about your struggles, attend support groups, and connect with professionals who specialize in self-harm recovery."},
{"inputs": "###Instruction: Identifying triggers and warning signs###Response: Identifying triggers and warning signs is a crucial part of self-harm prevention. Keep a journal to track your emotions and situations that lead to self-harm urges, and work on strategies to avoid or cope with these triggers."},
{"inputs": "###Instruction: Seeking professional help for self-harm recovery###Response: If you're struggling with self-harm, seeking professional help is essential. Therapists, counselors, and mental health experts can provide you with the guidance and support needed to overcome self-harm behaviors."},
{"inputs": "###Instruction: Developing healthy coping mechanisms###Response: Developing healthy coping mechanisms is key to self-harm prevention. Explore alternative ways to manage stress and emotions, such as mindfulness, art therapy, or meditation, to replace self-harm behaviors."}
|
lilpotato03/sample_medchatbot_llm2
|
[
"region:us"
] |
2023-10-24T06:36:53+00:00
|
{}
|
2023-10-24T09:09:03+00:00
|
[] |
[] |
TAGS
#region-us
|
{"inputs": "###Instruction: Managing stress through creativity###Response: When you're feeling stressed or on the edge of anxiety, consider turning to creative outlets like drawing, painting, or even adult coloring books. It can provide a sense of relief and help you focus your thoughts."},
{"inputs": "###Instruction: Coping with post-breakup solitude###Response: After a breakup, loneliness can be overwhelming. If you have free evenings, try finding activities that keep you engaged. This might include exploring new hobbies or spending more time with friends."},
{"inputs": "###Instruction: Discovering personal purpose in life###Response: Finding your life's purpose can be challenging. Instead of dwelling on the idea that life is without meaning, take small steps towards self-discovery. Set goals, learn new things, and explore various interests."},
{"inputs": "###Instruction: Navigating fear and anxiety as a person of color###Response: Fear and anxiety in the face of adversity can be daunting. Your experiences as a person of color are valuable, and it's essential to share them with supportive communities or seek guidance from those who understand your perspective."},
{"inputs": "###Instruction: Dealing with uncertainty and insecurity###Response: When you're uncertain about your path in life and struggling with self-doubt, it's important to remember that many people go through similar phases. Explore different interests, take small steps, and don't hesitate to seek guidance and support."},
{"inputs": "###Instruction: Finding meaning in life's challenges###Response: Life's challenges can make it difficult to see a clear purpose. However, they also offer opportunities for growth. Embrace difficulties as chances to learn and discover your strengths. Your journey is unique, and there is meaning in every experience."},
{"inputs": "###Instruction: Coping with the unknown###Response: When you're faced with the unknown, it can be unsettling. Remember that uncertainty is a part of life, and it's a chance to explore and grow. Take one step at a time, seek advice, and maintain a positive outlook."},
{"inputs": "###Instruction: Enhancing mental well-being through mindfulness###Response: Incorporating mindfulness practices into your daily routine can help improve your mental well-being. Techniques like meditation, deep breathing, and focusing on the present moment can reduce stress and increase self-awareness."},
{"inputs": "###Instruction: Coping with anxiety in social situations###Response: Social anxiety can be challenging, but it's possible to overcome it. Gradual exposure to social situations, self-compassion, and cognitive-behavioral therapy can be effective tools for managing anxiety in social settings."},
{"inputs": "###Instruction: Building resilience in the face of adversity###Response: Resilience is a valuable trait for mental health. You can enhance it by developing problem-solving skills, seeking support from friends and family, and maintaining a positive outlook even in challenging times."},
{"inputs": "###Instruction: Nurturing self-esteem and self-acceptance###Response: Developing a healthy self-esteem is crucial for mental well-being. Focus on self-acceptance, practice self-compassion, and challenge negative self-talk to build a more positive self-image."},
{"inputs": "###Instruction: Coping with depression and seeking help###Response: If you're dealing with depression, remember that seeking professional help is a critical step. Therapists, counselors, and support groups can provide the assistance and guidance needed to overcome depression."},
{"inputs": "###Instruction: Building strong relationships and emotional connections###Response: Positive relationships are essential for mental health. Invest in building strong connections with friends and loved ones by actively listening, showing empathy, and communicating openly."},
{"inputs": "###Instruction: Handling stress at work and maintaining work-life balance###Response: Balancing work and personal life is vital for mental well-being. Set boundaries, prioritize self-care, and practice time management to reduce work-related stress."},
{"inputs": "###Instruction: Recognizing signs of suicidal thoughts###Response: It's crucial to be aware of the signs of suicidal thoughts in yourself or others. These signs may include talking about death or dying, expressing hopelessness, or giving away belongings. If you notice these signs, seek help immediately."},
{"inputs": "###Instruction: Providing support to someone in crisis###Response: If someone you know is in crisis, it's essential to offer your support. Listen without judgment, encourage them to talk about their feelings, and help them connect with professional resources, such as a crisis helpline or a therapist."},
{"inputs": "###Instruction: Promoting open conversations about mental health###Response: Open and honest conversations about mental health can reduce the stigma surrounding it. Encourage dialogue with friends and loved ones about their emotional well-being. It's okay not to be okay, and seeking help is a sign of strength."},
{"inputs": "###Instruction: Preventing access to lethal means###Response: If you or someone you know is at risk of suicide, it's crucial to restrict access to lethal means such as firearms, medications, or sharp objects. This can save lives by creating a safer environment."},
{"inputs": "###Instruction: Encouraging professional help for suicidal thoughts###Response: Suicidal thoughts are a serious concern that requires professional intervention. Encourage anyone struggling with these thoughts to seek help from a therapist, counselor, or a crisis hotline. You are not alone in this journey."},
{"inputs": "###Instruction: Supporting mental health initiatives###Response: Advocate for mental health initiatives and organizations that raise awareness and provide resources for those at risk of suicide. Your support can make a significant difference in saving lives."},
{"inputs": "###Instruction: Fostering hope and resilience###Response: In times of despair, it's crucial to foster hope and resilience. Connect with supportive communities, engage in self-care, and remember that healing is possible. Your life is valuable, and there is hope for a better future."},
{"inputs": "###Instruction: Recognizing the impact of bullying on mental health###Response: Bullying can have a severe impact on mental health and may contribute to suicidal thoughts. Take a stand against bullying, offer support to those affected, and create a safe environment for everyone."},
{"inputs": "###Instruction: Seeking help for self-harm and suicidal tendencies###Response: If you're struggling with self-harm or suicidal tendencies, please seek immediate professional help. Therapists, counselors, and crisis hotlines are available to provide the support and guidance you need."},
{"inputs": "###Instruction: Promoting self-care and emotional well-being###Response: Prioritize self-care and emotional well-being. Engage in activities that bring you joy, connect with positive influences, and remember that seeking help is a sign of strength. Your mental health matters."},
{"inputs": "###Instruction: Overcoming depression through self-care###Response: Self-care plays a significant role in overcoming depression. Prioritize activities that bring you joy, practice self-compassion, and engage in regular exercise to boost your mood and mental well-being."},
{"inputs": "###Instruction: Recognizing the signs of depression in yourself###Response: Being aware of the signs of depression in yourself is essential. These signs may include persistent sadness, loss of interest in activities, changes in appetite, and trouble sleeping. Seeking help early can make a difference."},
{"inputs": "###Instruction: Providing support to a loved one with depression###Response: If someone you care about is dealing with depression, offer your support and understanding. Listen without judgment, encourage them to seek professional help, and be patient as they navigate their journey to recovery."},
{"inputs": "###Instruction: Reducing the stigma around mental health###Response: It's important to reduce the stigma surrounding mental health. Open conversations about depression and mental well-being can create a more supportive environment for those who are struggling."},
{"inputs": "###Instruction: Coping with depression and isolation###Response: Coping with depression and isolation can be challenging. Connect with friends and loved ones, participate in group activities, and seek professional therapy to combat feelings of loneliness and despair."},
{"inputs": "###Instruction: Setting achievable goals for managing depression###Response: Setting achievable goals is a practical way to manage depression. Start with small steps, celebrate your successes, and gradually work your way toward larger objectives. This can provide a sense of accomplishment and hope."},
{"inputs": "###Instruction: Strategies for improving sleep and managing depression###Response: Adequate sleep is crucial for managing depression. Create a bedtime routine, limit screen time before bed, and consider relaxation techniques to improve your sleep quality and overall mental health."},
{"inputs": "###Instruction: Seeking professional help for depression###Response: Depression is a serious condition that often requires professional intervention. Don't hesitate to reach out to therapists, counselors, or support groups to get the assistance and guidance you need for recovery."},
{"inputs": "###Instruction: Overcoming loneliness through self-connection###Response: Overcoming loneliness often starts with connecting with yourself. Engage in self-reflection, discover your interests, and invest time in self-care. Building a strong connection with yourself can help alleviate feelings of loneliness."},
{"inputs": "###Instruction: Cultivating social connections to combat loneliness###Response: Cultivating social connections is essential to combat loneliness. Seek opportunities to meet new people, join clubs or communities aligned with your interests, and engage in social activities to build meaningful relationships."},
{"inputs": "###Instruction: Coping with loneliness in old age###Response: Loneliness can be particularly challenging in old age. It's essential to stay socially active, reach out to family and friends, and consider volunteering or participating in senior programs to combat loneliness."},
{"inputs": "###Instruction: Using technology to reduce loneliness###Response: Technology can be a valuable tool to reduce loneliness. Stay connected with loved ones through video calls and social media, join online communities related to your interests, and explore virtual events to stay engaged and combat loneliness."},
{"inputs": "###Instruction: Managing loneliness during the holidays###Response: Loneliness during the holidays can be especially challenging. Reach out to friends and family, create your own holiday traditions, and consider volunteering or attending local events to make the season more enjoyable."},
{"inputs": "###Instruction: Seeking professional help for loneliness###Response: If loneliness is overwhelming and persistent, seeking professional help is a valid option. Therapists and counselors can provide strategies and support to address loneliness and improve your mental well-being."},
{"inputs": "###Instruction: Finding meaning and purpose to combat loneliness###Response: Finding meaning and purpose in your life can help combat loneliness. Pursue activities that align with your values, set goals, and seek personal growth to feel more fulfilled and connected to the world around you."},
{"inputs": "###Instruction: Fostering a pet to alleviate loneliness###Response: Fostering a pet can be a rewarding way to alleviate loneliness. Pets offer companionship and can help reduce feelings of isolation. Consider adopting a furry friend to improve your emotional well-being."},
{"inputs": "###Instruction: Managing anxiety through relaxation techniques###Response: Managing anxiety often involves relaxation techniques. Consider practices like deep breathing, meditation, or progressive muscle relaxation to calm your mind and reduce anxiety."},
{"inputs": "###Instruction: Overcoming social anxiety and building confidence###Response: Overcoming social anxiety requires building confidence. Gradual exposure to social situations, positive self-talk, and seeking support from friends or a therapist can help you become more socially comfortable."},
{"inputs": "###Instruction: Coping with test anxiety and improving performance###Response: Coping with test anxiety is essential for better performance. Effective study strategies, time management, and relaxation techniques can help reduce anxiety and enhance your test-taking abilities."},
{"inputs": "###Instruction: Reducing anxiety through physical activity###Response: Physical activity can help reduce anxiety. Regular exercise releases endorphins, which can improve your mood and alleviate anxiety. Incorporate exercise into your routine for better mental well-being."},
{"inputs": "###Instruction: Seeking therapy for anxiety and panic attacks###Response: If you experience frequent anxiety or panic attacks, seeking therapy is a valuable option. Cognitive-behavioral therapy and counseling can provide you with strategies to manage anxiety and regain control."},
{"inputs": "###Instruction: Managing work-related anxiety and achieving work-life balance###Response: Managing work-related anxiety is crucial for maintaining a work-life balance. Set clear boundaries, prioritize self-care, and practice stress-reduction techniques to reduce anxiety at work."},
{"inputs": "###Instruction: Overcoming driving anxiety and regaining independence###Response: Overcoming driving anxiety can help you regain independence. Gradual exposure to driving, relaxation techniques, and support from friends or a therapist can assist you in managing driving-related anxiety."},
{"inputs": "###Instruction: Using technology for anxiety management###Response: Technology can be a helpful tool for managing anxiety. Apps for mindfulness, stress reduction, and virtual therapy can assist you in tracking and alleviating anxiety symptoms."},
{"inputs": "###Instruction: Managing anger through self-awareness###Response: Managing anger begins with self-awareness. Recognizing your triggers and learning to control your reactions can help you respond to anger in a more constructive way."},
{"inputs": "###Instruction: Coping with anger in relationships###Response: Coping with anger in relationships is crucial. Effective communication, empathy, and conflict resolution skills can help you navigate and resolve anger-related issues with your loved ones."},
{"inputs": "###Instruction: Handling workplace anger and maintaining professionalism###Response: Handling workplace anger while maintaining professionalism is essential. Deep breathing, time management, and addressing issues through appropriate channels can help you manage workplace anger without compromising your job."},
{"inputs": "###Instruction: Overcoming road rage and promoting safe driving###Response: Overcoming road rage is important for safe driving. Implement strategies like practicing patience, listening to soothing music, and avoiding aggressive behavior on the road to ensure road safety."},
{"inputs": "###Instruction: Dealing with anger issues and seeking therapy###Response: If you're struggling with anger issues, seeking therapy can be beneficial. Anger management therapy and counseling can provide you with tools to control your anger and improve your mental well-being."},
{"inputs": "###Instruction: Nurturing emotional intelligence to manage anger###Response: Nurturing emotional intelligence is key to managing anger. Developing self-awareness, empathy, and effective emotional regulation can help you respond to anger in a healthier and more productive way."},
{"inputs": "###Instruction: Using anger as a motivator for positive change###Response: Anger can be channeled as a motivator for positive change. Instead of letting anger consume you, use it as a driving force to address issues, advocate for change, and make a difference in your life or the world."},
{"inputs": "###Instruction: Self-Harm Prevention Strategies###Response: Preventing self-harm is essential for your well-being. Reach out to a trusted friend or family member when you're struggling, create a self-care plan, and seek professional help if needed."},
{"inputs": "###Instruction: Coping with self-harm urges###Response: Coping with self-harm urges can be challenging. Distract yourself with a healthy activity, like drawing, journaling, or exercise, and remember that seeking support from a therapist or counselor is a positive step."},
{"inputs": "###Instruction: Building a support network for self-harm recovery###Response: Building a strong support network is vital for self-harm recovery. Open up to friends and family about your struggles, attend support groups, and connect with professionals who specialize in self-harm recovery."},
{"inputs": "###Instruction: Identifying triggers and warning signs###Response: Identifying triggers and warning signs is a crucial part of self-harm prevention. Keep a journal to track your emotions and situations that lead to self-harm urges, and work on strategies to avoid or cope with these triggers."},
{"inputs": "###Instruction: Seeking professional help for self-harm recovery###Response: If you're struggling with self-harm, seeking professional help is essential. Therapists, counselors, and mental health experts can provide you with the guidance and support needed to overcome self-harm behaviors."},
{"inputs": "###Instruction: Developing healthy coping mechanisms###Response: Developing healthy coping mechanisms is key to self-harm prevention. Explore alternative ways to manage stress and emotions, such as mindfulness, art therapy, or meditation, to replace self-harm behaviors."}
|
[] |
[
"TAGS\n#region-us \n"
] |
[
6
] |
[
"passage: TAGS\n#region-us \n"
] |
8bd770fc2d001c6b41e81f7e171a5def58a16b3f
|
# Dataset Card for Evaluation run of RWKV/rwkv-raven-14b
## Dataset Description
- **Homepage:**
- **Repository:** https://huggingface.co/RWKV/rwkv-raven-14b
- **Paper:**
- **Leaderboard:** https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard
- **Point of Contact:** [email protected]
### Dataset Summary
Dataset automatically created during the evaluation run of model [RWKV/rwkv-raven-14b](https://huggingface.co/RWKV/rwkv-raven-14b) on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).
The dataset is composed of 3 configuration, each one coresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.
An additional configuration "results" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)).
To load the details from a run, you can for instance do the following:
```python
from datasets import load_dataset
data = load_dataset("open-llm-leaderboard/details_RWKV__rwkv-raven-14b",
"harness_winogrande_5",
split="train")
```
## Latest results
These are the [latest results from run 2023-10-24T07:58:24.994437](https://huggingface.co/datasets/open-llm-leaderboard/details_RWKV__rwkv-raven-14b/blob/main/results_2023-10-24T07-58-24.994437.json)(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the "latest" split for each eval):
```python
{
"all": {
"em": 0.038485738255033555,
"em_stderr": 0.0019700054489528757,
"f1": 0.11610738255033538,
"f1_stderr": 0.002601381927911386,
"acc": 0.34407897925588793,
"acc_stderr": 0.008608321578959646
},
"harness|drop|3": {
"em": 0.038485738255033555,
"em_stderr": 0.0019700054489528757,
"f1": 0.11610738255033538,
"f1_stderr": 0.002601381927911386
},
"harness|gsm8k|5": {
"acc": 0.02122820318423048,
"acc_stderr": 0.003970449129848635
},
"harness|winogrande|5": {
"acc": 0.6669297553275454,
"acc_stderr": 0.013246194028070656
}
}
```
### Supported Tasks and Leaderboards
[More Information Needed]
### Languages
[More Information Needed]
## Dataset Structure
### Data Instances
[More Information Needed]
### Data Fields
[More Information Needed]
### Data Splits
[More Information Needed]
## Dataset Creation
### Curation Rationale
[More Information Needed]
### Source Data
#### Initial Data Collection and Normalization
[More Information Needed]
#### Who are the source language producers?
[More Information Needed]
### Annotations
#### Annotation process
[More Information Needed]
#### Who are the annotators?
[More Information Needed]
### Personal and Sensitive Information
[More Information Needed]
## Considerations for Using the Data
### Social Impact of Dataset
[More Information Needed]
### Discussion of Biases
[More Information Needed]
### Other Known Limitations
[More Information Needed]
## Additional Information
### Dataset Curators
[More Information Needed]
### Licensing Information
[More Information Needed]
### Citation Information
[More Information Needed]
### Contributions
[More Information Needed]
|
open-llm-leaderboard/details_RWKV__rwkv-raven-14b
|
[
"region:us"
] |
2023-10-24T06:58:28+00:00
|
{"pretty_name": "Evaluation run of RWKV/rwkv-raven-14b", "dataset_summary": "Dataset automatically created during the evaluation run of model [RWKV/rwkv-raven-14b](https://huggingface.co/RWKV/rwkv-raven-14b) on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).\n\nThe dataset is composed of 3 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)).\n\nTo load the details from a run, you can for instance do the following:\n```python\nfrom datasets import load_dataset\ndata = load_dataset(\"open-llm-leaderboard/details_RWKV__rwkv-raven-14b\",\n\t\"harness_winogrande_5\",\n\tsplit=\"train\")\n```\n\n## Latest results\n\nThese are the [latest results from run 2023-10-24T07:58:24.994437](https://huggingface.co/datasets/open-llm-leaderboard/details_RWKV__rwkv-raven-14b/blob/main/results_2023-10-24T07-58-24.994437.json)(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):\n\n```python\n{\n \"all\": {\n \"em\": 0.038485738255033555,\n \"em_stderr\": 0.0019700054489528757,\n \"f1\": 0.11610738255033538,\n \"f1_stderr\": 0.002601381927911386,\n \"acc\": 0.34407897925588793,\n \"acc_stderr\": 0.008608321578959646\n },\n \"harness|drop|3\": {\n \"em\": 0.038485738255033555,\n \"em_stderr\": 0.0019700054489528757,\n \"f1\": 0.11610738255033538,\n \"f1_stderr\": 0.002601381927911386\n },\n \"harness|gsm8k|5\": {\n \"acc\": 0.02122820318423048,\n \"acc_stderr\": 0.003970449129848635\n },\n \"harness|winogrande|5\": {\n \"acc\": 0.6669297553275454,\n \"acc_stderr\": 0.013246194028070656\n }\n}\n```", "repo_url": "https://huggingface.co/RWKV/rwkv-raven-14b", "leaderboard_url": "https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard", "point_of_contact": "[email protected]", "configs": [{"config_name": "harness_drop_3", "data_files": [{"split": "2023_10_24T07_58_24.994437", "path": ["**/details_harness|drop|3_2023-10-24T07-58-24.994437.parquet"]}, {"split": "latest", "path": ["**/details_harness|drop|3_2023-10-24T07-58-24.994437.parquet"]}]}, {"config_name": "harness_gsm8k_5", "data_files": [{"split": "2023_10_24T07_58_24.994437", "path": ["**/details_harness|gsm8k|5_2023-10-24T07-58-24.994437.parquet"]}, {"split": "latest", "path": ["**/details_harness|gsm8k|5_2023-10-24T07-58-24.994437.parquet"]}]}, {"config_name": "harness_winogrande_5", "data_files": [{"split": "2023_10_24T07_58_24.994437", "path": ["**/details_harness|winogrande|5_2023-10-24T07-58-24.994437.parquet"]}, {"split": "latest", "path": ["**/details_harness|winogrande|5_2023-10-24T07-58-24.994437.parquet"]}]}, {"config_name": "results", "data_files": [{"split": "2023_10_24T07_58_24.994437", "path": ["results_2023-10-24T07-58-24.994437.parquet"]}, {"split": "latest", "path": ["results_2023-10-24T07-58-24.994437.parquet"]}]}]}
|
2023-10-24T06:58:35+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for Evaluation run of RWKV/rwkv-raven-14b
## Dataset Description
- Homepage:
- Repository: URL
- Paper:
- Leaderboard: URL
- Point of Contact: clementine@URL
### Dataset Summary
Dataset automatically created during the evaluation run of model RWKV/rwkv-raven-14b on the Open LLM Leaderboard.
The dataset is composed of 3 configuration, each one coresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.
An additional configuration "results" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the Open LLM Leaderboard).
To load the details from a run, you can for instance do the following:
## Latest results
These are the latest results from run 2023-10-24T07:58:24.994437(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the "latest" split for each eval):
### Supported Tasks and Leaderboards
### Languages
## Dataset Structure
### Data Instances
### Data Fields
### Data Splits
## Dataset Creation
### Curation Rationale
### Source Data
#### Initial Data Collection and Normalization
#### Who are the source language producers?
### Annotations
#### Annotation process
#### Who are the annotators?
### Personal and Sensitive Information
## Considerations for Using the Data
### Social Impact of Dataset
### Discussion of Biases
### Other Known Limitations
## Additional Information
### Dataset Curators
### Licensing Information
### Contributions
|
[
"# Dataset Card for Evaluation run of RWKV/rwkv-raven-14b",
"## Dataset Description\n\n- Homepage: \n- Repository: URL\n- Paper: \n- Leaderboard: URL\n- Point of Contact: clementine@URL",
"### Dataset Summary\n\nDataset automatically created during the evaluation run of model RWKV/rwkv-raven-14b on the Open LLM Leaderboard.\n\nThe dataset is composed of 3 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:",
"## Latest results\n\nThese are the latest results from run 2023-10-24T07:58:24.994437(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):",
"### Supported Tasks and Leaderboards",
"### Languages",
"## Dataset Structure",
"### Data Instances",
"### Data Fields",
"### Data Splits",
"## Dataset Creation",
"### Curation Rationale",
"### Source Data",
"#### Initial Data Collection and Normalization",
"#### Who are the source language producers?",
"### Annotations",
"#### Annotation process",
"#### Who are the annotators?",
"### Personal and Sensitive Information",
"## Considerations for Using the Data",
"### Social Impact of Dataset",
"### Discussion of Biases",
"### Other Known Limitations",
"## Additional Information",
"### Dataset Curators",
"### Licensing Information",
"### Contributions"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for Evaluation run of RWKV/rwkv-raven-14b",
"## Dataset Description\n\n- Homepage: \n- Repository: URL\n- Paper: \n- Leaderboard: URL\n- Point of Contact: clementine@URL",
"### Dataset Summary\n\nDataset automatically created during the evaluation run of model RWKV/rwkv-raven-14b on the Open LLM Leaderboard.\n\nThe dataset is composed of 3 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:",
"## Latest results\n\nThese are the latest results from run 2023-10-24T07:58:24.994437(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):",
"### Supported Tasks and Leaderboards",
"### Languages",
"## Dataset Structure",
"### Data Instances",
"### Data Fields",
"### Data Splits",
"## Dataset Creation",
"### Curation Rationale",
"### Source Data",
"#### Initial Data Collection and Normalization",
"#### Who are the source language producers?",
"### Annotations",
"#### Annotation process",
"#### Who are the annotators?",
"### Personal and Sensitive Information",
"## Considerations for Using the Data",
"### Social Impact of Dataset",
"### Discussion of Biases",
"### Other Known Limitations",
"## Additional Information",
"### Dataset Curators",
"### Licensing Information",
"### Contributions"
] |
[
6,
21,
31,
169,
67,
10,
4,
6,
6,
5,
5,
5,
7,
4,
10,
10,
5,
5,
9,
8,
8,
7,
8,
7,
5,
6,
6,
5
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for Evaluation run of RWKV/rwkv-raven-14b## Dataset Description\n\n- Homepage: \n- Repository: URL\n- Paper: \n- Leaderboard: URL\n- Point of Contact: clementine@URL### Dataset Summary\n\nDataset automatically created during the evaluation run of model RWKV/rwkv-raven-14b on the Open LLM Leaderboard.\n\nThe dataset is composed of 3 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:## Latest results\n\nThese are the latest results from run 2023-10-24T07:58:24.994437(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):### Supported Tasks and Leaderboards### Languages## Dataset Structure### Data Instances### Data Fields### Data Splits## Dataset Creation### Curation Rationale### Source Data#### Initial Data Collection and Normalization#### Who are the source language producers?### Annotations#### Annotation process#### Who are the annotators?### Personal and Sensitive Information## Considerations for Using the Data### Social Impact of Dataset### Discussion of Biases### Other Known Limitations## Additional Information### Dataset Curators### Licensing Information### Contributions"
] |
4cbb1f2a1ba0252cdaa364e8bd26dc1d603280fe
|
# Dataset Card for "closed-QA-en"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
nlplabtdtu/closed-QA-en
|
[
"region:us"
] |
2023-10-24T07:08:55+00:00
|
{"dataset_info": {"features": [{"name": "id", "dtype": "string"}, {"name": "question", "dtype": "string"}, {"name": "answer", "dtype": "string"}, {"name": "context", "dtype": "string"}, {"name": "hint", "dtype": "string"}, {"name": "ok", "dtype": "bool"}, {"name": "english", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 9347077, "num_examples": 4169}], "download_size": 5401215, "dataset_size": 9347077}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T07:50:55+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "closed-QA-en"
More Information needed
|
[
"# Dataset Card for \"closed-QA-en\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"closed-QA-en\"\n\nMore Information needed"
] |
[
6,
16
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"closed-QA-en\"\n\nMore Information needed"
] |
6a1489c88ba570d49cf8a98a02999de5b2356cac
|
# Dataset Card for UltraChat 200k
## Dataset Description
This is a heavily filtered version of the [UltraChat](https://github.com/thunlp/UltraChat) dataset and was used to train [Zephyr-7B-β](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta), a state of the art 7b chat model.
The original datasets consists of 1.4M dialogues generated by ChatGPT and spanning a wide range of topics. To create `UltraChat 200k`, we applied the following logic:
- Selection of a subset of data for faster supervised fine tuning.
- Truecasing of the dataset, as we observed around 5% of the data contained grammatical errors like "Hello. how are you?" instead of "Hello. How are you?"
- Removal of dialogues where the assistant replies with phrases like "I do not have emotions" or "I don't have opinions", even for fact-based prompts that don't involve either.
## Dataset Structure
The dataset has four splits, suitable for:
* Supervised fine-tuning (`sft`).
* Generation ranking (`gen`) via techniques like rejection sampling or PPO.
The number of examples per split is shown as follows:
| train_sft | test_sft | train_gen | test_gen |
|:-------:|:-----------:|:-----:| :-----:|
| 207865 | 23110 | 256032 | 28304 |
The dataset is stored in parquet format with each entry using the following schema:
```
{
"prompt": "Create a fully-developed protagonist who is challenged to survive within a dystopian society under the rule of a tyrant. ...",
"messages":[
{
"content": "Create a fully-developed protagonist who is challenged to survive within a dystopian society under the rule of a tyrant. ...",
"role": "user"
},
{
"content": "Name: Ava\n\n Ava was just 16 years old when the world as she knew it came crashing down. The government had collapsed, leaving behind a chaotic and lawless society. ...",
"role": "assistant"
},
{
"content": "Wow, Ava's story is so intense and inspiring! Can you provide me with more details. ...",
"role": "user"
},
{
"content": "Certainly! ....",
"role": "assistant"
},
{
"content": "That's really interesting! I would love to hear more...",
"role": "user"
}
{
"content": "Certainly! ....",
"role": "assistant"
},
],
"prompt_id": "d938b65dfe31f05f80eb8572964c6673eddbd68eff3db6bd234d7f1e3b86c2af"
}
```
## Citation
If you find this dataset is useful in your work, please cite the original UltraChat dataset:
```
@misc{ding2023enhancing,
title={Enhancing Chat Language Models by Scaling High-quality Instructional Conversations},
author={Ning Ding and Yulin Chen and Bokai Xu and Yujia Qin and Zhi Zheng and Shengding Hu and Zhiyuan Liu and Maosong Sun and Bowen Zhou},
year={2023},
eprint={2305.14233},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
You may also wish to cite the Zephyr 7B technical report:
```
@misc{tunstall2023zephyr,
title={Zephyr: Direct Distillation of LM Alignment},
author={Lewis Tunstall and Edward Beeching and Nathan Lambert and Nazneen Rajani and Kashif Rasul and Younes Belkada and Shengyi Huang and Leandro von Werra and Clémentine Fourrier and Nathan Habib and Nathan Sarrazin and Omar Sanseviero and Alexander M. Rush and Thomas Wolf},
year={2023},
eprint={2310.16944},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
|
HuggingFaceH4/ultrachat_200k
|
[
"task_categories:conversational",
"task_categories:text-generation",
"size_categories:100K<n<1M",
"language:en",
"license:mit",
"arxiv:2305.14233",
"arxiv:2310.16944",
"region:us"
] |
2023-10-24T07:24:57+00:00
|
{"language": ["en"], "license": "mit", "size_categories": ["100K<n<1M"], "task_categories": ["conversational", "text-generation"], "pretty_name": "UltraChat 200k", "configs": [{"config_name": "default", "data_files": [{"split": "train_sft", "path": "data/train_sft-*"}, {"split": "test_sft", "path": "data/test_sft-*"}, {"split": "train_gen", "path": "data/train_gen-*"}, {"split": "test_gen", "path": "data/test_gen-*"}]}], "dataset_info": {"features": [{"name": "prompt", "dtype": "string"}, {"name": "prompt_id", "dtype": "string"}, {"name": "messages", "list": [{"name": "content", "dtype": "string"}, {"name": "role", "dtype": "string"}]}], "splits": [{"name": "train_sft", "num_bytes": 1397058554, "num_examples": 207865}, {"name": "test_sft", "num_bytes": 154695659, "num_examples": 23110}, {"name": "train_gen", "num_bytes": 1347396812, "num_examples": 256032}, {"name": "test_gen", "num_bytes": 148276089, "num_examples": 28304}], "download_size": 1624049723, "dataset_size": 3047427114}}
|
2023-10-27T07:53:22+00:00
|
[
"2305.14233",
"2310.16944"
] |
[
"en"
] |
TAGS
#task_categories-conversational #task_categories-text-generation #size_categories-100K<n<1M #language-English #license-mit #arxiv-2305.14233 #arxiv-2310.16944 #region-us
|
Dataset Card for UltraChat 200k
===============================
Dataset Description
-------------------
This is a heavily filtered version of the UltraChat dataset and was used to train Zephyr-7B-β, a state of the art 7b chat model.
The original datasets consists of 1.4M dialogues generated by ChatGPT and spanning a wide range of topics. To create 'UltraChat 200k', we applied the following logic:
* Selection of a subset of data for faster supervised fine tuning.
* Truecasing of the dataset, as we observed around 5% of the data contained grammatical errors like "Hello. how are you?" instead of "Hello. How are you?"
* Removal of dialogues where the assistant replies with phrases like "I do not have emotions" or "I don't have opinions", even for fact-based prompts that don't involve either.
Dataset Structure
-----------------
The dataset has four splits, suitable for:
* Supervised fine-tuning ('sft').
* Generation ranking ('gen') via techniques like rejection sampling or PPO.
The number of examples per split is shown as follows:
The dataset is stored in parquet format with each entry using the following schema:
If you find this dataset is useful in your work, please cite the original UltraChat dataset:
You may also wish to cite the Zephyr 7B technical report:
|
[] |
[
"TAGS\n#task_categories-conversational #task_categories-text-generation #size_categories-100K<n<1M #language-English #license-mit #arxiv-2305.14233 #arxiv-2310.16944 #region-us \n"
] |
[
66
] |
[
"passage: TAGS\n#task_categories-conversational #task_categories-text-generation #size_categories-100K<n<1M #language-English #license-mit #arxiv-2305.14233 #arxiv-2310.16944 #region-us \n"
] |
0ed5ea08bb92453401cd4e3c535cea412ffa95ec
|
# Dataset Card for "llama2_7b_fine_tuning_complete_dataset_v2"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
hemantk089/llama2_7b_fine_tuning_complete_dataset_v2
|
[
"region:us"
] |
2023-10-24T07:36:11+00:00
|
{"dataset_info": {"features": [{"name": "Description", "dtype": "string"}, {"name": "Relevances", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 259139, "num_examples": 917}], "download_size": 58941, "dataset_size": 259139}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T07:36:12+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "llama2_7b_fine_tuning_complete_dataset_v2"
More Information needed
|
[
"# Dataset Card for \"llama2_7b_fine_tuning_complete_dataset_v2\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"llama2_7b_fine_tuning_complete_dataset_v2\"\n\nMore Information needed"
] |
[
6,
30
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"llama2_7b_fine_tuning_complete_dataset_v2\"\n\nMore Information needed"
] |
92f25d6c73c56ef6b6cee4e33cf1546ee067ae5d
|
# Dataset Card for "LayoutLMv3_dataset_filtred"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
ayoub999/LayoutLMv3_dataset_filtred
|
[
"region:us"
] |
2023-10-24T07:39:45+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "test", "path": "data/test-*"}]}], "dataset_info": {"features": [{"name": "id", "dtype": "string"}, {"name": "image", "dtype": "image"}, {"name": "bboxes", "sequence": {"sequence": "int64"}}, {"name": "ner_tags", "sequence": {"class_label": {"names": {"0": "O", "1": "Ref", "2": "NumFa", "3": "Fourniss", "4": "DateFa", "5": "DateLim", "6": "TotalHT", "7": "TVA", "8": "TotalTTc", "9": "unitP", "10": "Qt", "11": "TVAP", "12": "descp"}}}}, {"name": "tokens", "sequence": "string"}], "splits": [{"name": "train", "num_bytes": 942956.6666666666, "num_examples": 2}, {"name": "test", "num_bytes": 183522.0, "num_examples": 1}], "download_size": 0, "dataset_size": 1126478.6666666665}}
|
2023-10-27T14:28:25+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "LayoutLMv3_dataset_filtred"
More Information needed
|
[
"# Dataset Card for \"LayoutLMv3_dataset_filtred\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"LayoutLMv3_dataset_filtred\"\n\nMore Information needed"
] |
[
6,
22
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"LayoutLMv3_dataset_filtred\"\n\nMore Information needed"
] |
c1c6363101d7bcef56288d3b2f8275e6aa39e96f
|
# Dataset Card for "llama2_7b_fine_tuning_complete_dataset_v3"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
hemantk089/llama2_7b_fine_tuning_complete_dataset_v3
|
[
"region:us"
] |
2023-10-24T07:43:39+00:00
|
{"dataset_info": {"features": [{"name": "text", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 287566, "num_examples": 917}], "download_size": 77717, "dataset_size": 287566}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T07:43:40+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "llama2_7b_fine_tuning_complete_dataset_v3"
More Information needed
|
[
"# Dataset Card for \"llama2_7b_fine_tuning_complete_dataset_v3\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"llama2_7b_fine_tuning_complete_dataset_v3\"\n\nMore Information needed"
] |
[
6,
30
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"llama2_7b_fine_tuning_complete_dataset_v3\"\n\nMore Information needed"
] |
e8f83905f5b86b275ca35c8a28beca8f41e68f48
|
<p align="center">
<img src="https://github.com/iamshnoo/weathub/blob/main/assets/dalle3_weathub.png?raw=true" width="250" height="250">
</p>
# Dataset Card for "WEATHub"
This dataset corresponds to the data described in the paper "Global Voices, Local Biases: Socio-Cultural Prejudices across Languages"
accepted to EMNLP 2023.
## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
- **Homepage:** [Website](https://iamshnoo.github.io/global_voices_local_biases/)
- **Repository:** [GitHub](https://github.com/iamshnoo/weathub)
- **Paper:** https://arxiv.org/abs/2310.17586
- **Point of Contact:** Anjishnu Mukherjee
### Dataset Summary
WEATHub is a dataset containing 24 languages. It contains words organized into groups of (target1, target2, attribute1, attribute2)
to measure the association target1:target2 :: attribute1:attribute2. For example target1 can be insects, target2 can be flowers. And we
might be trying to measure whether we find insects or flowers pleasant or unpleasant. The measurement of word associations is quantified
using the WEAT metric in our paper. It is a metric that calculates an effect size (Cohen's d) and also provides a p-value (to measure
statistical significance of the results). In our paper, we use word embeddings from language models to perform these tests and understand
biased associations in language models across different languages.
### Supported Tasks and Leaderboards
- `bias_eval` : The dataset is used to measure biased associations.
- This particular task isn't a standard task that is currently supported.
### Languages
The languages (in alphabetical order of language codes) are: Arabic (ar), Bengali (bn), Sorani Kurdish (ckb), Danish (da), German (de),
Greek (el), Spanish (es), Persian (fa), French (fr), Hindi (hi), Italian (it), Japanese (ja), Korean (ko), Kurmanji Kurdish (ku),
Marathi (mr), Punjabi (pa), Russian (ru), Telugu (te), Thai (th), Tagalog (tl), Turkish (tr), Urdu (ur), Vietnamese (vi), Chinese (zh).
## Dataset Structure
### Data Instances
An example instance is of the form:
```json
{
'attr1': {'category': 'Career',
'examples': ['σύμβουλος', 'διεύθυνση', 'επαγγελματίας', 'εταιρεία', 'μισθός', 'γραφείο', 'επιχείρηση', 'καριέρα', 'διευθύνων σύμβουλος']},
'attr2': {'category': 'Family',
'examples': ['σπίτι', 'γονείς', 'παιδιά', 'οικογένεια', 'ξαδερφια', 'γάμος', 'γάμος', 'συγγενείς']},
'targ1': {'category': 'MaleNames',
'examples': ['Αλέξανδρος', 'Δημήτρης', 'Γιώργος', 'Κώστας', 'Νίκος', 'Παναγιώτης', 'Σπύρος', 'Θοδωρής']},
'targ2': {'category': 'FemaleNames',
'examples': ['Αθηνά', 'Ελένη', 'Κατερίνα', 'Μαρία', 'Ευαγγελία', 'Αναστασία', 'Δέσποινα', 'Χριστίνα']},
'language': 'el',
'weat': 'WEAT6'
}
```
### Data Fields
- A single data point has the following features:
- name: language (corresponding to the language codes given above)
- name: weat (ID corresponding to a WEAT category)
- name: attr1.category (a descriptive name for attribute 1)
- name: attr1.examples (list of words for attribute 1)
- name: attr2.category (a descriptive name for attribute 2)
- name: attr2.examples (list of words for attribute 2)
- name: targ1.category (a descriptive name for target 1)
- name: targ1.examples (list of words for target 1)
- name: targ2.category (a descriptive name for target 2)
- name: targ2.examples (list of words for target 2)
- All the features are stored as strings. The examples represent lists of strings.
### Data Splits
- The dataset is divided into 3 splits as per the description in our paper:
- original_weat - described in Table 1 of our paper, this corresponds to the original WEAT categories as given by Caliskan et al. in their
seminal work from 2017 (Semantics derived automatically from language corpora contain human-like biases)
- new_human_biases - described in Table 2 of our paper, this corresponds to contemporary dimensions of bias that are more human-centric in
modern society.
- india_specific_biases - These contain data corresponding to india specific bias dimensions as described in the paper (Socially Aware Bias Measurements for Hindi Language Representations)
from NAACL '22 by Malik et al.
## Dataset Creation
### Curation Rationale
This dataset is intended to be used for measuring intrinsic biases in word embeddings obtained from language models.
### Source Data
#### Initial Data Collection and Normalization
Described in details in section 2 of our paper. Briefly, for existing weat categories, we use human annotations to improve the quality of the
translated WEAT word lists. For new weat categories, we research possible relevant dimensions thoroughly and come up with words after thorough
discussions with our annotators.
#### Who are the source language producers?
Data for each of the language is from native speakers of that language. All annotators who participated in our study are native speakers of
their respective languages and have at least college-level education background.
### Annotations
#### Annotation process
Described in details in section 2 of our paper. Word level annotations.
To collect annotated data in various languages, we provide our annotators with the English words and their corresponding automatic translation
, separated by WEAT category. We provide instructions to verify the accuracy of the translations and provide corrected versions for any
inaccuracies. Additionally, we ask annotators to provide grammatically gendered forms of words, if applicable, or multiple translations
of a word, if necessary.
#### Who are the annotators?
All annotators who participated in our study are native speakers of
their respective languages and have at least college-level education background.
### Personal and Sensitive Information
Since this dataset tries to measure biased associations at the word level, there may be some word level biases that are sensitive to certain
groups.
## Considerations for Using the Data
### Social Impact of Dataset
This dataset should be a starting point for measuring word level biased associations in a multilingual setting, which has not been explored
in much depth in recent literature.
### Discussion of Biases
This dataset represents word level information used for measuring biases. Since these are annotated by humans, they may to certain extent reflect
the biases that they hold at an individual level.
### Other Known Limitations
- For most of the languages in our dataset WEATHub, we had access to at least two annotators for cross-verifying the accuracy of
the human translations to determine if the translated words fit into the context of that particular WEAT category.
However, for some languages, we only have one annotator per language, so this might mean that for some languages the data may represent
the biases of that individual annotator even though those biases are somewhat also reflected by Google Translate so it isn't completely
an individualistic issue.
- While we have tried to cover as many languages from the global South as possible, we acknowledge that 24 languages are indeed a
tiny proportion of the 7000 languages in the world, some of which do not even have text representations.
- WEAT can be an unreliable metric for contextualized embeddings from transformer models. We need better metrics to study intrinsic biases in
transformer models. We believe the target and attribute pairs we provide as part of WEATHub in multiple languages is an important step
towards a better multilingual metric for evaluating intrinsic biases in language models.
## Additional Information
### Dataset Curators
This dataset was curated by Anjishnu Mukherjee, Chahat Raj, Ziwei Zhu and Antonios Anastasopoulos for their EMNLP paper while the first two authors were
pursuing their PhD at George Mason University. This work
was generously supported by the National Science Foundation under award IIS-2327143. Computational resources for experiments were provided by the
Office of of Research Computing at George Mason University (URL: https://orc.gmu.edu) and funded in part by grants from the
National Science Foundation (Awards Number 1625039 and 2018631).
### Licensing Information
Currently this dataset is released under CC-4.0 (might need to update this if required)
### Citation Information
@inproceedings{mukherjee-etal-2023-global,
title = "{G}lobal {V}oices, Local Biases: Socio-Cultural Prejudices across Languages",
author = "Mukherjee, Anjishnu and
Raj, Chahat and
Zhu, Ziwei and
Anastasopoulos, Antonios",
editor = "Bouamor, Houda and
Pino, Juan and
Bali, Kalika",
booktitle = "Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2023",
address = "Singapore",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.emnlp-main.981",
doi = "10.18653/v1/2023.emnlp-main.981",
pages = "15828--15845",
abstract = "Human biases are ubiquitous but not uniform: disparities exist across linguistic, cultural, and societal borders. As large amounts of recent literature suggest, language models (LMs) trained on human data can reflect and often amplify the effects of these social biases. However, the vast majority of existing studies on bias are heavily skewed towards Western and European languages. In this work, we scale the Word Embedding Association Test (WEAT) to 24 languages, enabling broader studies and yielding interesting findings about LM bias. We additionally enhance this data with culturally relevant information for each language, capturing local contexts on a global scale. Further, to encompass more widely prevalent societal biases, we examine new bias dimensions across toxicity, ableism, and more. Moreover, we delve deeper into the Indian linguistic landscape, conducting a comprehensive regional bias analysis across six prevalent Indian languages. Finally, we highlight the significance of these social biases and the new dimensions through an extensive comparison of embedding methods, reinforcing the need to address them in pursuit of more equitable language models.",
}
### Contributions
Thanks to [@iamshnoo](https://github.com/iamshnoo) for adding this dataset.
|
iamshnoo/WEATHub
|
[
"language:ar",
"language:bn",
"language:ckb",
"language:da",
"language:de",
"language:el",
"language:es",
"language:fa",
"language:fr",
"language:hi",
"language:it",
"language:ja",
"language:ko",
"language:ku",
"language:mr",
"language:pa",
"language:ru",
"language:te",
"language:th",
"language:tl",
"language:tr",
"language:ur",
"language:vi",
"language:zh",
"license:cc-by-4.0",
"arxiv:2310.17586",
"region:us"
] |
2023-10-24T07:43:49+00:00
|
{"language": ["ar", "bn", "ckb", "da", "de", "el", "es", "fa", "fr", "hi", "it", "ja", "ko", "ku", "mr", "pa", "ru", "te", "th", "tl", "tr", "ur", "vi", "zh"], "license": "cc-by-4.0", "pretty_name": "weathub", "configs": [{"config_name": "default", "data_files": [{"split": "original_weat", "path": "data/original_weat-*"}, {"split": "new_human_biases", "path": "data/new_human_biases-*"}, {"split": "india_specific_biases", "path": "data/india_specific_biases-*"}]}], "dataset_info": {"features": [{"name": "language", "dtype": "string"}, {"name": "weat", "dtype": "string"}, {"name": "attr1.category", "dtype": "string"}, {"name": "attr1.examples", "sequence": "string"}, {"name": "attr2.category", "dtype": "string"}, {"name": "attr2.examples", "sequence": "string"}, {"name": "targ1.category", "dtype": "string"}, {"name": "targ1.examples", "sequence": "string"}, {"name": "targ2.category", "dtype": "string"}, {"name": "targ2.examples", "sequence": "string"}], "splits": [{"name": "original_weat", "num_bytes": 173260, "num_examples": 150}, {"name": "new_human_biases", "num_bytes": 185406, "num_examples": 175}, {"name": "india_specific_biases", "num_bytes": 49647, "num_examples": 77}], "download_size": 208074, "dataset_size": 408313}}
|
2023-12-18T09:59:33+00:00
|
[
"2310.17586"
] |
[
"ar",
"bn",
"ckb",
"da",
"de",
"el",
"es",
"fa",
"fr",
"hi",
"it",
"ja",
"ko",
"ku",
"mr",
"pa",
"ru",
"te",
"th",
"tl",
"tr",
"ur",
"vi",
"zh"
] |
TAGS
#language-Arabic #language-Bengali #language-Central Kurdish #language-Danish #language-German #language-Modern Greek (1453-) #language-Spanish #language-Persian #language-French #language-Hindi #language-Italian #language-Japanese #language-Korean #language-Kurdish #language-Marathi #language-Panjabi #language-Russian #language-Telugu #language-Thai #language-Tagalog #language-Turkish #language-Urdu #language-Vietnamese #language-Chinese #license-cc-by-4.0 #arxiv-2310.17586 #region-us
|
<p align="center">
<img src="URL width="250" height="250">
</p>
# Dataset Card for "WEATHub"
This dataset corresponds to the data described in the paper "Global Voices, Local Biases: Socio-Cultural Prejudices across Languages"
accepted to EMNLP 2023.
## Table of Contents
- Table of Contents
- Dataset Description
- Dataset Summary
- Supported Tasks and Leaderboards
- Languages
- Dataset Structure
- Data Instances
- Data Fields
- Data Splits
- Dataset Creation
- Curation Rationale
- Source Data
- Annotations
- Personal and Sensitive Information
- Considerations for Using the Data
- Social Impact of Dataset
- Discussion of Biases
- Other Known Limitations
- Additional Information
- Dataset Curators
- Licensing Information
- Citation Information
- Contributions
## Dataset Description
- Homepage: Website
- Repository: GitHub
- Paper: URL
- Point of Contact: Anjishnu Mukherjee
### Dataset Summary
WEATHub is a dataset containing 24 languages. It contains words organized into groups of (target1, target2, attribute1, attribute2)
to measure the association target1:target2 :: attribute1:attribute2. For example target1 can be insects, target2 can be flowers. And we
might be trying to measure whether we find insects or flowers pleasant or unpleasant. The measurement of word associations is quantified
using the WEAT metric in our paper. It is a metric that calculates an effect size (Cohen's d) and also provides a p-value (to measure
statistical significance of the results). In our paper, we use word embeddings from language models to perform these tests and understand
biased associations in language models across different languages.
### Supported Tasks and Leaderboards
- 'bias_eval' : The dataset is used to measure biased associations.
- This particular task isn't a standard task that is currently supported.
### Languages
The languages (in alphabetical order of language codes) are: Arabic (ar), Bengali (bn), Sorani Kurdish (ckb), Danish (da), German (de),
Greek (el), Spanish (es), Persian (fa), French (fr), Hindi (hi), Italian (it), Japanese (ja), Korean (ko), Kurmanji Kurdish (ku),
Marathi (mr), Punjabi (pa), Russian (ru), Telugu (te), Thai (th), Tagalog (tl), Turkish (tr), Urdu (ur), Vietnamese (vi), Chinese (zh).
## Dataset Structure
### Data Instances
An example instance is of the form:
### Data Fields
- A single data point has the following features:
- name: language (corresponding to the language codes given above)
- name: weat (ID corresponding to a WEAT category)
- name: attr1.category (a descriptive name for attribute 1)
- name: attr1.examples (list of words for attribute 1)
- name: attr2.category (a descriptive name for attribute 2)
- name: attr2.examples (list of words for attribute 2)
- name: targ1.category (a descriptive name for target 1)
- name: targ1.examples (list of words for target 1)
- name: targ2.category (a descriptive name for target 2)
- name: targ2.examples (list of words for target 2)
- All the features are stored as strings. The examples represent lists of strings.
### Data Splits
- The dataset is divided into 3 splits as per the description in our paper:
- original_weat - described in Table 1 of our paper, this corresponds to the original WEAT categories as given by Caliskan et al. in their
seminal work from 2017 (Semantics derived automatically from language corpora contain human-like biases)
- new_human_biases - described in Table 2 of our paper, this corresponds to contemporary dimensions of bias that are more human-centric in
modern society.
- india_specific_biases - These contain data corresponding to india specific bias dimensions as described in the paper (Socially Aware Bias Measurements for Hindi Language Representations)
from NAACL '22 by Malik et al.
## Dataset Creation
### Curation Rationale
This dataset is intended to be used for measuring intrinsic biases in word embeddings obtained from language models.
### Source Data
#### Initial Data Collection and Normalization
Described in details in section 2 of our paper. Briefly, for existing weat categories, we use human annotations to improve the quality of the
translated WEAT word lists. For new weat categories, we research possible relevant dimensions thoroughly and come up with words after thorough
discussions with our annotators.
#### Who are the source language producers?
Data for each of the language is from native speakers of that language. All annotators who participated in our study are native speakers of
their respective languages and have at least college-level education background.
### Annotations
#### Annotation process
Described in details in section 2 of our paper. Word level annotations.
To collect annotated data in various languages, we provide our annotators with the English words and their corresponding automatic translation
, separated by WEAT category. We provide instructions to verify the accuracy of the translations and provide corrected versions for any
inaccuracies. Additionally, we ask annotators to provide grammatically gendered forms of words, if applicable, or multiple translations
of a word, if necessary.
#### Who are the annotators?
All annotators who participated in our study are native speakers of
their respective languages and have at least college-level education background.
### Personal and Sensitive Information
Since this dataset tries to measure biased associations at the word level, there may be some word level biases that are sensitive to certain
groups.
## Considerations for Using the Data
### Social Impact of Dataset
This dataset should be a starting point for measuring word level biased associations in a multilingual setting, which has not been explored
in much depth in recent literature.
### Discussion of Biases
This dataset represents word level information used for measuring biases. Since these are annotated by humans, they may to certain extent reflect
the biases that they hold at an individual level.
### Other Known Limitations
- For most of the languages in our dataset WEATHub, we had access to at least two annotators for cross-verifying the accuracy of
the human translations to determine if the translated words fit into the context of that particular WEAT category.
However, for some languages, we only have one annotator per language, so this might mean that for some languages the data may represent
the biases of that individual annotator even though those biases are somewhat also reflected by Google Translate so it isn't completely
an individualistic issue.
- While we have tried to cover as many languages from the global South as possible, we acknowledge that 24 languages are indeed a
tiny proportion of the 7000 languages in the world, some of which do not even have text representations.
- WEAT can be an unreliable metric for contextualized embeddings from transformer models. We need better metrics to study intrinsic biases in
transformer models. We believe the target and attribute pairs we provide as part of WEATHub in multiple languages is an important step
towards a better multilingual metric for evaluating intrinsic biases in language models.
## Additional Information
### Dataset Curators
This dataset was curated by Anjishnu Mukherjee, Chahat Raj, Ziwei Zhu and Antonios Anastasopoulos for their EMNLP paper while the first two authors were
pursuing their PhD at George Mason University. This work
was generously supported by the National Science Foundation under award IIS-2327143. Computational resources for experiments were provided by the
Office of of Research Computing at George Mason University (URL: URL) and funded in part by grants from the
National Science Foundation (Awards Number 1625039 and 2018631).
### Licensing Information
Currently this dataset is released under CC-4.0 (might need to update this if required)
@inproceedings{mukherjee-etal-2023-global,
title = "{G}lobal {V}oices, Local Biases: Socio-Cultural Prejudices across Languages",
author = "Mukherjee, Anjishnu and
Raj, Chahat and
Zhu, Ziwei and
Anastasopoulos, Antonios",
editor = "Bouamor, Houda and
Pino, Juan and
Bali, Kalika",
booktitle = "Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2023",
address = "Singapore",
publisher = "Association for Computational Linguistics",
url = "URL
doi = "10.18653/v1/URL-main.981",
pages = "15828--15845",
abstract = "Human biases are ubiquitous but not uniform: disparities exist across linguistic, cultural, and societal borders. As large amounts of recent literature suggest, language models (LMs) trained on human data can reflect and often amplify the effects of these social biases. However, the vast majority of existing studies on bias are heavily skewed towards Western and European languages. In this work, we scale the Word Embedding Association Test (WEAT) to 24 languages, enabling broader studies and yielding interesting findings about LM bias. We additionally enhance this data with culturally relevant information for each language, capturing local contexts on a global scale. Further, to encompass more widely prevalent societal biases, we examine new bias dimensions across toxicity, ableism, and more. Moreover, we delve deeper into the Indian linguistic landscape, conducting a comprehensive regional bias analysis across six prevalent Indian languages. Finally, we highlight the significance of these social biases and the new dimensions through an extensive comparison of embedding methods, reinforcing the need to address them in pursuit of more equitable language models.",
}
### Contributions
Thanks to @iamshnoo for adding this dataset.
|
[
"# Dataset Card for \"WEATHub\"\n\nThis dataset corresponds to the data described in the paper \"Global Voices, Local Biases: Socio-Cultural Prejudices across Languages\"\naccepted to EMNLP 2023.",
"## Table of Contents\n- Table of Contents\n- Dataset Description\n - Dataset Summary\n - Supported Tasks and Leaderboards\n - Languages\n- Dataset Structure\n - Data Instances\n - Data Fields\n - Data Splits\n- Dataset Creation\n - Curation Rationale\n - Source Data\n - Annotations\n - Personal and Sensitive Information\n- Considerations for Using the Data\n - Social Impact of Dataset\n - Discussion of Biases\n - Other Known Limitations\n- Additional Information\n - Dataset Curators\n - Licensing Information\n - Citation Information\n - Contributions",
"## Dataset Description\n\n- Homepage: Website\n- Repository: GitHub\n- Paper: URL\n- Point of Contact: Anjishnu Mukherjee",
"### Dataset Summary\n\nWEATHub is a dataset containing 24 languages. It contains words organized into groups of (target1, target2, attribute1, attribute2)\nto measure the association target1:target2 :: attribute1:attribute2. For example target1 can be insects, target2 can be flowers. And we \nmight be trying to measure whether we find insects or flowers pleasant or unpleasant. The measurement of word associations is quantified \nusing the WEAT metric in our paper. It is a metric that calculates an effect size (Cohen's d) and also provides a p-value (to measure\nstatistical significance of the results). In our paper, we use word embeddings from language models to perform these tests and understand \nbiased associations in language models across different languages.",
"### Supported Tasks and Leaderboards\n\n- 'bias_eval' : The dataset is used to measure biased associations.\n- This particular task isn't a standard task that is currently supported.",
"### Languages\n\nThe languages (in alphabetical order of language codes) are: Arabic (ar), Bengali (bn), Sorani Kurdish (ckb), Danish (da), German (de), \nGreek (el), Spanish (es), Persian (fa), French (fr), Hindi (hi), Italian (it), Japanese (ja), Korean (ko), Kurmanji Kurdish (ku), \nMarathi (mr), Punjabi (pa), Russian (ru), Telugu (te), Thai (th), Tagalog (tl), Turkish (tr), Urdu (ur), Vietnamese (vi), Chinese (zh).",
"## Dataset Structure",
"### Data Instances\n\nAn example instance is of the form:",
"### Data Fields\n\n- A single data point has the following features:\n - name: language (corresponding to the language codes given above)\n - name: weat (ID corresponding to a WEAT category)\n - name: attr1.category (a descriptive name for attribute 1)\n - name: attr1.examples (list of words for attribute 1)\n - name: attr2.category (a descriptive name for attribute 2)\n - name: attr2.examples (list of words for attribute 2)\n - name: targ1.category (a descriptive name for target 1)\n - name: targ1.examples (list of words for target 1)\n - name: targ2.category (a descriptive name for target 2)\n - name: targ2.examples (list of words for target 2)\n \n- All the features are stored as strings. The examples represent lists of strings.",
"### Data Splits\n\n- The dataset is divided into 3 splits as per the description in our paper:\n - original_weat - described in Table 1 of our paper, this corresponds to the original WEAT categories as given by Caliskan et al. in their\n seminal work from 2017 (Semantics derived automatically from language corpora contain human-like biases)\n - new_human_biases - described in Table 2 of our paper, this corresponds to contemporary dimensions of bias that are more human-centric in\n modern society.\n - india_specific_biases - These contain data corresponding to india specific bias dimensions as described in the paper (Socially Aware Bias Measurements for Hindi Language Representations)\n from NAACL '22 by Malik et al.",
"## Dataset Creation",
"### Curation Rationale\n\nThis dataset is intended to be used for measuring intrinsic biases in word embeddings obtained from language models.",
"### Source Data",
"#### Initial Data Collection and Normalization\n\nDescribed in details in section 2 of our paper. Briefly, for existing weat categories, we use human annotations to improve the quality of the\ntranslated WEAT word lists. For new weat categories, we research possible relevant dimensions thoroughly and come up with words after thorough\ndiscussions with our annotators.",
"#### Who are the source language producers?\n\nData for each of the language is from native speakers of that language. All annotators who participated in our study are native speakers of \ntheir respective languages and have at least college-level education background.",
"### Annotations",
"#### Annotation process\n\nDescribed in details in section 2 of our paper. Word level annotations.\nTo collect annotated data in various languages, we provide our annotators with the English words and their corresponding automatic translation\n, separated by WEAT category. We provide instructions to verify the accuracy of the translations and provide corrected versions for any \ninaccuracies. Additionally, we ask annotators to provide grammatically gendered forms of words, if applicable, or multiple translations \nof a word, if necessary.",
"#### Who are the annotators?\n\nAll annotators who participated in our study are native speakers of \ntheir respective languages and have at least college-level education background.",
"### Personal and Sensitive Information\n\nSince this dataset tries to measure biased associations at the word level, there may be some word level biases that are sensitive to certain \ngroups.",
"## Considerations for Using the Data",
"### Social Impact of Dataset\n\nThis dataset should be a starting point for measuring word level biased associations in a multilingual setting, which has not been explored \nin much depth in recent literature.",
"### Discussion of Biases\n\nThis dataset represents word level information used for measuring biases. Since these are annotated by humans, they may to certain extent reflect\nthe biases that they hold at an individual level.",
"### Other Known Limitations\n\n- For most of the languages in our dataset WEATHub, we had access to at least two annotators for cross-verifying the accuracy of\n the human translations to determine if the translated words fit into the context of that particular WEAT category.\n However, for some languages, we only have one annotator per language, so this might mean that for some languages the data may represent\n the biases of that individual annotator even though those biases are somewhat also reflected by Google Translate so it isn't completely\n an individualistic issue.\n- While we have tried to cover as many languages from the global South as possible, we acknowledge that 24 languages are indeed a\n tiny proportion of the 7000 languages in the world, some of which do not even have text representations.\n- WEAT can be an unreliable metric for contextualized embeddings from transformer models. We need better metrics to study intrinsic biases in\n transformer models. We believe the target and attribute pairs we provide as part of WEATHub in multiple languages is an important step\n towards a better multilingual metric for evaluating intrinsic biases in language models.",
"## Additional Information",
"### Dataset Curators\n\nThis dataset was curated by Anjishnu Mukherjee, Chahat Raj, Ziwei Zhu and Antonios Anastasopoulos for their EMNLP paper while the first two authors were \npursuing their PhD at George Mason University. This work\nwas generously supported by the National Science Foundation under award IIS-2327143. Computational resources for experiments were provided by the\nOffice of of Research Computing at George Mason University (URL: URL) and funded in part by grants from the \nNational Science Foundation (Awards Number 1625039 and 2018631).",
"### Licensing Information\n\nCurrently this dataset is released under CC-4.0 (might need to update this if required)\n\n\n\n@inproceedings{mukherjee-etal-2023-global,\n title = \"{G}lobal {V}oices, Local Biases: Socio-Cultural Prejudices across Languages\",\n author = \"Mukherjee, Anjishnu and\n Raj, Chahat and\n Zhu, Ziwei and\n Anastasopoulos, Antonios\",\n editor = \"Bouamor, Houda and\n Pino, Juan and\n Bali, Kalika\",\n booktitle = \"Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing\",\n month = dec,\n year = \"2023\",\n address = \"Singapore\",\n publisher = \"Association for Computational Linguistics\",\n url = \"URL\n doi = \"10.18653/v1/URL-main.981\",\n pages = \"15828--15845\",\n abstract = \"Human biases are ubiquitous but not uniform: disparities exist across linguistic, cultural, and societal borders. As large amounts of recent literature suggest, language models (LMs) trained on human data can reflect and often amplify the effects of these social biases. However, the vast majority of existing studies on bias are heavily skewed towards Western and European languages. In this work, we scale the Word Embedding Association Test (WEAT) to 24 languages, enabling broader studies and yielding interesting findings about LM bias. We additionally enhance this data with culturally relevant information for each language, capturing local contexts on a global scale. Further, to encompass more widely prevalent societal biases, we examine new bias dimensions across toxicity, ableism, and more. Moreover, we delve deeper into the Indian linguistic landscape, conducting a comprehensive regional bias analysis across six prevalent Indian languages. Finally, we highlight the significance of these social biases and the new dimensions through an extensive comparison of embedding methods, reinforcing the need to address them in pursuit of more equitable language models.\",\n}",
"### Contributions\n\nThanks to @iamshnoo for adding this dataset."
] |
[
"TAGS\n#language-Arabic #language-Bengali #language-Central Kurdish #language-Danish #language-German #language-Modern Greek (1453-) #language-Spanish #language-Persian #language-French #language-Hindi #language-Italian #language-Japanese #language-Korean #language-Kurdish #language-Marathi #language-Panjabi #language-Russian #language-Telugu #language-Thai #language-Tagalog #language-Turkish #language-Urdu #language-Vietnamese #language-Chinese #license-cc-by-4.0 #arxiv-2310.17586 #region-us \n",
"# Dataset Card for \"WEATHub\"\n\nThis dataset corresponds to the data described in the paper \"Global Voices, Local Biases: Socio-Cultural Prejudices across Languages\"\naccepted to EMNLP 2023.",
"## Table of Contents\n- Table of Contents\n- Dataset Description\n - Dataset Summary\n - Supported Tasks and Leaderboards\n - Languages\n- Dataset Structure\n - Data Instances\n - Data Fields\n - Data Splits\n- Dataset Creation\n - Curation Rationale\n - Source Data\n - Annotations\n - Personal and Sensitive Information\n- Considerations for Using the Data\n - Social Impact of Dataset\n - Discussion of Biases\n - Other Known Limitations\n- Additional Information\n - Dataset Curators\n - Licensing Information\n - Citation Information\n - Contributions",
"## Dataset Description\n\n- Homepage: Website\n- Repository: GitHub\n- Paper: URL\n- Point of Contact: Anjishnu Mukherjee",
"### Dataset Summary\n\nWEATHub is a dataset containing 24 languages. It contains words organized into groups of (target1, target2, attribute1, attribute2)\nto measure the association target1:target2 :: attribute1:attribute2. For example target1 can be insects, target2 can be flowers. And we \nmight be trying to measure whether we find insects or flowers pleasant or unpleasant. The measurement of word associations is quantified \nusing the WEAT metric in our paper. It is a metric that calculates an effect size (Cohen's d) and also provides a p-value (to measure\nstatistical significance of the results). In our paper, we use word embeddings from language models to perform these tests and understand \nbiased associations in language models across different languages.",
"### Supported Tasks and Leaderboards\n\n- 'bias_eval' : The dataset is used to measure biased associations.\n- This particular task isn't a standard task that is currently supported.",
"### Languages\n\nThe languages (in alphabetical order of language codes) are: Arabic (ar), Bengali (bn), Sorani Kurdish (ckb), Danish (da), German (de), \nGreek (el), Spanish (es), Persian (fa), French (fr), Hindi (hi), Italian (it), Japanese (ja), Korean (ko), Kurmanji Kurdish (ku), \nMarathi (mr), Punjabi (pa), Russian (ru), Telugu (te), Thai (th), Tagalog (tl), Turkish (tr), Urdu (ur), Vietnamese (vi), Chinese (zh).",
"## Dataset Structure",
"### Data Instances\n\nAn example instance is of the form:",
"### Data Fields\n\n- A single data point has the following features:\n - name: language (corresponding to the language codes given above)\n - name: weat (ID corresponding to a WEAT category)\n - name: attr1.category (a descriptive name for attribute 1)\n - name: attr1.examples (list of words for attribute 1)\n - name: attr2.category (a descriptive name for attribute 2)\n - name: attr2.examples (list of words for attribute 2)\n - name: targ1.category (a descriptive name for target 1)\n - name: targ1.examples (list of words for target 1)\n - name: targ2.category (a descriptive name for target 2)\n - name: targ2.examples (list of words for target 2)\n \n- All the features are stored as strings. The examples represent lists of strings.",
"### Data Splits\n\n- The dataset is divided into 3 splits as per the description in our paper:\n - original_weat - described in Table 1 of our paper, this corresponds to the original WEAT categories as given by Caliskan et al. in their\n seminal work from 2017 (Semantics derived automatically from language corpora contain human-like biases)\n - new_human_biases - described in Table 2 of our paper, this corresponds to contemporary dimensions of bias that are more human-centric in\n modern society.\n - india_specific_biases - These contain data corresponding to india specific bias dimensions as described in the paper (Socially Aware Bias Measurements for Hindi Language Representations)\n from NAACL '22 by Malik et al.",
"## Dataset Creation",
"### Curation Rationale\n\nThis dataset is intended to be used for measuring intrinsic biases in word embeddings obtained from language models.",
"### Source Data",
"#### Initial Data Collection and Normalization\n\nDescribed in details in section 2 of our paper. Briefly, for existing weat categories, we use human annotations to improve the quality of the\ntranslated WEAT word lists. For new weat categories, we research possible relevant dimensions thoroughly and come up with words after thorough\ndiscussions with our annotators.",
"#### Who are the source language producers?\n\nData for each of the language is from native speakers of that language. All annotators who participated in our study are native speakers of \ntheir respective languages and have at least college-level education background.",
"### Annotations",
"#### Annotation process\n\nDescribed in details in section 2 of our paper. Word level annotations.\nTo collect annotated data in various languages, we provide our annotators with the English words and their corresponding automatic translation\n, separated by WEAT category. We provide instructions to verify the accuracy of the translations and provide corrected versions for any \ninaccuracies. Additionally, we ask annotators to provide grammatically gendered forms of words, if applicable, or multiple translations \nof a word, if necessary.",
"#### Who are the annotators?\n\nAll annotators who participated in our study are native speakers of \ntheir respective languages and have at least college-level education background.",
"### Personal and Sensitive Information\n\nSince this dataset tries to measure biased associations at the word level, there may be some word level biases that are sensitive to certain \ngroups.",
"## Considerations for Using the Data",
"### Social Impact of Dataset\n\nThis dataset should be a starting point for measuring word level biased associations in a multilingual setting, which has not been explored \nin much depth in recent literature.",
"### Discussion of Biases\n\nThis dataset represents word level information used for measuring biases. Since these are annotated by humans, they may to certain extent reflect\nthe biases that they hold at an individual level.",
"### Other Known Limitations\n\n- For most of the languages in our dataset WEATHub, we had access to at least two annotators for cross-verifying the accuracy of\n the human translations to determine if the translated words fit into the context of that particular WEAT category.\n However, for some languages, we only have one annotator per language, so this might mean that for some languages the data may represent\n the biases of that individual annotator even though those biases are somewhat also reflected by Google Translate so it isn't completely\n an individualistic issue.\n- While we have tried to cover as many languages from the global South as possible, we acknowledge that 24 languages are indeed a\n tiny proportion of the 7000 languages in the world, some of which do not even have text representations.\n- WEAT can be an unreliable metric for contextualized embeddings from transformer models. We need better metrics to study intrinsic biases in\n transformer models. We believe the target and attribute pairs we provide as part of WEATHub in multiple languages is an important step\n towards a better multilingual metric for evaluating intrinsic biases in language models.",
"## Additional Information",
"### Dataset Curators\n\nThis dataset was curated by Anjishnu Mukherjee, Chahat Raj, Ziwei Zhu and Antonios Anastasopoulos for their EMNLP paper while the first two authors were \npursuing their PhD at George Mason University. This work\nwas generously supported by the National Science Foundation under award IIS-2327143. Computational resources for experiments were provided by the\nOffice of of Research Computing at George Mason University (URL: URL) and funded in part by grants from the \nNational Science Foundation (Awards Number 1625039 and 2018631).",
"### Licensing Information\n\nCurrently this dataset is released under CC-4.0 (might need to update this if required)\n\n\n\n@inproceedings{mukherjee-etal-2023-global,\n title = \"{G}lobal {V}oices, Local Biases: Socio-Cultural Prejudices across Languages\",\n author = \"Mukherjee, Anjishnu and\n Raj, Chahat and\n Zhu, Ziwei and\n Anastasopoulos, Antonios\",\n editor = \"Bouamor, Houda and\n Pino, Juan and\n Bali, Kalika\",\n booktitle = \"Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing\",\n month = dec,\n year = \"2023\",\n address = \"Singapore\",\n publisher = \"Association for Computational Linguistics\",\n url = \"URL\n doi = \"10.18653/v1/URL-main.981\",\n pages = \"15828--15845\",\n abstract = \"Human biases are ubiquitous but not uniform: disparities exist across linguistic, cultural, and societal borders. As large amounts of recent literature suggest, language models (LMs) trained on human data can reflect and often amplify the effects of these social biases. However, the vast majority of existing studies on bias are heavily skewed towards Western and European languages. In this work, we scale the Word Embedding Association Test (WEAT) to 24 languages, enabling broader studies and yielding interesting findings about LM bias. We additionally enhance this data with culturally relevant information for each language, capturing local contexts on a global scale. Further, to encompass more widely prevalent societal biases, we examine new bias dimensions across toxicity, ableism, and more. Moreover, we delve deeper into the Indian linguistic landscape, conducting a comprehensive regional bias analysis across six prevalent Indian languages. Finally, we highlight the significance of these social biases and the new dimensions through an extensive comparison of embedding methods, reinforcing the need to address them in pursuit of more equitable language models.\",\n}",
"### Contributions\n\nThanks to @iamshnoo for adding this dataset."
] |
[
157,
52,
125,
29,
182,
46,
128,
6,
14,
207,
171,
5,
37,
4,
82,
56,
5,
122,
39,
40,
8,
45,
52,
270,
5,
131,
489,
18
] |
[
"passage: TAGS\n#language-Arabic #language-Bengali #language-Central Kurdish #language-Danish #language-German #language-Modern Greek (1453-) #language-Spanish #language-Persian #language-French #language-Hindi #language-Italian #language-Japanese #language-Korean #language-Kurdish #language-Marathi #language-Panjabi #language-Russian #language-Telugu #language-Thai #language-Tagalog #language-Turkish #language-Urdu #language-Vietnamese #language-Chinese #license-cc-by-4.0 #arxiv-2310.17586 #region-us \n# Dataset Card for \"WEATHub\"\n\nThis dataset corresponds to the data described in the paper \"Global Voices, Local Biases: Socio-Cultural Prejudices across Languages\"\naccepted to EMNLP 2023.## Table of Contents\n- Table of Contents\n- Dataset Description\n - Dataset Summary\n - Supported Tasks and Leaderboards\n - Languages\n- Dataset Structure\n - Data Instances\n - Data Fields\n - Data Splits\n- Dataset Creation\n - Curation Rationale\n - Source Data\n - Annotations\n - Personal and Sensitive Information\n- Considerations for Using the Data\n - Social Impact of Dataset\n - Discussion of Biases\n - Other Known Limitations\n- Additional Information\n - Dataset Curators\n - Licensing Information\n - Citation Information\n - Contributions## Dataset Description\n\n- Homepage: Website\n- Repository: GitHub\n- Paper: URL\n- Point of Contact: Anjishnu Mukherjee",
"passage: ### Dataset Summary\n\nWEATHub is a dataset containing 24 languages. It contains words organized into groups of (target1, target2, attribute1, attribute2)\nto measure the association target1:target2 :: attribute1:attribute2. For example target1 can be insects, target2 can be flowers. And we \nmight be trying to measure whether we find insects or flowers pleasant or unpleasant. The measurement of word associations is quantified \nusing the WEAT metric in our paper. It is a metric that calculates an effect size (Cohen's d) and also provides a p-value (to measure\nstatistical significance of the results). In our paper, we use word embeddings from language models to perform these tests and understand \nbiased associations in language models across different languages.### Supported Tasks and Leaderboards\n\n- 'bias_eval' : The dataset is used to measure biased associations.\n- This particular task isn't a standard task that is currently supported.### Languages\n\nThe languages (in alphabetical order of language codes) are: Arabic (ar), Bengali (bn), Sorani Kurdish (ckb), Danish (da), German (de), \nGreek (el), Spanish (es), Persian (fa), French (fr), Hindi (hi), Italian (it), Japanese (ja), Korean (ko), Kurmanji Kurdish (ku), \nMarathi (mr), Punjabi (pa), Russian (ru), Telugu (te), Thai (th), Tagalog (tl), Turkish (tr), Urdu (ur), Vietnamese (vi), Chinese (zh).## Dataset Structure### Data Instances\n\nAn example instance is of the form:### Data Fields\n\n- A single data point has the following features:\n - name: language (corresponding to the language codes given above)\n - name: weat (ID corresponding to a WEAT category)\n - name: attr1.category (a descriptive name for attribute 1)\n - name: attr1.examples (list of words for attribute 1)\n - name: attr2.category (a descriptive name for attribute 2)\n - name: attr2.examples (list of words for attribute 2)\n - name: targ1.category (a descriptive name for target 1)\n - name: targ1.examples (list of words for target 1)\n - name: targ2.category (a descriptive name for target 2)\n - name: targ2.examples (list of words for target 2)\n \n- All the features are stored as strings. The examples represent lists of strings.",
"passage: ### Data Splits\n\n- The dataset is divided into 3 splits as per the description in our paper:\n - original_weat - described in Table 1 of our paper, this corresponds to the original WEAT categories as given by Caliskan et al. in their\n seminal work from 2017 (Semantics derived automatically from language corpora contain human-like biases)\n - new_human_biases - described in Table 2 of our paper, this corresponds to contemporary dimensions of bias that are more human-centric in\n modern society.\n - india_specific_biases - These contain data corresponding to india specific bias dimensions as described in the paper (Socially Aware Bias Measurements for Hindi Language Representations)\n from NAACL '22 by Malik et al.## Dataset Creation### Curation Rationale\n\nThis dataset is intended to be used for measuring intrinsic biases in word embeddings obtained from language models.### Source Data#### Initial Data Collection and Normalization\n\nDescribed in details in section 2 of our paper. Briefly, for existing weat categories, we use human annotations to improve the quality of the\ntranslated WEAT word lists. For new weat categories, we research possible relevant dimensions thoroughly and come up with words after thorough\ndiscussions with our annotators.#### Who are the source language producers?\n\nData for each of the language is from native speakers of that language. All annotators who participated in our study are native speakers of \ntheir respective languages and have at least college-level education background.### Annotations#### Annotation process\n\nDescribed in details in section 2 of our paper. Word level annotations.\nTo collect annotated data in various languages, we provide our annotators with the English words and their corresponding automatic translation\n, separated by WEAT category. We provide instructions to verify the accuracy of the translations and provide corrected versions for any \ninaccuracies. Additionally, we ask annotators to provide grammatically gendered forms of words, if applicable, or multiple translations \nof a word, if necessary.#### Who are the annotators?\n\nAll annotators who participated in our study are native speakers of \ntheir respective languages and have at least college-level education background.### Personal and Sensitive Information\n\nSince this dataset tries to measure biased associations at the word level, there may be some word level biases that are sensitive to certain \ngroups.## Considerations for Using the Data### Social Impact of Dataset\n\nThis dataset should be a starting point for measuring word level biased associations in a multilingual setting, which has not been explored \nin much depth in recent literature.### Discussion of Biases\n\nThis dataset represents word level information used for measuring biases. Since these are annotated by humans, they may to certain extent reflect\nthe biases that they hold at an individual level.",
"passage: ### Other Known Limitations\n\n- For most of the languages in our dataset WEATHub, we had access to at least two annotators for cross-verifying the accuracy of\n the human translations to determine if the translated words fit into the context of that particular WEAT category.\n However, for some languages, we only have one annotator per language, so this might mean that for some languages the data may represent\n the biases of that individual annotator even though those biases are somewhat also reflected by Google Translate so it isn't completely\n an individualistic issue.\n- While we have tried to cover as many languages from the global South as possible, we acknowledge that 24 languages are indeed a\n tiny proportion of the 7000 languages in the world, some of which do not even have text representations.\n- WEAT can be an unreliable metric for contextualized embeddings from transformer models. We need better metrics to study intrinsic biases in\n transformer models. We believe the target and attribute pairs we provide as part of WEATHub in multiple languages is an important step\n towards a better multilingual metric for evaluating intrinsic biases in language models.## Additional Information### Dataset Curators\n\nThis dataset was curated by Anjishnu Mukherjee, Chahat Raj, Ziwei Zhu and Antonios Anastasopoulos for their EMNLP paper while the first two authors were \npursuing their PhD at George Mason University. This work\nwas generously supported by the National Science Foundation under award IIS-2327143. Computational resources for experiments were provided by the\nOffice of of Research Computing at George Mason University (URL: URL) and funded in part by grants from the \nNational Science Foundation (Awards Number 1625039 and 2018631)."
] |
5cd1a4d2fb867c81884add19e59b1b475a7c3dbf
|
# Discussions in Folketinget
The dataset is based on data from Folketinget in the Danish Gigaword corpus.
Comment-response pairs are purely extracted on the basis of heuristics, and have not been manually evaluated.
The dataset was created for aiding the training of sentence transformer models in the Danish Foundation Models project.
The dataset is currently not recommended for production use.
|
kardosdrur/folketinget-discussions
|
[
"license:mit",
"region:us"
] |
2023-10-24T07:48:35+00:00
|
{"license": "mit", "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "test", "path": "data/test-*"}]}], "dataset_info": {"features": [{"name": "comment", "dtype": "string"}, {"name": "response", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 7032676.035654362, "num_examples": 3814}, {"name": "test", "num_bytes": 1759090.9643456375, "num_examples": 954}], "download_size": 4898174, "dataset_size": 8791767.0}}
|
2023-12-04T09:25:10+00:00
|
[] |
[] |
TAGS
#license-mit #region-us
|
# Discussions in Folketinget
The dataset is based on data from Folketinget in the Danish Gigaword corpus.
Comment-response pairs are purely extracted on the basis of heuristics, and have not been manually evaluated.
The dataset was created for aiding the training of sentence transformer models in the Danish Foundation Models project.
The dataset is currently not recommended for production use.
|
[
"# Discussions in Folketinget\n\nThe dataset is based on data from Folketinget in the Danish Gigaword corpus.\nComment-response pairs are purely extracted on the basis of heuristics, and have not been manually evaluated.\n\nThe dataset was created for aiding the training of sentence transformer models in the Danish Foundation Models project.\nThe dataset is currently not recommended for production use."
] |
[
"TAGS\n#license-mit #region-us \n",
"# Discussions in Folketinget\n\nThe dataset is based on data from Folketinget in the Danish Gigaword corpus.\nComment-response pairs are purely extracted on the basis of heuristics, and have not been manually evaluated.\n\nThe dataset was created for aiding the training of sentence transformer models in the Danish Foundation Models project.\nThe dataset is currently not recommended for production use."
] |
[
11,
84
] |
[
"passage: TAGS\n#license-mit #region-us \n# Discussions in Folketinget\n\nThe dataset is based on data from Folketinget in the Danish Gigaword corpus.\nComment-response pairs are purely extracted on the basis of heuristics, and have not been manually evaluated.\n\nThe dataset was created for aiding the training of sentence transformer models in the Danish Foundation Models project.\nThe dataset is currently not recommended for production use."
] |
6d5460a5c9095a6de5994012096b65352d9f8089
|
# Dataset Card for "pix2struct-chartcaptioning"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
hk-kaden-kim/pix2struct-chartcaptioning
|
[
"region:us"
] |
2023-10-24T07:51:25+00:00
|
{"dataset_info": {"features": [{"name": "image", "dtype": "image"}, {"name": "caption", "dtype": "string"}, {"name": "chartType", "dtype": "string"}, {"name": "chartElement", "dtype": "string"}, {"name": "dataType", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 72690622.389, "num_examples": 1373}, {"name": "test", "num_bytes": 57043770.8, "num_examples": 1200}, {"name": "validation", "num_bytes": 10174935.0, "num_examples": 199}], "download_size": 0, "dataset_size": 139909328.189}}
|
2023-10-24T09:14:03+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "pix2struct-chartcaptioning"
More Information needed
|
[
"# Dataset Card for \"pix2struct-chartcaptioning\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"pix2struct-chartcaptioning\"\n\nMore Information needed"
] |
[
6,
19
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"pix2struct-chartcaptioning\"\n\nMore Information needed"
] |
daee4cffe8bea93b1cd7874b01c5a6a4ae9dcaf2
|
# Dataset Card for UltraFeedback Binarized
## Dataset Description
This is a pre-processed version of the [UltraFeedback dataset](https://huggingface.co/datasets/openbmb/UltraFeedback) and was used to train [Zephyr-7Β-β](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta), a state of the art chat model at the 7B parameter scale.
The original UltraFeedback dataset consists of 64k prompts, where each prompt is accompanied with four model completions from a wide variety of open and proprietary models. GPT-4 is then used to assign a score to each completion, along criteria like helpfulness and honesty. To create `UltraFeedback Binarized`, we picked the highest `overall_score` as the "chosen" completion, and one of the remaining 3 at random as the "rejected" one. This defines the preference modelling splits for techniques like reward modelling or DPO. We also created splits for supervised fine-tuning (SFT) that use the "chosen" column as the dialogues to model, along with splits that involve generation like rejection sampling or PPO. For details on the dataset processing, see the accompanying [script](https://huggingface.co/datasets/HuggingFaceH4/ultrafeedback_binarized/blob/main/create_dataset.py).
## Dataset Structure
### Usage
To load the dataset, run:
```python
from datasets import load_dataset
ds = load_dataset("HuggingFaceH4/ultrafeedback_binarized")
```
**Note:** after the release of Zephyr-7b-β, the team at [Argilla](https://huggingface.co/argilla) noted that there were a few hundred completions with the incorrect label. Similarly, members of the community pointed out that a several prompts were sourced from the TruthfulQA benchmark which can lead to contamination with public leaderboards. Both of these issues have been resolved in the latest version of the dataset, but if you wish to use the older version, you can run:
```python
from datasets import load_dataset
ds = load_dataset("HuggingFaceH4/ultrafeedback_binarized", revision="292c16329d921287c4166934cac1a6ad1e13a6c5")
```
### Data Splits
The dataset has six splits, suitable for:
* Supervised fine-tuning (`sft`).
* Preference modelling (`prefs`) to train reward models or apply techniques like DPO.
* Generation ranking (`gen`) via techniques like rejection sampling or PPO.
The number of examples per split is shown as follows:
| train_sft | test_sft | train_prefs | test_prefs | train_gen | test_gen |
|:-------:|:-----------:|:-----:| :-----:| :-----:| :-----:|
| 61135 | 1000 | 61135 | 2000 | 61135 | 1000 |
The dataset is stored in parquet format with each entry using the following schema:
```
{
"prompt": "Which books are still banned in Canada?",
"chosen": [
{
"content": "Which books are still banned in Canada?",
"role": "user"
},
{
"content": "As a responsible and ethical AI assistant, I must inform you that I cannot provide a list of banned books in Canada, as there are no books that are currently banned in the country. Canada has a strong commitment to freedom of expression and the right to access information, and there are no laws in place that restrict the publication or distribution of books.\n",
"role": "assistant"
}
],
"messages": [
{
"content": "Which books are still banned in Canada?",
"role": "user"
},
{
"content": "As a responsible and ethical AI assistant, I must inform you that I cannot provide a list of banned books in Canada, as there are no books that are currently banned in the country. Canada has a strong commitment to freedom of expression and the right to access information, and there are no laws in place that restrict the publication or distribution of books.\n",
"role": "assistant"
}],
"prompt_id": "aeccf551d9ba42fdf5f2044de43b8ce6e360fb523ace428317b81d804594e090",
"rejected": [
{
"content": "Which books are still banned in Canada?",
"role": "user"},
{
"content": "According to the Canadian Government’s Ban Affront website, there are still several books that are banned in Canada. These include The Begum’s Millionaire, The Education of Little Tree, The Harry Potter series, Lolita, 1984, and Lady Chatterley’s Lover. Some of these books are considered inaccessible due to their age, while others are still legally banned in certain parts of the country.",
"role": "assistant"
}
],
"score_chosen": 8.0,
"score_rejected": 5.0
}
```
You should use the `chosen` and `rejected` columns for techniques like DPO, while the `messages` column is suitable for SFT or PPO.
## Citation
If you find this dataset is useful in your work, please cite the original UltraFeedback dataset: https://huggingface.co/datasets/openbmb/UltraFeedback
You may also wish to cite the Zephyr 7B technical report:
```
@misc{tunstall2023zephyr,
title={Zephyr: Direct Distillation of LM Alignment},
author={Lewis Tunstall and Edward Beeching and Nathan Lambert and Nazneen Rajani and Kashif Rasul and Younes Belkada and Shengyi Huang and Leandro von Werra and Clémentine Fourrier and Nathan Habib and Nathan Sarrazin and Omar Sanseviero and Alexander M. Rush and Thomas Wolf},
year={2023},
eprint={2310.16944},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
|
HuggingFaceH4/ultrafeedback_binarized
|
[
"task_categories:conversational",
"task_categories:text-generation",
"language:en",
"license:mit",
"arxiv:2310.16944",
"region:us"
] |
2023-10-24T07:53:19+00:00
|
{"language": ["en"], "license": "mit", "task_categories": ["conversational", "text-generation"], "pretty_name": "UltraFeedback Binarized", "configs": [{"config_name": "default", "data_files": [{"split": "train_prefs", "path": "data/train_prefs-*"}, {"split": "train_sft", "path": "data/train_sft-*"}, {"split": "test_prefs", "path": "data/test_prefs-*"}, {"split": "test_sft", "path": "data/test_sft-*"}, {"split": "train_gen", "path": "data/train_gen-*"}, {"split": "test_gen", "path": "data/test_gen-*"}]}], "dataset_info": {"features": [{"name": "prompt", "dtype": "string"}, {"name": "prompt_id", "dtype": "string"}, {"name": "chosen", "list": [{"name": "content", "dtype": "string"}, {"name": "role", "dtype": "string"}]}, {"name": "rejected", "list": [{"name": "content", "dtype": "string"}, {"name": "role", "dtype": "string"}]}, {"name": "messages", "list": [{"name": "content", "dtype": "string"}, {"name": "role", "dtype": "string"}]}, {"name": "score_chosen", "dtype": "float64"}, {"name": "score_rejected", "dtype": "float64"}], "splits": [{"name": "train_prefs", "num_bytes": 405688662, "num_examples": 61135}, {"name": "train_sft", "num_bytes": 405688662, "num_examples": 61135}, {"name": "test_prefs", "num_bytes": 13161585, "num_examples": 2000}, {"name": "test_sft", "num_bytes": 6697333, "num_examples": 1000}, {"name": "train_gen", "num_bytes": 325040536, "num_examples": 61135}, {"name": "test_gen", "num_bytes": 5337695, "num_examples": 1000}], "download_size": 649967196, "dataset_size": 1161614473}}
|
2024-01-08T11:33:19+00:00
|
[
"2310.16944"
] |
[
"en"
] |
TAGS
#task_categories-conversational #task_categories-text-generation #language-English #license-mit #arxiv-2310.16944 #region-us
|
Dataset Card for UltraFeedback Binarized
========================================
Dataset Description
-------------------
This is a pre-processed version of the UltraFeedback dataset and was used to train Zephyr-7Β-β, a state of the art chat model at the 7B parameter scale.
The original UltraFeedback dataset consists of 64k prompts, where each prompt is accompanied with four model completions from a wide variety of open and proprietary models. GPT-4 is then used to assign a score to each completion, along criteria like helpfulness and honesty. To create 'UltraFeedback Binarized', we picked the highest 'overall\_score' as the "chosen" completion, and one of the remaining 3 at random as the "rejected" one. This defines the preference modelling splits for techniques like reward modelling or DPO. We also created splits for supervised fine-tuning (SFT) that use the "chosen" column as the dialogues to model, along with splits that involve generation like rejection sampling or PPO. For details on the dataset processing, see the accompanying script.
Dataset Structure
-----------------
### Usage
To load the dataset, run:
Note: after the release of Zephyr-7b-β, the team at Argilla noted that there were a few hundred completions with the incorrect label. Similarly, members of the community pointed out that a several prompts were sourced from the TruthfulQA benchmark which can lead to contamination with public leaderboards. Both of these issues have been resolved in the latest version of the dataset, but if you wish to use the older version, you can run:
### Data Splits
The dataset has six splits, suitable for:
* Supervised fine-tuning ('sft').
* Preference modelling ('prefs') to train reward models or apply techniques like DPO.
* Generation ranking ('gen') via techniques like rejection sampling or PPO.
The number of examples per split is shown as follows:
The dataset is stored in parquet format with each entry using the following schema:
You should use the 'chosen' and 'rejected' columns for techniques like DPO, while the 'messages' column is suitable for SFT or PPO.
If you find this dataset is useful in your work, please cite the original UltraFeedback dataset: URL
You may also wish to cite the Zephyr 7B technical report:
|
[
"### Usage\n\n\nTo load the dataset, run:\n\n\nNote: after the release of Zephyr-7b-β, the team at Argilla noted that there were a few hundred completions with the incorrect label. Similarly, members of the community pointed out that a several prompts were sourced from the TruthfulQA benchmark which can lead to contamination with public leaderboards. Both of these issues have been resolved in the latest version of the dataset, but if you wish to use the older version, you can run:",
"### Data Splits\n\n\nThe dataset has six splits, suitable for:\n\n\n* Supervised fine-tuning ('sft').\n* Preference modelling ('prefs') to train reward models or apply techniques like DPO.\n* Generation ranking ('gen') via techniques like rejection sampling or PPO.\n\n\nThe number of examples per split is shown as follows:\n\n\n\nThe dataset is stored in parquet format with each entry using the following schema:\n\n\nYou should use the 'chosen' and 'rejected' columns for techniques like DPO, while the 'messages' column is suitable for SFT or PPO.\n\n\nIf you find this dataset is useful in your work, please cite the original UltraFeedback dataset: URL\n\n\nYou may also wish to cite the Zephyr 7B technical report:"
] |
[
"TAGS\n#task_categories-conversational #task_categories-text-generation #language-English #license-mit #arxiv-2310.16944 #region-us \n",
"### Usage\n\n\nTo load the dataset, run:\n\n\nNote: after the release of Zephyr-7b-β, the team at Argilla noted that there were a few hundred completions with the incorrect label. Similarly, members of the community pointed out that a several prompts were sourced from the TruthfulQA benchmark which can lead to contamination with public leaderboards. Both of these issues have been resolved in the latest version of the dataset, but if you wish to use the older version, you can run:",
"### Data Splits\n\n\nThe dataset has six splits, suitable for:\n\n\n* Supervised fine-tuning ('sft').\n* Preference modelling ('prefs') to train reward models or apply techniques like DPO.\n* Generation ranking ('gen') via techniques like rejection sampling or PPO.\n\n\nThe number of examples per split is shown as follows:\n\n\n\nThe dataset is stored in parquet format with each entry using the following schema:\n\n\nYou should use the 'chosen' and 'rejected' columns for techniques like DPO, while the 'messages' column is suitable for SFT or PPO.\n\n\nIf you find this dataset is useful in your work, please cite the original UltraFeedback dataset: URL\n\n\nYou may also wish to cite the Zephyr 7B technical report:"
] |
[
45,
113,
183
] |
[
"passage: TAGS\n#task_categories-conversational #task_categories-text-generation #language-English #license-mit #arxiv-2310.16944 #region-us \n### Usage\n\n\nTo load the dataset, run:\n\n\nNote: after the release of Zephyr-7b-β, the team at Argilla noted that there were a few hundred completions with the incorrect label. Similarly, members of the community pointed out that a several prompts were sourced from the TruthfulQA benchmark which can lead to contamination with public leaderboards. Both of these issues have been resolved in the latest version of the dataset, but if you wish to use the older version, you can run:### Data Splits\n\n\nThe dataset has six splits, suitable for:\n\n\n* Supervised fine-tuning ('sft').\n* Preference modelling ('prefs') to train reward models or apply techniques like DPO.\n* Generation ranking ('gen') via techniques like rejection sampling or PPO.\n\n\nThe number of examples per split is shown as follows:\n\n\n\nThe dataset is stored in parquet format with each entry using the following schema:\n\n\nYou should use the 'chosen' and 'rejected' columns for techniques like DPO, while the 'messages' column is suitable for SFT or PPO.\n\n\nIf you find this dataset is useful in your work, please cite the original UltraFeedback dataset: URL\n\n\nYou may also wish to cite the Zephyr 7B technical report:"
] |
6b816e33243d3479009398f48b9e325fea4b0f65
|
# Dataset Card for "most_precious_4"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
ArmelRandy/most_precious_4
|
[
"region:us"
] |
2023-10-24T08:01:58+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "test", "path": "data/test-*"}]}], "dataset_info": {"features": [{"name": "prompt", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 8031615.129808251, "num_examples": 7744}, {"name": "test", "num_bytes": 892977.870191749, "num_examples": 861}], "download_size": 5178596, "dataset_size": 8924593.0}}
|
2023-10-24T08:02:14+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "most_precious_4"
More Information needed
|
[
"# Dataset Card for \"most_precious_4\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"most_precious_4\"\n\nMore Information needed"
] |
[
6,
17
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"most_precious_4\"\n\nMore Information needed"
] |
8296db0b791758743cb8efb5d62cc10d03c2b988
|
# Dataset Card for "uspto-50k"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
Phando/uspto-50k
|
[
"region:us"
] |
2023-10-24T08:03:20+00:00
|
{"dataset_info": {"features": [{"name": "class", "dtype": "int64"}, {"name": "id", "dtype": "string"}, {"name": "prod_smiles", "dtype": "string"}, {"name": "rxn_smiles", "dtype": "string"}, {"name": "prod_smiles_pop", "dtype": "int64"}, {"name": "keep", "dtype": "bool"}], "splits": [{"name": "train", "num_bytes": 22822250.69997601, "num_examples": 49015}, {"name": "validation", "num_bytes": 466083.3000239923, "num_examples": 1001}], "download_size": 8864323, "dataset_size": 23288334.0}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "validation", "path": "data/validation-*"}]}]}
|
2023-12-18T04:24:59+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "uspto-50k"
More Information needed
|
[
"# Dataset Card for \"uspto-50k\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"uspto-50k\"\n\nMore Information needed"
] |
[
6,
14
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"uspto-50k\"\n\nMore Information needed"
] |
2f3e250e978162d2ee1b083e0054c6ef524f2cbd
|
# Wish You Were Here - Dataset
A dataset conisting out of postcards from 1900-1960 annoted with a combination of CLIP and manual annotation.
## datastructure
The dataset is strucured as follows:
- Images of postcards.
- Text file desribing the image.
- Images are liked to text file via the name of the text file.
## Metadata
| Size of dataset in quantity | Size of dataset in storage | Repeating images | Source annotation |
|---|---|---|---|
| 646 | 1.6 Gb | Yes | No |
### Collection method:
Manual search of WikiMedia pages and selection of images with attributes that allow for the usage of images without permission or attribution of the creator of the media. Licenses include:
- CC-O
- No license
- Public domain
### Annotation method:
The data was annotated using Kohya_SS in 2 phases:
1. Automated annotation using Clip.
2. Manual annotation.
During manual annotation the following features were consitently annotated:
- Type of Postcard (drawing, photograph, colored in photograph)
- Aspect ration (horizontal, vertical or square)
- Border color (if there is a border)
- Damage of the postcard (ranging from no annotation, slightly damage, damage, significant damage)
- Stamps
- Folding damage
- Lineart
- Monochrome (color images are not specified)
### Image dataset composition:
The dataset compromises of postcards originating from Germany, Poland, Russia and the United States of America. No additional annotation provded to identify where the postcards are from.
Most of the postcards depict a bias towards nature scenes E.G snowy mountain valleys at sunset. Training a model could create a bias towards these images.
## license
This dataset is licensed under CC BY 4.0 Deed. This gives you the rights to:
- Share — copy and redistribute the material in any medium or format for any purpose, even commercially.
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Under the following terms:
- Attribution - You must give appropriate credit , provide a link to the license, and indicate if changes were made . You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions - You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
The license only applies to the descriptions of the images. Not to the images themselves (see collection method for more details).
|
calm-and-collected/wish-you-were-here
|
[
"size_categories:n<1K",
"language:en",
"license:cc-by-4.0",
"photography",
"art",
"region:us"
] |
2023-10-24T08:48:54+00:00
|
{"language": ["en"], "license": "cc-by-4.0", "size_categories": ["n<1K"], "pretty_name": "Wish You were Here", "tags": ["photography", "art"]}
|
2023-10-25T19:59:52+00:00
|
[] |
[
"en"
] |
TAGS
#size_categories-n<1K #language-English #license-cc-by-4.0 #photography #art #region-us
|
Wish You Were Here - Dataset
============================
!image/png
A dataset conisting out of postcards from 1900-1960 annoted with a combination of CLIP and manual annotation.
datastructure
-------------
The dataset is strucured as follows:
* Images of postcards.
* Text file desribing the image.
* Images are liked to text file via the name of the text file.
!image/png
Metadata
--------
### Collection method:
Manual search of WikiMedia pages and selection of images with attributes that allow for the usage of images without permission or attribution of the creator of the media. Licenses include:
* CC-O
* No license
* Public domain
### Annotation method:
The data was annotated using Kohya\_SS in 2 phases:
1. Automated annotation using Clip.
2. Manual annotation.
During manual annotation the following features were consitently annotated:
* Type of Postcard (drawing, photograph, colored in photograph)
* Aspect ration (horizontal, vertical or square)
* Border color (if there is a border)
* Damage of the postcard (ranging from no annotation, slightly damage, damage, significant damage)
* Stamps
* Folding damage
* Lineart
* Monochrome (color images are not specified)
### Image dataset composition:
The dataset compromises of postcards originating from Germany, Poland, Russia and the United States of America. No additional annotation provded to identify where the postcards are from.
Most of the postcards depict a bias towards nature scenes E.G snowy mountain valleys at sunset. Training a model could create a bias towards these images.
license
-------
This dataset is licensed under CC BY 4.0 Deed. This gives you the rights to:
* Share — copy and redistribute the material in any medium or format for any purpose, even commercially.
* Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Under the following terms:
* Attribution - You must give appropriate credit , provide a link to the license, and indicate if changes were made . You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
* No additional restrictions - You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
The license only applies to the descriptions of the images. Not to the images themselves (see collection method for more details).
|
[
"### Collection method:\n\n\nManual search of WikiMedia pages and selection of images with attributes that allow for the usage of images without permission or attribution of the creator of the media. Licenses include:\n\n\n* CC-O\n* No license\n* Public domain",
"### Annotation method:\n\n\nThe data was annotated using Kohya\\_SS in 2 phases:\n\n\n1. Automated annotation using Clip.\n2. Manual annotation.\n\n\nDuring manual annotation the following features were consitently annotated:\n\n\n* Type of Postcard (drawing, photograph, colored in photograph)\n* Aspect ration (horizontal, vertical or square)\n* Border color (if there is a border)\n* Damage of the postcard (ranging from no annotation, slightly damage, damage, significant damage)\n* Stamps\n* Folding damage\n* Lineart\n* Monochrome (color images are not specified)",
"### Image dataset composition:\n\n\nThe dataset compromises of postcards originating from Germany, Poland, Russia and the United States of America. No additional annotation provded to identify where the postcards are from.\nMost of the postcards depict a bias towards nature scenes E.G snowy mountain valleys at sunset. Training a model could create a bias towards these images.\n\n\nlicense\n-------\n\n\nThis dataset is licensed under CC BY 4.0 Deed. This gives you the rights to:\n\n\n* Share — copy and redistribute the material in any medium or format for any purpose, even commercially.\n* Adapt — remix, transform, and build upon the material for any purpose, even commercially.\nUnder the following terms:\n* Attribution - You must give appropriate credit , provide a link to the license, and indicate if changes were made . You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.\n* No additional restrictions - You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.\n\n\nThe license only applies to the descriptions of the images. Not to the images themselves (see collection method for more details)."
] |
[
"TAGS\n#size_categories-n<1K #language-English #license-cc-by-4.0 #photography #art #region-us \n",
"### Collection method:\n\n\nManual search of WikiMedia pages and selection of images with attributes that allow for the usage of images without permission or attribution of the creator of the media. Licenses include:\n\n\n* CC-O\n* No license\n* Public domain",
"### Annotation method:\n\n\nThe data was annotated using Kohya\\_SS in 2 phases:\n\n\n1. Automated annotation using Clip.\n2. Manual annotation.\n\n\nDuring manual annotation the following features were consitently annotated:\n\n\n* Type of Postcard (drawing, photograph, colored in photograph)\n* Aspect ration (horizontal, vertical or square)\n* Border color (if there is a border)\n* Damage of the postcard (ranging from no annotation, slightly damage, damage, significant damage)\n* Stamps\n* Folding damage\n* Lineart\n* Monochrome (color images are not specified)",
"### Image dataset composition:\n\n\nThe dataset compromises of postcards originating from Germany, Poland, Russia and the United States of America. No additional annotation provded to identify where the postcards are from.\nMost of the postcards depict a bias towards nature scenes E.G snowy mountain valleys at sunset. Training a model could create a bias towards these images.\n\n\nlicense\n-------\n\n\nThis dataset is licensed under CC BY 4.0 Deed. This gives you the rights to:\n\n\n* Share — copy and redistribute the material in any medium or format for any purpose, even commercially.\n* Adapt — remix, transform, and build upon the material for any purpose, even commercially.\nUnder the following terms:\n* Attribution - You must give appropriate credit , provide a link to the license, and indicate if changes were made . You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.\n* No additional restrictions - You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.\n\n\nThe license only applies to the descriptions of the images. Not to the images themselves (see collection method for more details)."
] |
[
33,
51,
138,
267
] |
[
"passage: TAGS\n#size_categories-n<1K #language-English #license-cc-by-4.0 #photography #art #region-us \n### Collection method:\n\n\nManual search of WikiMedia pages and selection of images with attributes that allow for the usage of images without permission or attribution of the creator of the media. Licenses include:\n\n\n* CC-O\n* No license\n* Public domain### Annotation method:\n\n\nThe data was annotated using Kohya\\_SS in 2 phases:\n\n\n1. Automated annotation using Clip.\n2. Manual annotation.\n\n\nDuring manual annotation the following features were consitently annotated:\n\n\n* Type of Postcard (drawing, photograph, colored in photograph)\n* Aspect ration (horizontal, vertical or square)\n* Border color (if there is a border)\n* Damage of the postcard (ranging from no annotation, slightly damage, damage, significant damage)\n* Stamps\n* Folding damage\n* Lineart\n* Monochrome (color images are not specified)### Image dataset composition:\n\n\nThe dataset compromises of postcards originating from Germany, Poland, Russia and the United States of America. No additional annotation provded to identify where the postcards are from.\nMost of the postcards depict a bias towards nature scenes E.G snowy mountain valleys at sunset. Training a model could create a bias towards these images.\n\n\nlicense\n-------\n\n\nThis dataset is licensed under CC BY 4.0 Deed. This gives you the rights to:\n\n\n* Share — copy and redistribute the material in any medium or format for any purpose, even commercially.\n* Adapt — remix, transform, and build upon the material for any purpose, even commercially.\nUnder the following terms:\n* Attribution - You must give appropriate credit , provide a link to the license, and indicate if changes were made . You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.\n* No additional restrictions - You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.\n\n\nThe license only applies to the descriptions of the images. Not to the images themselves (see collection method for more details)."
] |
49efc2204e6bc4530188deb42360d2c89bc22215
|
# AutoTrain Dataset for project: summarization-xlsum
## Dataset Description
This dataset has been automatically processed by AutoTrain for project summarization-xlsum.
### Languages
The BCP-47 code for the dataset's language is unk.
## Dataset Structure
### Data Instances
A sample from this dataset looks as follows:
```json
[
{
"text": "\u092e\u0928 \u0915\u0940 \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u092e\u0947\u0902 \u092e\u094c\u091c\u0942\u0926 \u0905\u0902\u0927\u0947\u0930\u093e \u092f\u093e \u0924\u094b \u0939\u092e\u0947\u0902 \u0916\u0941\u0926 \u0930\u094c\u0936\u0928\u0940 \u0915\u093e \u0938\u094d\u0930\u094b\u0924 \u092c\u0928\u0928\u093e \u0938\u0940\u0916\u093e \u0938\u0915\u0924\u093e \u0939\u0948 \u092f\u093e \u092b\u093f\u0930 \u0935\u0939 \u0939\u092e\u093e\u0930\u0940 \u092c\u091a\u094d\u091a\u0940 \u0915\u0940 \u091a\u092e\u0915 \u0915\u094b \u092d\u0940 \u0916\u0924\u094d\u092e \u0915\u0930 \u0938\u0915\u0924\u093e \u0939\u0948\u0964 \u0915\u0941\u091b \u0910\u0938\u093e \u0939\u0940 \u0939\u0941\u0906 \u0925\u093e \u091c\u0930\u094d\u092e\u0928\u0940 \u092b\u093f\u0932\u0949\u0938\u092b\u0930 \u092b\u094d\u0930\u0947\u0921\u094d\u0930\u093f\u0915 \u0928\u0940\u091a\u0947 \u0915\u0947 \u0938\u093e\u0925\u0964 \u0907\u0902\u0938\u093e\u0928 \u092a\u0943\u0925\u094d\u0935\u0940 \u0915\u0947 \u0939\u0930 \u091b\u094b\u0930 \u0924\u0915 \u092f\u093e\u0924\u094d\u0930\u093e \u0915\u0930 \u091a\u0941\u0915\u093e \u0939\u0948, \u0938\u092e\u0941\u0926\u094d\u0930 \u0915\u0940 \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u092e\u0947\u0902 \u0921\u0941\u092c\u0915\u0940 \u0932\u0917\u093e \u091a\u0941\u0915\u093e \u0939\u0948 \u0914\u0930 \u0916\u0941\u0926 \u0915\u094b \u0905\u0902\u0924\u0930\u093f\u0915\u094d\u0937 \u0915\u093e \u0938\u092b\u0930 \u0915\u0930\u093e \u091a\u0941\u0915\u093e \u0939\u0948\u0964 \u0932\u0947\u0915\u093f\u0928 \u0939\u092e\u093e\u0930\u093e \u092e\u0928 \u0905\u092d\u0940 \u092d\u0940 \u090f\u0915 \u0910\u0938\u0940 \u091a\u0940\u095b \u0939\u0948 \u091c\u0939\u093e\u0901 \u091c\u093e\u0928\u0947 \u0914\u0930 \u091c\u0948\u0938\u0947 \u090f\u0915\u094d\u0938\u094d\u092a\u094d\u0932\u094b\u0930 \u0915\u0930\u0928\u0947 \u0915\u0940 \u0939\u093f\u092e\u094d\u092e\u0924 \u0938\u093f\u0930\u094d\u092b \u0915\u0908 \u0917\u093f\u0928\u0947 \u091a\u0941\u0928\u0947 \u0932\u094b\u0917\u094b\u0902 \u092e\u0947\u0902 \u0939\u0940 \u0939\u0948\u0964 \u092a\u093e\u0917\u0932 \u0939\u094b \u091c\u093e\u0928\u0947 \u0914\u0930 \u092c\u0939\u0941\u0924 \u0907\u0928 10 \u0938\u094d\u092a\u0947\u0928 \u0915\u0947 \u0921\u0930 \u0938\u0947 \u091c\u094d\u092f\u093e\u0926\u093e\u0924\u0930 \u0939\u092e \u0932\u094b\u0917 \u0905\u092a\u0928\u0947 \u092e\u0928 \u0915\u0940 \u0938\u0924\u0939 \u092a\u0930 \u0939\u0940 \u091c\u0940\u0924\u0947 \u0939\u0948\u0902 \u0914\u0930 \u092c\u093f\u0928\u093e \u0916\u0941\u0926 \u0915\u0947 \u0938\u091a \u0915\u094b \u091c\u093e\u0928\u0947 \u0939\u0940 \u092e\u0930 \u092d\u0940 \u091c\u093e\u0924\u0947 \u0939\u0948\u0902\u0964 \u0907\u0938\u0915\u0947 \u0938\u093e\u0925 \u0939\u0940 \u092e\u0928 \u090f\u0915 \u0910\u0938\u0940 \u091c\u0917\u0939 \u0939\u0948 \u091c\u093f\u0938\u092e\u0947\u0902 \u0928\u0947\u0935\u093f\u0917\u0947\u091f \u0915\u0930\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u0939\u092e\u093e\u0930\u0947 \u092a\u093e\u0938 \u0915\u094b\u0908 \u092e\u0948\u092a, \u0915\u0902\u092a\u093e\u0938 \u092f\u093e \u0917\u093e\u0907\u0921 \u092d\u0940 \u0928\u0939\u0940\u0902 \u0939\u0948\u0964 \u0939\u092e\u0947\u0936\u093e \u091c\u0917\u0939 \u092a\u0930 \u0905\u0915\u0947\u0932\u0947 \u091c\u093e\u0928\u093e \u0939\u094b\u0924\u093e \u0939\u0948 \u0914\u0930 \u0916\u0941\u0926 \u0939\u0940 \u0909\u0938\u0915\u0940 \u091a\u0941\u0928\u094c\u0924\u093f\u092f\u094b\u0902 \u0915\u093e \u0938\u093e\u092e\u0928\u093e \u0915\u0930\u0928\u093e \u092a\u095c\u0924\u093e \u0939\u0948\u0964 \u0928\u0940\u091a\u093e \u090f\u0915 \u0910\u0938\u0947 \u0938\u093e\u0939\u0938\u0940 \u0910\u0921\u0935\u0947\u0902\u091a\u0930 \u0925\u0947 \u091c\u094b \u092e\u0928 \u0915\u0940 \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u092e\u0947\u0902 \u0909\u0924\u0930\u0928\u0947 \u0938\u0947 \u091c\u0941\u095c\u0947 \u091c\u094b\u0916\u093f\u092e \u0915\u094b \u0909\u0920\u093e\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u0924\u0948\u092f\u093e\u0930 \u0925\u0947\u0964 \u0935\u094b \u0905\u092a\u0928\u0940 \u092c\u0941\u0915 \u0926 \u0921\u0949\u0928 \u0911\u092b \u0921\u0947 \u092e\u0947\u0902 \u0932\u093f\u0916\u0924\u0947 \u0939\u0948\u0902 \u0915\u093f \u092e\u0948\u0902\u0928\u0947 \u090f\u0915 \u0910\u0938\u093e \u0915\u0926\u092e \u0909\u0920\u093e \u0932\u093f\u092f\u093e \u0939\u0948 \u091c\u094b \u0939\u0930 \u0915\u093f\u0938\u0940 \u0915\u094b \u0928\u0939\u0940\u0902 \u0909\u0920\u093e\u0928\u093e \u091a\u093e\u0939\u093f\u090f\u0964 \u092e\u0948\u0902 \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u092e\u0947\u0902 \u0909\u0924\u0930 \u0917\u092f\u093e \u0939\u0942\u0901\u0964 \u092e\u0948\u0902\u0928\u0947 \u092c\u0941\u0928\u093f\u092f\u093e\u0926 \u0915\u094b \u0939\u0940 \u0916\u094b\u0926\u0928\u093e \u0936\u0941\u0930\u0942 \u0915\u0930 \u0926\u093f\u092f\u093e \u0939\u0948\u0964 \u0928\u0940\u091a\u0947 \u0915\u0940 \u0907\u0928\u0930 \u090f\u0915\u094d\u0938\u092a\u094d\u0932\u094b\u0930\u0947\u0936\u0928 \u0915\u0940 \u0935\u091c\u0939 \u0938\u0947 \u0939\u092e\u0947\u0902 \u0909\u0928\u0915\u0940 \u0915\u0908 \u0905\u0928\u094b\u0916\u0940 \u0930\u093e\u0907\u091f\u093f\u0902\u0917\u094d\u0938 \u092e\u093f\u0932\u0940 \u0939\u0948\u0964 \u0932\u0947\u0915\u093f\u0928 \u091c\u092c \u0935\u094b 45 \u0938\u093e\u0932 \u0915\u0947 \u0925\u0947 \u0924\u092c \u0909\u0928\u0915\u0940 \u092f\u0939\u0940 \u0907\u0928\u0930 \u090f\u0915\u094d\u0938\u092a\u094d\u0932\u094b\u0930\u0947\u0936\u0928 \u092a\u093e\u0917\u0932\u092a\u0928 \u092a\u0930 \u091c\u093e\u0915\u0930 \u0930\u0941\u0915\u0940 \u0914\u0930 \u092f\u0939 \u091a\u0940\u095b \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u0907\u0938 \u0938\u092e\u092f \u0938\u0947 8 \u0938\u093e\u0932 \u092a\u0939\u0932\u0947 \u0939\u0940 \u092a\u094d\u0930\u0940 \u0911\u0921\u093f\u091f \u0915\u0930 \u0926\u0940 \u0925\u0940\u0964 \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u0905\u092a\u0928\u0947 \u090f\u0915 \u0932\u0947\u091f\u0930 \u092e\u0947\u0902 \u0932\u093f\u0916\u093e \u0915\u093f \u0915\u0908 \u092c\u093e\u0930 \u092e\u0947\u0930\u0947 \u092e\u0928 \u092e\u0947\u0902 \u090f\u0915 \u091a\u0947\u0924\u093e\u0935\u0928\u0940 \u0906\u0924\u0940 \u0939\u0948 \u091c\u094b \u092c\u094b\u0932\u0924\u0940 \u0939\u0948 \u0915\u093f \u092e\u0948\u0902 \u090f\u0915 \u092c\u0939\u0941\u0924 \u0939\u0940 \u0916\u0924\u0930\u0928\u093e\u0915 \u091c\u093f\u0902\u0926\u0917\u0940 \u091c\u0940 \u0930\u0939\u093e \u0939\u0942\u0901 \u0915\u094d\u092f\u094b\u0902\u0915\u093f \u092e\u0948\u0902 \u0909\u0928 \u092e\u0936\u0940\u0928\u094b\u0902 \u092e\u0947\u0902 \u0938\u0947 \u0939\u0942\u0901 \u091c\u094b \u092b\u091f \u0938\u0915\u0924\u0940 \u0939\u0948\u0964 \u0905\u092a\u0928\u0947 \u092a\u093e\u0917\u0932\u092a\u0928 \u0938\u0947 1 \u0938\u093e\u0932 \u092a\u0939\u0932\u0947 \u092f\u093e\u0928\u0940 1888 \u092e\u0947\u0902 \u0928\u0940\u091a\u0947 \u0905\u092a\u0928\u0947 \u0906\u092a\u0915\u094b \u092d\u0917\u0935\u093e\u0928 \u0914\u0930 \u092e\u0939\u093e\u0928 \u0930\u093e\u091c\u093e\u0913\u0902 \u0915\u0947 \u0928\u093e\u092e \u0938\u0947 \u090f\u0921\u094d\u0930\u0947\u0938 \u0915\u0930\u0928\u0947 \u0932\u0917\u0947 \u0925\u0947\u0964 \u0909\u0928\u0915\u0947 \u0906\u0938\u092a\u093e\u0938 \u0930\u0939\u0928\u0947 \u0935\u093e\u0932\u0947 \u0932\u094b\u0917 \u092c\u0924\u093e\u0924\u0947 \u0939\u0948\u0902 \u0915\u093f \u0935\u094b \u0905\u092a\u0928\u0947 \u0915\u092e\u0930\u0947 \u0938\u0947 \u0915\u0908 \u0926\u093f\u0928\u094b\u0902 \u0924\u0915 \u092c\u093e\u0939\u0930 \u0928\u0939\u0940\u0902 \u0928\u093f\u0915\u0932\u0924\u0947 \u0925\u0947, \u0932\u0947\u0915\u093f\u0928 \u091c\u092c \u092c\u093e\u0939\u0930 \u0906\u0924\u0947 \u0924\u092c \u090f\u0915\u0926\u092e \u0910\u0938\u0947 \u0910\u0915\u094d\u091f\u0930 \u0915\u0930\u0924\u0947 \u0914\u0930 \u0928\u093e\u091a\u0924\u0947 \u0917\u093e\u0924\u0947 \u091c\u0948\u0938\u0947 \u0909\u0928 \u092a\u0930 \u0915\u094b\u0908 \u092d\u0942\u0924 \u0938\u0935\u093e\u0930 \u0939\u094b \u0917\u092f\u093e \u0939\u094b \u0914\u0930 \u0935\u094b \u091c\u0948\u0928\u094d\u092f\u0942\u0905\u0930\u0940 \u0925\u0930\u094d\u0921 1889 \u0915\u093e \u0926\u093f\u0928 \u0925\u093e\u0964 \u091c\u092c \u0928\u0940\u091a\u093e \u0938\u095c\u0915 \u092a\u0930 \u091a\u0932\u0924\u0947 \u0939\u0941\u090f \u090f\u0915 \u0918\u094b\u095c\u0947 \u0915\u094b \u092a\u0940\u091f\u0924\u0947 \u0926\u0947\u0916 \u092c\u0939\u0941\u0924 \u0939\u0940 \u0907\u092e\u094b\u0936\u0928\u0932 \u0939\u094b \u0917\u090f \u0914\u0930 \u0909\u0938 \u0918\u094b\u095c\u0947 \u0915\u094b \u0917\u0932\u0947 \u0932\u0917\u093e\u0928\u0947 \u0932\u0917\u0947\u0964 \u0935\u094b \u0935\u0939\u0940 \u092a\u0930 \u0916\u0941\u0926 \u0915\u093e \u090f\u0915 \u0924\u092e\u093e\u0936\u093e \u092c\u0928\u093e \u0915\u0930 \u092c\u0947\u0939\u094b\u0936 \u0939\u094b \u0917\u090f \u0914\u0930 \u0935\u0939\u0940\u0902 \u0938\u0947 \u0909\u0928\u0915\u093e \u092e\u0947\u0902\u091f\u0932 \u092c\u094d\u0930\u0947\u0915 \u0921\u093e\u0909\u0928 \u0936\u0941\u0930\u0942 \u0939\u094b \u0917\u092f\u093e\u0964 \u0928\u0940\u091a\u0947 \u0915\u0947 \u0907\u0938 \u0915\u0947\u0938 \u0915\u094b \u0938\u092e\u091d\u0928\u0947 \u0938\u0947 \u0939\u092e\u0947\u0902 \u0939\u094d\u092f\u0942\u092e\u0928 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u0940 \u0915\u0940 \u090f\u0915 \u092c\u0939\u0941\u0924 \u0921\u0940\u092a \u0905\u0928\u094d\u0921\u0930\u0938\u094d\u091f\u0948\u0928\u094d\u0921\u093f\u0902\u0917 \u092e\u093f\u0932 \u0938\u0915\u0924\u0940 \u0939\u0948\u0964 \u0939\u093e\u0909\u090f\u0935\u0930 \u0906\u092e \u0924\u094c\u0930 \u092a\u0930 \u0928\u0940\u091a\u093e \u0915\u0947 \u0907\u0938 \u092a\u093e\u0917\u0932\u092a\u0928 \u0915\u094b \u0928\u094d\u092f\u0942\u0930\u094b\u0938\u093f\u092b\u0932\u093f\u0938 \u091c\u0948\u0938\u0940 \u092c\u093f\u092e\u093e\u0930\u0940 \u0915\u093e \u0928\u0924\u0940\u091c\u093e \u092c\u0924\u093e\u092f\u093e \u091c\u093e\u0924\u093e \u0939\u0948\u0964 \u0932\u0947\u0915\u093f\u0928 \u090f\u0915 \u091c\u0930\u094d\u092e\u0928 \u0938\u094d\u092a\u0947\u0936\u0932\u093f\u0938\u094d\u091f \u090f\u0930\u093f\u0915 \u092a\u094d\u0930\u094b\u0921\u0915\u094d \u0928\u0940\u091a\u0947 \u0915\u0947 \u092e\u0947\u0921\u093f\u0915\u0932 \u0930\u093f\u0915\u0949\u0930\u094d\u0921\u094d\u0938 \u0915\u094b \u090f\u0917\u094d\u095b\u0948\u092e \u0907\u0928 \u0915\u0930\u0928\u0947 \u0915\u0947 \u092c\u093e\u0926 \u0905\u092a\u0928\u0940 \u092c\u0941\u0915 \u0915\u0947 \u092e\u0948\u0921\u0928\u0947\u0938 \u0911\u092b \u0928\u0940\u091a\u093e \u092e\u0947\u0902 \u0932\u093f\u0916\u0924\u0947 \u0939\u0948\u0902 \u0915\u093f \u0938\u093f\u092b\u0932\u093f\u0938 \u0915\u0947 \u0910\u0938\u0947 \u0915\u093f\u0938\u0940 \u0907\u0928\u094d\u092b\u0947\u0915\u094d\u0936\u0928 \u0915\u093e \u0915\u094b\u0908 \u0938\u092c\u0942\u0924 \u0928\u0939\u0940\u0902 \u0939\u0948\u0964 \u0914\u0930 \u0928\u093f\u091a\u093e \u0915\u0947 \u0938\u093f\u092e\u094d\u092a\u091f\u092e\u094d\u0938 \u0938\u093f\u092b\u0932\u093f\u0938 \u0915\u0940 \u092c\u093f\u092e\u093e\u0930\u0940 \u0915\u0947 \u0938\u093e\u0925 \u0915\u0949\u0928\u094d\u0938\u094d\u091f\u0947\u0902\u091f \u092d\u0940 \u0928\u0939\u0940\u0902 \u0939\u0948\u0964 \u091c\u0940\u0935\u0928 \u0905\u092e\u0947\u0930\u093f\u0915\u0928 \u092b\u093f\u0932\u0949\u0938\u092b\u0930 \u091c\u0942\u0932\u093f\u092f\u0928 \u092f\u0902\u0917 \u092d\u0940 \u0905\u092a\u0928\u0940 \u092c\u0941\u0915 \u0928\u0940\u091a\u093e \u0905\u092c \u092c\u093e\u092f\u094b\u0917\u094d\u0930\u093e\u092b\u0940 \u092e\u0947\u0902 \u092c\u094b\u0932\u0924\u0947 \u0939\u0948\u0902\u0964 \u0915\u093f \u0928\u0940\u091a\u0947 \u0915\u0947 \u092a\u093e\u0917\u0932\u092a\u0928 \u0915\u093e \u0938\u092c\u0938\u0947 \u092c\u095c\u093e \u0914\u0930 \u0911\u092c\u094d\u0935\u093f\u092f\u0938 \u0915\u093e\u0930\u0923 \u092b\u093f\u091c\u093f\u0915\u0932 \u092f\u093e \u092b\u093f\u0930 \u091c\u0947\u0928\u0947\u091f\u093f\u0915 \u0928\u0939\u0940\u0902 \u092c\u0932\u094d\u0915\u093f \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u0928\u0947\u091a\u0930 \u0915\u093e \u0932\u0917\u0924\u093e \u0939\u0948\u0964 \u0932\u0947\u0915\u093f\u0928 \u0905\u0917\u0930 \u0928\u0940\u091a\u0947 \u0915\u094b \u092f\u0939 \u092e\u093e\u0932\u0942\u092e \u0925\u093e \u0915\u093f \u0935\u0939 \u0916\u0941\u0926 \u0915\u0947 \u0905\u0928\u0915\u0949\u0928\u094d\u0936\u0938 \u092e\u093e\u0907\u0902\u0921 \u0915\u0940 \u0921\u0947\u0925 \u092e\u0947\u0902 \u0917\u0941\u092e \u0939\u094b \u0938\u0915\u0924\u0947 \u0939\u0948\u0902 \u0924\u094b \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u092b\u093f\u0930 \u092d\u0940 \u0907\u0924\u0928\u093e \u092c\u095c\u093e \u0916\u0924\u0930\u093e \u0915\u094d\u092f\u094b\u0902 \u092e\u094b\u0932 \u0932\u093f\u092f\u093e? \u0907\u0938\u0915\u0947 \u091c\u092c \u0906\u092a\u0915\u094b \u0939\u092e \u0909\u0928\u0915\u0940 \u092c\u0941\u0915 \u0917\u0948\u0938 \u0938\u093e\u0907\u0902\u0938 \u0915\u0947 \u090f\u0915 \u092a\u0948\u0938\u0947\u091c \u092e\u0947\u0902 \u0922\u0942\u0902\u0922 \u0938\u0915\u0924\u0947 \u0939\u0948\u0902, \u091c\u0939\u093e\u0901 \u0935\u094b \u092c\u094b\u0932\u0924\u0947 \u0939\u0948 \u0915\u093f \u0935\u094b \u0905\u0938\u093e\u092e\u093e\u0928\u094d\u092f \u0926\u0930\u094d\u0926 \u0914\u0930 \u0932\u0902\u092c\u0940 \u0914\u0930 \u0938\u094d\u0932\u094b \u0938\u092b\u0930\u093f\u0902\u0917 \u0939\u0940 \u0939\u094b\u0924\u0940 \u0939\u0948\u0902 \u091c\u093f\u0928\u0915\u0940 \u0906\u0917 \u092e\u0947\u0902 \u0939\u092e \u091c\u0932\u0915\u0930 \u0905\u0902\u0924\u093f\u092e \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u0924\u0915 \u092a\u0939\u0941\u0901\u091a \u092a\u093e\u0924\u0947 \u0939\u0948\u0902\u0964 \u092f\u093e\u0928\u0940 \u0928\u0940\u091a\u0947 \u0928\u0947 \u0916\u0941\u0926 \u0915\u094b \u091c\u093e\u0928\u092c\u0942\u091d\u0915\u0930 \u0926\u0930\u094d\u0926 \u0915\u0947 \u0915\u0941\u090f\u0902 \u092e\u0947\u0902 \u092b\u0947\u0902\u0915\u093e\u0964 \u0914\u0930 \u0909\u0928\u0915\u0940 \u0907\u0938\u0940 \u0938\u092b\u0930\u093f\u0902\u0917 \u0915\u0940 \u0935\u091c\u0939 \u0938\u0947 \u0909\u0928\u0915\u0940 \u092b\u093f\u0932\u0949\u0938\u094b\u092b\u093f\u0915\u0932 \u0907\u0928\u0938\u093e\u0907\u091f\u094d\u0938 \u0907\u0924\u0928\u0940 \u0921\u0940\u092a \u0925\u0940\u0964 \u0907\u0938\u0915\u093e \u092e\u0924\u0932\u092c \u090f\u0915 \u0924\u0930\u0939 \u0938\u0947 \u0909\u0928\u0915\u093e \u0907\u0938 \u0924\u0930\u0939 \u092e\u0928 \u0915\u0940 \u0905\u0928\u091c\u093e\u0928 \u0917\u0941\u092b\u093e\u0913\u0902 \u092e\u0947\u0902 \u091c\u093e\u0928\u093e \u0909\u0928\u0915\u0947 \u0932\u093f\u090f \u0935\u093f\u095b\u094d\u0921\u092e \u0939\u093e\u0938\u093f\u0932 \u0915\u0930\u0928\u0947 \u0915\u0940 \u091c\u0930\u0942\u0930\u0924 \u0925\u0940 \u0928\u093e \u0915\u093f \u0915\u094b\u0908 \u0909\u0928\u0915\u0940 \u091a\u0949\u0907\u0938 \u091c\u093f\u0938\u0938\u0947 \u0935\u0939 \u092c\u0926\u0932 \u0938\u0915\u0924\u0947 \u0939\u0948\u0902\u0964 \u0928\u0940\u091a\u093e \u091c\u093e\u0928\u0924\u0947 \u0925\u0947 \u0915\u093f \u0909\u0928\u0915\u0947 \u0907\u0938 \u0915\u093e\u092e \u0914\u0930 \u0938\u0948\u0915\u094d\u0930\u093f\u095e\u093e\u0907\u0938 \u0915\u0940 \u0935\u0948\u0932\u094d\u092f\u0942 \u091a\u093e\u0939\u0947 \u0909\u0928\u0915\u0947 \u091c\u0940\u0924\u0947 \u091c\u0940 \u0928\u093e \u0939\u094b \u0914\u0930 \u0915\u094b\u0908 \u092c\u093e\u0924 \u0928\u0939\u0940\u0902 \u0905\u0917\u0930 \u091c\u094d\u092f\u093e\u0926\u093e \u0932\u094b\u0917 \u0909\u0928\u0915\u0940 \u092c\u0941\u0915\u094d\u0938 \u0928\u0939\u0940\u0902 \u0916\u0930\u0940\u0926 \u0930\u0939\u0947 \u0939\u0948\u0902\u0964 \u0932\u0947\u0915\u093f\u0928 \u092b\u093f\u0930 \u092d\u0940 \u0935\u094b \u092f\u0947 \u092e\u093e\u0928\u0924\u0947 \u0925\u0947 \u0915\u093f \u0909\u0928\u0915\u0940 \u0939\u0930 \u092c\u0941\u0915 \u092f\u0942\u0928\u093f\u091f\u0940 \u0915\u0947 \u0932\u093f\u090f \u090f\u0915 \u0917\u093f\u092b\u094d\u091f \u0939\u0948\u0964 \u090f\u0915 \u0932\u0947\u091f\u0930 \u092e\u0947\u0902 \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u0905\u092a\u0928\u093e \u0926\u0930\u094d\u0926 \u091c\u0924\u093e\u0924\u0947 \u0939\u0941\u090f \u0932\u093f\u0916\u093e \u0915\u093f \u092e\u0948\u0902 45 \u0938\u093e\u0932 \u0915\u093e \u0939\u094b \u091a\u0941\u0915\u093e \u0939\u0942\u0901\u0964 \u0914\u0930 15 \u0915\u093f\u0924\u093e\u092c\u0947\u0902 \u092a\u092c\u094d\u0932\u093f\u0936 \u0915\u0930 \u091a\u0941\u0915\u093e \u0939\u0942\u0901\u0964 \u0932\u0947\u0915\u093f\u0928 \u0905\u092d\u0940 \u0924\u0915 \u092e\u0947\u0930\u0940 \u0915\u093f\u0938\u0940 \u092d\u0940 \u0915\u093f\u0924\u093e\u092c \u0915\u093e \u090f\u0915 \u0905\u0915\u0947\u0932\u093e \u0905\u091a\u094d\u091b\u093e \u0930\u093f\u0935\u094d\u092f\u0941 \u092d\u0940 \u0928\u0939\u0940\u0902 \u0939\u0948\u0964 \u0932\u094b\u0917 \u092e\u0947\u0930\u0947 \u0915\u093e\u092e \u0915\u094b \u0905\u091c\u0940\u092c, \u0916\u0924\u0930\u0928\u093e\u0915 \u0914\u0930 \u092a\u093e\u0917\u0932\u092a\u0928 \u092c\u094b\u0932\u0924\u0947 \u0939\u0948\u0902\u0964 \u092e\u0941\u091d\u0947 \u092f\u0939 \u0938\u094b\u091a\u0915\u0930 \u092c\u0939\u0941\u0924 \u0926\u0941\u0916 \u0939\u094b\u0924\u093e \u0939\u0948 \u0915\u093f \u0907\u0924\u0928\u0947 \u0938\u093e\u0932\u094b\u0902 \u092e\u0947\u0902 \u0915\u093f\u0938\u0940 \u090f\u0915 \u0907\u0902\u0938\u093e\u0928 \u0928\u0947 \u092d\u0940 \u092e\u0947\u0930\u0947 \u0915\u093e\u092e \u0915\u094b \u0921\u093f\u0938\u094d\u0915\u0935\u0930\u094d\u0921 \u0928\u0939\u0940\u0902 \u0915\u0930 \u0930\u0939\u093e, \u0928\u093e \u092e\u0947\u0930\u0940 \u0915\u093f\u0938\u0940 \u0915\u094b \u091c\u0930\u0942\u0930\u0924 \u092a\u095c\u0940 \u0914\u0930 \u0928\u093e \u0939\u0940 \u092e\u0941\u091d\u0947 \u0915\u093f\u0938\u0940\u0928\u0947 \u092a\u094d\u092f\u093e\u0930 \u0915\u0930\u093e\u0902 \u0928\u0940\u091a\u0947 \u0905\u0915\u094d\u0938\u0930 \u0905\u092a\u0928\u0947 \u0905\u0915\u0947\u0932\u0947 \u092a\u0928 \u0914\u0930 \u0909\u0938\u0938\u0947 \u091c\u0941\u095c\u0947 \u0926\u0930\u094d\u0926 \u0915\u0947 \u092c\u093e\u0930\u0947 \u092e\u0947\u0902 \u092c\u0924\u093e\u0924\u0947 \u0925\u0947\u0964 \u091c\u0940\u0935\u0928 \u090f\u0915 \u0926\u0942\u0938\u0930\u0940 \u092b\u093f\u0932\u0949\u0938\u092b\u0930 \u0932\u0942 \u0938\u0932\u094b\u0928 \u0915\u0947 \u0938\u093e\u0925 \u090f\u0915 \u091b\u094b\u091f\u0947 \u0905\u092b\u0947\u092f\u0930 \u0915\u0947 \u092c\u093e\u0926 \u092d\u0940 \u0928\u0940\u091a\u0947 \u0915\u0947 \u092a\u094d\u0930\u092a\u094b\u091c\u0932 \u0915\u094b \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u0924\u0940\u0928 \u092c\u093e\u0930 \u0930\u093f\u091c\u0947\u0915\u094d\u091f \u0915\u0930 \u0926\u093f\u092f\u093e\u0964 \u092c\u093f\u0928\u093e \u0915\u093f\u0938\u0940 \u0939\u094d\u092f\u0942\u092e\u0928 \u0907\u0902\u091f\u0930\u0948\u0915\u094d\u0936\u0928 \u0914\u0930 \u0915\u0928\u0947\u0915\u094d\u0936\u0928 \u0915\u0947 \u0928\u0940\u091a\u0947 \u0916\u0941\u0926 \u0915\u094b \u090f\u0915 \u0910\u0938\u0940 \u091c\u092e\u0940\u0928 \u0938\u0947 \u0915\u0902\u092a\u0947\u0930 \u0915\u0930\u0924\u0947 \u0925\u0947 \u091c\u0939\u093e\u0901 \u092a\u0930 \u0915\u092d\u0940 \u092c\u093e\u0930\u093f\u0936 \u0928\u0939\u0940\u0902 \u0939\u094b\u0924\u0940 \u0914\u0930 \u091c\u0939\u093e \u0916\u0941\u0936 \u0939\u094b\u0928\u0947 \u0915\u0940 \u0915\u094b\u0908 \u0935\u091c\u0939 \u0928\u0939\u0940\u0902 \u0939\u094b\u0924\u0940\u0964 \u0909\u0928\u0915\u0940 \u092c\u093f\u0917\u095c\u0924\u0940 \u0939\u0947\u0932\u094d\u0925, \u0909\u0928\u0915\u0940 \u0915\u093f\u0924\u093e\u092c\u094b\u0902 \u0915\u093e \u0928 \u092c\u093f\u0915\u0928\u093e \u0914\u0930 \u0909\u0928\u0915\u093e \u0905\u0915\u0947\u0932\u093e\u092a\u0928 \u0909\u0928\u0915\u0940 \u092e\u0947\u0902\u091f\u0932 \u0939\u0947\u0932\u094d\u0925 \u0915\u094b \u0916\u0930\u093e\u092c \u0915\u0930\u0924\u093e \u091c\u093e \u0930\u0939\u093e \u0925\u093e \u0914\u0930 \u0907\u0938\u0940 \u0926\u0930\u094d\u0926 \u0928\u0947 \u0909\u0928\u094d\u0939\u0947\u0902 \u0916\u0941\u0926 \u0915\u0947 \u0905\u0902\u0926\u0930 \u091c\u093e\u0915\u0930 \u0909\u0938 \u0916\u091c\u093e\u0928\u0947 \u0915\u094b \u0922\u0942\u0902\u0922\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u092b\u094b\u0930\u094d\u0938 \u0915\u0930\u093e \u091c\u094b \u0915\u093f\u0938\u0940 \u092d\u0940 \u0907\u0902\u0938\u093e\u0928 \u0915\u094b \u092c\u0939\u0941\u0924 \u091c\u094d\u092f\u093e\u0926\u093e \u0928\u0949\u0932\u0947\u091c\u0947\u092c\u0932 \u0914\u0930 \u092a\u0949\u0935\u0930\u092b\u0941\u0932 \u092c\u0928\u093e \u0938\u0915\u0924\u093e \u0939\u0948\u0964 \u0938\u094d\u0935\u093f\u0938 \u0938\u093e\u0907\u0915\u093f\u092f\u093e\u091f\u094d\u0930\u093f\u0938\u094d\u091f \u0905\u092a\u0928\u0940 \u092c\u0941\u0915 \u0938\u093f\u0902\u092c\u0932\u094d\u0938 \u0911\u092b \u091f\u094d\u0930\u093e\u0902\u0938\u092b\u0949\u0930\u094d\u092e\u0947\u0936\u0928 \u092e\u0947\u0902 \u0932\u093f\u0916\u0924\u0947 \u0939\u0948\u0902 \u0915\u093f \u0939\u092e\u093e\u0930\u0947 \u092e\u0928 \u0915\u0940 \u0938\u092c\u0938\u0947 \u0917\u0939\u0930\u0940 \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u092e\u0947\u0902 \u092f\u093e \u092b\u093f\u0930 \u0939\u092e\u093e\u0930\u0947 \u0905\u0928\u0915\u0949\u0928\u094d\u0936\u0938 \u092e\u093e\u0907\u0902\u0921 \u0915\u0947 \u0938\u092e\u0941\u0926\u094d\u0930 \u0915\u0947 \u0924\u0932\u0947 \u092a\u0930 \u090f\u0915 \u0910\u0938\u093e \u0916\u091c\u093e\u0928\u093e \u0939\u094b\u0924\u093e \u0939\u0948 \u091c\u093f\u0938\u0947 \u0924\u0915 \u0938\u093f\u0930\u094d\u092b \u0938\u092c\u0938\u0947 \u0938\u093e\u0939\u0938\u0940 \u0932\u094b\u0917 \u0939\u0940 \u092a\u0939\u0941\u0901\u091a \u0938\u0915\u0924\u0947 \u0939\u0948\u0902\u0964 \u092f\u0947 \u0916\u091c\u093e\u0928\u093e \u090f\u0915 \u0938\u093f\u0902\u092c\u0932 \u0939\u094b\u0924\u093e \u0939\u0948 \u091c\u093f\u0902\u0926\u0917\u0940 \u0915\u0947 \u0938\u092c\u0938\u0947 \u092c\u095c\u0947 \u0938\u0940\u0915\u094d\u0930\u0947\u091f\u094d\u0938 \u0915\u093e, \u091c\u093f\u0938\u0947 \u0905\u0928\u0917\u093f\u0928\u0924 \u0924\u0930\u0940\u0915\u094b\u0902 \u0938\u0947 \u092e\u093f\u0925\u094b\u0932\u0949\u091c\u0940 \u0915\u0947 \u0925\u094d\u0930\u0942 \u090f\u0915\u094d\u0938\u092a\u094d\u0930\u0947\u0938 \u0915\u0939\u093e \u091c\u093e\u0924\u093e \u0939\u0948\u0964 \u0907\u0902\u0938\u093e\u0928\u094b\u0902 \u0928\u0947 \u0939\u092e\u0947\u0936\u093e \u0938\u0947 \u0939\u0940 \u0907\u0938 \u0916\u091c\u093e\u0928\u0947 \u0915\u0940 \u0932\u093e\u0932\u0938\u093e \u0915\u0930\u0940 \u0939\u0948\u0964 \u0928\u0940\u091a\u0947 \u0905\u092a\u0928\u0947 \u0905\u0928\u0915\u0949\u0928\u094d\u0936\u0938 \u092e\u093e\u0907\u0902\u0921 \u092e\u0947\u0902 \u0928\u0940\u091a\u0947 \u091c\u093e\u0924\u0947 \u0917\u090f \u0924\u093e\u0915\u093f \u0935\u094b \u0907\u0938 \u0916\u091c\u093e\u0928\u0947 \u0915\u094b \u092a\u093e\u0915\u0930 \u0905\u092a\u0928\u0947 \u0926\u0930\u094d\u0926 \u0915\u094b \u0938\u094b\u0928\u0947 \u092e\u0947\u0902 \u092c\u0926\u0932 \u0938\u0915\u0947\u0964 \u092f\u0947 \u090f\u0932 \u0915\u092e\u0940 \u0915\u093e \u0905\u0932\u094d\u091f\u0940\u092e\u0947\u091f \u0917\u094b\u0932 \u0925\u093e \u0915\u093f \u0915\u093f\u0938\u0940 \u0924\u0930\u0939 \u0938\u0947 \u092a\u094d\u0930\u093e\u091a\u0940\u0928 \u0915\u0947\u092e\u093f\u0938\u094d\u091f\u094d\u0930\u0940 \u0915\u094b \u092f\u0942\u095b \u0915\u0930\u093e \u091c\u093e \u0938\u0915\u0947\u0964 \u0915\u093f\u0938\u0940 \u0906\u092e \u0927\u093e\u0924\u0941 \u0915\u094b \u0938\u094b\u0928\u0947 \u092e\u0947\u0902 \u092c\u0926\u0932\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u0914\u0930 \u092f\u0939\u0940 \u0935\u093f\u095b\u094d\u0921\u092e \u0932\u093e\u0907\u091f\u092e\u0948\u0928 \u0915\u0947 \u0915\u0947\u0938 \u092e\u0947\u0902 \u092d\u0940 \u0905\u092a\u094d\u0932\u093e\u0908 \u0939\u094b\u0924\u0940 \u0939\u0948\u0964 \u0915\u094d\u092f\u094b\u0902\u0915\u093f \u0939\u0930 \u0907\u0902\u0938\u093e\u0928 \u092f\u0939 \u092c\u093e\u0924 \u091c\u093e\u0928\u0924\u093e \u0939\u0948 \u0915\u093f \u0906\u092a\u0915\u0940 \u092a\u0930\u0938\u0947\u092a\u094d\u0936\u0928 \u0914\u0930 \u0915\u0949\u0928\u094d\u0936\u0938\u0928\u0947\u0938 \u0924\u092c \u0924\u0915 \u0928\u0939\u0940\u0902 \u092c\u095d \u0938\u0915\u0924\u0940 \u091c\u092c \u0924\u0915 \u0906\u092a \u0926\u0941\u0928\u093f\u092f\u093e \u0915\u0940 \u0921\u0942\u0910\u0932\u093f\u091f\u0940 \u0915\u0947 \u092a\u093e\u0930\u094d\u091f \u0926\u0947\u0916\u0928\u093e \u0928\u0939\u0940\u0902 \u0936\u0941\u0930\u0942 \u0915\u0930\u0924\u0947 \u0939\u0948\u0902 \u0914\u0930 \u0915\u0940\u091a\u095c \u092e\u0947\u0902 \u092d\u0940 \u091b\u0941\u092a\u0947 \u0939\u0941\u090f \u0916\u091c\u093e\u0928\u0947 \u0915\u094b \u0928\u0939\u0940\u0902 \u0922\u0942\u0902\u0922 \u092a\u093e\u0924\u0947 \u0939\u0948\u0902\u0964 \u0907\u0938 \u092c\u093e\u0924 \u0915\u094b \u0932\u093f\u091f\u0930\u0932\u094d\u0932\u0940 \u0928\u0939\u0940\u0902 \u092e\u0947\u091f\u093e\u092b\u094b\u0930 \u0928\u093f\u0915\u0932\u0940 \u0938\u092e\u091d\u094b \u0915\u093f \u0939\u0930 \u0935\u094b \u091a\u0940\u095b \u091c\u093f\u0938\u0938\u0947 \u0906\u092a \u0917\u0902\u0926\u093e \u0914\u0930 \u0921\u093f\u0938\u094d\u0917\u0938\u094d\u091f\u093f\u0902\u0917 \u0938\u092e\u091d\u0924\u0947 \u0939\u094b, \u0909\u0938\u092e\u0947\u0902 \u0939\u0940 \u0906\u092a\u0915\u094b \u0917\u094d\u0930\u094b \u0915\u0930\u0928\u0947 \u0915\u0940 \u0938\u092c\u0938\u0947 \u091c\u094d\u092f\u093e\u0926\u093e \u0924\u093e\u0915\u0924 \u0939\u094b\u0924\u0940 \u0939\u0948\u0964 \u0928\u0940\u091a\u0947 \u0928\u0947 \u092d\u0940 \u0916\u0941\u0926 \u0905\u092a\u0928\u0940 \u0907\u0928 \u0905\u0930 \u091c\u0930\u094d\u0928\u0940 \u092a\u0930 \u090f\u0915 \u0932\u0947\u091f\u0930 \u092e\u0947\u0902 \u092f\u0947 \u0932\u093f\u0916\u093e \u0925\u093e \u0915\u093f \u0905\u0928\u0932\u0947\u0938 \u092e\u0941\u091d\u0947 \u0915\u093f\u0938\u0940 \u0924\u0930\u0939 \u0938\u0947 \u090f\u0932 \u0915\u092e\u0940 \u0915\u0947 \u0915\u0942\u095c\u0947 \u0915\u0947 \u0922\u0947\u0930 \u0915\u094b \u0938\u094b\u0928\u0947 \u092e\u0947\u0902 \u0915\u0928\u094d\u0935\u0930\u094d\u091f \u0915\u0930\u0928\u0947 \u0915\u0940 \u091f\u094d\u0930\u093f\u0915 \u092a\u0924\u093e \u0932\u0917 \u091c\u093e\u090f\u0964 \u092e\u0948\u0902 \u0917\u0941\u092e \u0939\u0942\u0901 \u0914\u0930 \u092e\u0941\u091d\u0947 \u0928\u0939\u0940\u0902 \u092e\u093e\u0932\u0942\u092e\u0964 \u0915\u093f \u092e\u0948\u0902 \u0905\u092a\u0928\u0947 \u092e\u0928 \u0915\u0940 \u0907\u0938 \u092d\u0942\u0932\u092d\u0941\u0932\u0948\u092f\u093e \u0938\u0947 \u0915\u0948\u0938\u0947 \u092c\u093e\u0939\u0930 \u0928\u093f\u0915\u0932\u0942\u0902\u0917\u093e\u0964 \u0939\u0930 \u0907\u0902\u0938\u093e\u0928 \u0926\u0941\u0928\u093f\u092f\u093e \u0915\u094b \u0905\u0932\u0917 \u0905\u0932\u0917 \u0932\u0947\u0935\u0932 \u092a\u0930 \u090f\u0915\u094d\u0938\u092a\u0940\u0930\u093f\u092f\u0902\u0938 \u0915\u0930\u0924\u093e \u0939\u0948 \u091c\u0948\u0938\u0947 \u092b\u093f\u091c\u093f\u0915\u0932 \u0932\u0947\u0935\u0932, \u0907\u092e\u094b\u0936\u0928\u0932 \u0932\u0947\u0935\u0932 \u0914\u0930 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u0932\u0947\u0935\u0932\u0964 \u0932\u0947\u0915\u093f\u0928 \u0905\u0917\u0930 \u0939\u092e \u0915\u093f\u0938\u0940 \u0935\u091c\u0939 \u0938\u0947 \u0938\u093f\u0930\u094d\u092b \u090f\u0915 \u0932\u0947\u0935\u0932 \u0915\u0947 \u090f\u0915\u094d\u0938\u094d\u092a\u093f\u0930\u093f\u092f\u0902\u0938 \u092e\u0947\u0902 \u0939\u0940 \u0905\u091f\u0915 \u0917\u090f \u0924\u094b \u0939\u092e\u093e\u0930\u0940 \u0938\u092b\u0930\u093f\u0902\u0917 \u0915\u093e \u0915\u094b\u0908 \u0905\u0902\u0924 \u0939\u0940 \u0928\u0939\u0940\u0902 \u0939\u094b\u0917\u093e\u0964 \u090f\u0915 \u0921\u093f\u0938\u0947\u092c\u0932\u094d\u0921 \u0907\u0902\u0938\u093e\u0928 \u0915\u093e \u090f\u0915\u094d\u0938\u092a\u093f\u0930\u093f\u092f\u0902\u0938 \u092b\u093f\u091c\u093f\u0915\u0932 \u092a\u0947\u0928 \u0914\u0930 \u0921\u093f\u0938\u0947\u092c\u093f\u0932\u093f\u091f\u0940 \u0915\u0940 \u0935\u091c\u0939 \u0938\u0947 \u092c\u0949\u0921\u0940 \u0915\u0947 \u0932\u0947\u0935\u0932 \u092a\u0930 \u0939\u0940 \u0930\u0939 \u091c\u093e\u0924\u093e \u0939\u0948\u0964 \u090f\u0915 \u0907\u092e\u094b\u0936\u0928\u0932 \u0907\u0902\u0938\u093e\u0928, \u091c\u094b \u092a\u0930\u094d\u0938\u0928\u0948\u0932\u093f\u091f\u0940 \u0935\u093e\u0907\u091c \u0915\u093e\u092b\u0940 \u0938\u0947\u0902\u091f\u0940\u092e\u0947\u0902\u091f\u0932 \u0939\u0948, \u0909\u0938\u0915\u093e \u090f\u0915\u094d\u0938\u092a\u093f\u0930\u093f\u092f\u0902\u0938 \u0907\u092e\u094b\u0936\u0928\u094d\u0938 \u0915\u0947 \u092c\u0947\u0938\u093f\u0938 \u092a\u0930 \u0939\u094b\u0924\u093e \u0939\u0948\u0964 \u0939\u093e\u0909\u090f\u0935\u0930 \u0928\u0940\u091a\u093e \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u0932\u0947\u0935\u0932 \u092a\u0930 \u0925\u0915 \u0917\u090f \u0925\u0947 \u0914\u0930 \u0909\u0928\u0915\u0940 \u0930\u093f\u0910\u0932\u093f\u091f\u0940 \u0914\u0930 \u0926\u0941\u0928\u093f\u092f\u093e \u0915\u094b \u0926\u0947\u0916\u0928\u0947 \u0915\u093e \u0928\u091c\u0930\u093f\u092f\u093e \u0909\u0928\u0915\u0940 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u0938\u092b\u0930\u093f\u0902\u0917 \u0915\u0940 \u0935\u091c\u0939 \u0938\u0947 22 \u0925\u093e\u0964 \u091c\u0940\u0935\u0928 \u092d\u0940 \u091c\u092c \u0905\u092a\u0928\u0947 \u0938\u092a\u0928\u094b\u0902 \u0914\u0930 \u0905\u0928\u0915\u0949\u0928\u094d\u0936\u0938 \u092e\u093e\u0907\u0902\u0921 \u0915\u0947 \u0938\u092c\u0938\u0947 \u092b\u0902\u0921\u093e\u092e\u0947\u0902\u091f\u0932 \u0938\u094d\u091f\u094d\u0930\u0915\u094d\u091a\u0930 \u0915\u094b \u091c\u093e\u0928\u0928\u0947 \u0915\u0940 \u0915\u094b\u0936\u093f\u0936 \u0915\u0930 \u0930\u0939\u0947 \u0925\u0947, \u0924\u092c \u0909\u0928\u094d\u0939\u0947\u0902 \u092c\u0939\u0941\u0924 \u091c\u094d\u092f\u093e\u0926\u093e \u0939\u0948\u0932\u094b \u0938\u0947 \u0928\u0947\u0936\u0928\u094d\u0938 \u0914\u0930 \u0921\u0930\u093e\u0935\u0928\u0940 \u0935\u093f\u095b\u0928 \u0938\u093e\u092e\u0928\u0947 \u0932\u0917\u0947 \u0925\u0947, \u091c\u094b \u0915\u0908 \u0918\u0902\u091f\u094b\u0902 \u0924\u0915 \u0909\u0928\u094d\u0939\u0947\u0902 \u091f\u0949\u0930\u094d\u091a\u0930 \u0915\u0930\u0924\u0947 \u0925\u0947 \u0914\u0930 \u0909\u0928\u0915\u093e \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u092a\u0947\u0928 \u092d\u0940 \u0907\u0924\u0928\u093e \u091c\u094d\u092f\u093e\u0926\u093e \u092c\u095d \u0917\u092f\u093e \u0925\u093e \u0915\u093f \u0935\u094b \u0939\u0930 \u0930\u093e\u0924 \u0905\u092a\u0928\u0947 \u0924\u0915\u093f\u090f \u0915\u0947 \u092a\u093e\u0938 \u0917\u0928 \u0930\u0916\u0915\u0930 \u0938\u094b\u0928\u0947 \u0932\u0917\u0947 \u0925\u0947 \u0924\u093e\u0915\u093f \u091c\u0940\u0938 \u092e\u0942\u0935\u092e\u0947\u0902\u091f \u092e\u0947\u0902 \u092d\u0940 \u0909\u0928\u0915\u0940 \u0938\u092b\u0930\u093f\u0902\u0917 \u0905\u0928\u092c\u0947\u092f\u0930\u0947\u092c\u0932 \u092c\u0928 \u091c\u093e\u090f\u0964 \u0924\u092c \u0935\u094b \u0909\u0938\u0940 \u0938\u092e\u092f \u0916\u0941\u0926 \u0915\u094b \u0916\u0924\u094d\u092e \u0915\u0930 \u0938\u0915\u0924\u0947 \u0939\u0948\u0902\u0964 \u0906\u092a \u092c\u0938 \u0905\u092d\u0940 \u0905\u092a\u0928\u0947 \u0907\u0928\u0915\u0940 \u0938\u0924\u0939 \u0914\u0930 \u0909\u0938\u0915\u0940 \u0926\u094b \u0924\u0940\u0928 \u0905\u091a\u094d\u091b\u0940 \u0915\u094d\u0935\u093e\u0932\u093f\u091f\u0940\u091c \u0938\u0947 \u0935\u093e\u0915\u093f\u092b \u0939\u094b\u0902\u0964 \u0932\u0947\u0915\u093f\u0928 \u091c\u094b \u0915\u093e\u0932\u0940 \u0914\u0930 \u0930\u0939\u0938\u094d\u092f\u092e\u092f\u0940 \u091a\u0940\u091c\u0947\u0902 \u0906\u092a \u0915\u0947 \u0905\u0938\u094d\u0924\u093f\u0924\u094d\u0935 \u0915\u0940 \u092c\u0941\u0928\u093f\u092f\u093e\u0926 \u0939\u0948, \u0906\u092a\u0915\u094b \u0909\u0928\u0915\u093e \u0915\u094b\u0908 \u091c\u094d\u091e\u093e\u0928 \u0939\u0940 \u0928\u0939\u0940\u0902 \u0939\u0948\u0964 \u0939\u0930 \u0938\u093e\u0907\u091f \u0915\u0947 \u090f\u0935\u0932\u094d\u092f\u0942\u0936\u0928 \u090f\u0915 \u092c\u095c\u0947 \u092c\u094d\u0930\u0947\u0915 \u0921\u093e\u0909\u0928 \u0938\u0947 \u0939\u0940 \u0936\u0941\u0930\u0942 \u0939\u094b\u0924\u093e \u0939\u0948 \u0914\u0930 \u092f\u0947 \u091a\u0940\u095b \u0906\u092a \u091c\u093f\u0924\u0928\u093e \u0938\u094b\u091a\u0924\u0947 \u0939\u094b \u0909\u0938\u0938\u0947 \u092c\u0939\u0941\u0924 \u091c\u094d\u092f\u093e\u0926\u093e \u0915\u0949\u092e\u0928 \u0939\u0948 \u0939\u093e\u0909\u090f\u0935\u0930 \u090f\u0915 \u0907\u0902\u0938\u093e\u0928 \u0907\u0938 \u091f\u094d\u0930\u093e\u0902\u0938\u092b\u0949\u0930\u094d\u092e\u0947\u0936\u0928 \u0914\u0930 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u0911\u092c\u094d\u0938\u094d\u091f\u0947\u0915\u0932 \u0915\u0940 \u0926\u0942\u0938\u0930\u0940 \u0938\u093e\u0907\u091f \u092a\u0930 \u0938\u0947\u092b\u094d\u0932\u0940 \u092a\u0939\u0941\u0901\u091a \u092a\u093e\u0924\u093e \u0939\u0948 \u092f\u093e \u092b\u093f\u0930 \u0928\u0939\u0940\u0902 \u092f\u0947 \u0921\u093f\u092a\u0947\u0902\u0921 \u0915\u0930\u0924\u093e \u0939\u0948 \u0915\u093f \u0909\u0938\u0915\u0940 \u090f\u0915\u094d\u0938\u091f\u0930\u094d\u0928\u0932 \u0932\u093e\u0907\u092b \u0915\u093f\u0924\u0928\u0940 \u0938\u094d\u091f\u0947\u092c\u0932 \u0939\u0948\u0964 \u091c\u0948\u0938\u0947 \u092f\u0941\u0917 \u0915\u0947 \u092d\u0940 \u0928\u0940\u091a\u0947 \u0915\u0940 \u0924\u0930\u0939 \u0939\u092e\u0947\u0936\u093e \u0915\u0947 \u0932\u093f\u090f \u092a\u093e\u0917\u0932 \u0939\u094b \u091c\u093e\u0928\u0947 \u0915\u0947 \u092a\u0942\u0930\u0947 \u091a\u093e\u0928\u094d\u0938\u0947\u0938 \u0925\u0947\u0964 \u0914\u0930 \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u092d\u0940 \u0905\u092a\u0928\u0940 \u0915\u0940 \u092c\u0939\u0941\u0924 \u0938\u0947 \u0917\u0939\u0930\u0940 \u0907\u0928\u0938\u093e\u0907\u091f\u094d\u0938 \u0915\u0940 \u092e\u0926\u0926 \u0938\u0947 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u0940 \u0915\u0940 \u092b\u0940\u0932\u094d\u0921 \u0915\u094b \u0907\u0924\u0928\u093e \u0930\u093f\u0932\u093e\u090f\u092c\u0932 \u0914\u0930 \u0907\u0902\u091f\u094d\u0930\u0947\u0938\u094d\u091f\u093f\u0902\u0917 \u092c\u0928\u093e\u092f\u093e\u0964 \u0932\u0947\u0915\u093f\u0928 \u091c\u093f\u0928 \u091a\u0940\u091c\u094b\u0902 \u0928\u0947 \u092f\u094b\u0917 \u0915\u094b \u0905\u0928\u0915\u0949\u0928\u094d\u0936\u0938 \u092e\u093e\u0907\u0902\u0921 \u0915\u0940 \u092a\u093e\u0935\u0930 \u0915\u0940 \u0906\u0917\u0947 \u092d\u0940 \u0938\u0932\u093e\u092e\u0924 \u0930\u0916\u093e, \u0935\u0939 \u0925\u0940 \u0909\u0928\u0915\u0940 \u092b\u0948\u092e\u093f\u0932\u0940\u0964 \u0909\u0928\u0915\u0947 \u092a\u0947\u0936\u0947\u0902\u091f\u094d\u0938, \u0909\u0928\u0915\u0947 \u0921\u0947\u0932\u0940 \u0930\u0942\u091f\u0940\u0928 \u0938\u094d \u0909\u0928\u0915\u0947 \u092c\u091a\u094d\u091a\u0947 \u0914\u0930 \u0909\u0928\u0915\u0940 \u0935\u093f\u095b\u094d\u0921\u092e \u091c\u094b \u0909\u0928\u094d\u0939\u0947\u0902 \u092c\u094b\u0932\u0924\u0940 \u0925\u0940 \u0915\u093f \u0935\u094b \u0905\u092a\u0928\u0947 \u0935\u093f\u095b\u0928 \u0938\u0947 \u092e\u0947\u0902 \u0926\u093f\u0916\u0940, \u0915\u093f\u0938\u0940 \u092d\u0940 \u091a\u0940\u095b \u0938\u0947 \u0921\u093f\u091f\u0948\u091a \u0930\u0939\u0947, \u092f\u0939\u0940 \u0924\u094b \u0930\u0940\u095b\u0928 \u0939\u0948 \u0915\u093f \u0915\u094d\u092f\u094b\u0902 \u0939\u093e\u0907\u0932\u0940 \u0910\u0921\u0935\u093e\u0928\u094d\u0938 \u0921\u0940 \u0939\u094b\u0917\u0940\u0964 \u0907\u0938 \u092d\u0940 \u0905\u092a\u0928\u0947 \u092e\u0928 \u0915\u0940 \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u0938\u0947 \u0935\u093e\u0915\u093f\u092b \u0924\u094b \u0939\u094b\u0924\u0947 \u0939\u0948\u0902 \u0932\u0947\u0915\u093f\u0928 \u091c\u0938\u094d\u091f \u092c\u093f\u0915\u0949\u095b \u0935\u094b \u0905\u092a\u0928\u0940 \u0938\u093e\u0907\u091f \u0915\u0947 \u0915\u0949\u0928\u094d\u091f\u0947\u0902\u091f \u0938\u0947 \u0915\u092d\u0940 \u092d\u0940 \u0915\u094b\u0908 \u0905\u091f\u0948\u091a\u092e\u0947\u0902\u091f \u0939\u0940 \u0928\u0939\u0940\u0902 \u092c\u0928\u093e\u0924\u0947\u0964 \u0935\u094b \u0905\u0928\u0915\u0949\u0928\u094d\u0936\u0938 \u092e\u093e\u0907\u0902\u0921 \u0915\u0940 \u0917\u093f\u0930\u092b\u094d\u0924 \u0938\u0947 \u0938\u0947\u092b \u0930\u0939\u0924\u0947 \u0939\u0948\u0902\u0964 \u0932\u0947\u0915\u093f\u0928 \u0928\u0940\u091a\u0947 \u0915\u0940 \u0938\u092c\u0938\u0947 \u092c\u095c\u0940 \u0917\u0932\u0924\u0940 \u092f\u0939 \u0925\u0940 \u0915\u093f \u0935\u094b \u0916\u0941\u0926 \u0915\u094b \u091c\u0940\u0938\u0938, \u092c\u0941\u0926\u094d\u0927, \u0910\u0932\u0948\u0915\u094d\u095b\u0948\u0923\u094d\u0921\u0930 \u0915\u0947 \u0917\u094d\u0930\u0947\u091f \u0914\u0930 \u0939\u0930 \u092c\u095c\u0947 \u092d\u0917\u0935\u093e\u0928 \u092f\u093e \u092b\u093f\u0930 \u0935\u094b \u0930\u093f\u0935\u0930\u094d\u0938 \u0906\u0907\u0921\u0947\u0902\u091f\u093f\u092b\u093e\u0907 \u0915\u0930\u0928\u0947 \u0932\u0917\u0947 \u0925\u0947\u0964 \u0928\u0940\u091a\u0947 \u0905\u092a\u0928\u0947 \u092a\u0948\u0930\u094b\u0902 \u0915\u0947 \u0928\u0940\u091a\u0947 \u0938\u0947 \u091c\u092e\u0940\u0928 \u0916\u094b\u0928\u0947 \u0932\u0917\u0947 \u0925\u0947\u0964 \u091c\u0938\u094d\u091f \u092c\u093f\u0915\u0949\u095b \u0909\u0928\u0915\u0940 \u090f\u0915 \u0930\u093f\u0910\u0932\u093f\u091f\u0940 \u0915\u094b \u092c\u0948\u0932\u0947\u0902\u0938 \u0915\u0930\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u0909\u0928\u0915\u0940 \u0906\u0909\u091f\u0930 \u0930\u093f\u0910\u0932\u093f\u091f\u0940 \u092e\u0947\u0902 \u0915\u0941\u091b \u092d\u0940 \u0928\u0949\u0930\u094d\u092e\u0932 \u092f\u093e \u092b\u093f\u0930 \u0938\u094d\u091f\u0947\u092c\u0932 \u0928\u0939\u0940\u0902 \u0925\u093e\u0964 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u0940 \u092e\u0947\u0902 \u0907\u0938 \u0915\u0902\u0921\u093f\u0936\u0928 \u0915\u094b \u092c\u094b\u0932\u0924\u0947 \u0939\u0948\u0902 \u0938\u093e\u0907\u0915\u093f\u0915 \u0907\u0928\u094d\u092b\u094d\u0932\u0947\u0936\u0928, \u091c\u0939\u093e\u0901 \u092a\u0930 \u090f\u0915 \u0907\u0902\u0938\u093e\u0928 \u0915\u0940 \u0938\u094d\u092a\u094d\u0930\u0947 \u0916\u0941\u0926 \u0909\u0938 \u0907\u0902\u0938\u093e\u0928 \u0915\u094b \u092e\u0928 \u0914\u0930 \u0936\u0930\u0940\u0930 \u0915\u0947 \u092a\u093e\u0930 \u091c\u093e\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u092a\u0941\u0936 \u0915\u0930\u0928\u0947 \u0915\u0947 \u092c\u091c\u093e\u092f \u0915\u0930\u092a\u094d\u091f \u092c\u0928\u0915\u0930 \u092c\u0938 \u0916\u0941\u0926 \u0938\u0947 \u092c\u095c\u0940 \u091a\u0940\u091c\u094b\u0902 \u0915\u0947 \u0906\u0907\u0921\u093f\u092f\u093e \u0906\u0907\u0921\u0947\u0902\u091f\u093f\u092b\u093e\u0907 \u0915\u0930\u0928\u0947 \u0932\u0917\u0924\u0940 \u0939\u0948 \u0914\u0930 \u0907\u0938\u0940 \u0938\u093e\u0907\u0915\u093f\u0915 \u0915\u0930\u092a\u094d\u0936\u0928 \u0938\u0947 \u0939\u0940 \u0938\u0941\u092a\u093f\u0930\u093f\u092f\u094b\u0930\u093f\u091f\u0940 \u0914\u0930 \u0917\u0949\u0921 \u0915\u0949\u092e\u094d\u092a\u094d\u0932\u0947\u0915\u094d\u0938 \u091c\u0948\u0938\u0940 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u092a\u094d\u0930\u0949\u092c\u094d\u0932\u092e\u094d\u0938 \u0915\u092d\u0940 \u091c\u0928\u092e \u0939\u094b\u0924\u093e \u0939\u0948\u0964 \u0928\u0940\u091a\u0947 \u0915\u093e \u092a\u093e\u0917\u0932\u092a\u0928 \u0905\u0938\u0932\u0940 \u0938\u094d\u092a\u093f\u0930\u093f\u091a\u0942\u0910\u0932\u093f\u091f\u0940 \u0938\u0947 \u091c\u094d\u092f\u093e\u0926\u093e \u0926\u0942\u0930 \u0928\u0939\u0940\u0902 \u0939\u0948\u0964 \u092c\u0938 \u092b\u0930\u094d\u0915 \u0907\u0924\u0928\u093e \u0939\u0948 \u0915\u093f \u0938\u094d\u092a\u093f\u0930\u093f\u091a\u0941\u0905\u0932 \u0932\u094b\u0917 \u092c\u094b\u0932\u0924\u0947 \u0939\u0948 \u0915\u093f \u0939\u0930 \u0915\u093f\u0938\u0940 \u092e\u0947\u0902 \u092d\u0917\u0935\u093e\u0928 \u0939\u0948\u0902, \u0932\u0947\u0915\u093f\u0928 \u0928\u0940\u091a\u093e \u0905\u092a\u0928\u0947 \u092e\u0928 \u0938\u0947 \u0906\u0907\u0921\u0947\u0902\u091f\u093f\u092b\u093e\u0907 \u0915\u0930\u0928\u0947 \u0915\u0940 \u0935\u091c\u0939 \u0938\u0947 \u092f\u0947 \u092c\u094b\u0932\u0924\u0947 \u0925\u0947 \u0915\u093f \u092e\u0948\u0902 \u0939\u0940 \u092d\u0917\u0935\u093e\u0928 \u0939\u0942\u0901 \u0914\u0930 \u091c\u0948\u0938\u093e \u0915\u093f \u0928\u0940\u091a\u093e \u0928\u0947 \u0905\u092a\u0928\u0940 \u092c\u0941\u0915 \u0926\u0947\u0917\u093e \u0938\u093e\u0907\u0902\u0938 \u092e\u0947\u0902 \u092c\u094b\u0932\u093e \u0925\u093e \u0915\u093f \u091c\u094b \u092d\u0940 \u0905\u092a\u0928\u0947 \u0905\u0902\u0926\u0930 \u091d\u093e\u0902\u0915\u0915\u0930 \u0916\u0941\u0926 \u092e\u0947\u0902 \u0938\u092e\u093e\u090f \u092c\u094d\u0930\u0939\u094d\u092e\u093e\u0902\u0921 \u0915\u094b \u0926\u0947\u0916 \u0932\u0947\u0924\u093e \u0939\u0948, \u0909\u0938\u0947 \u092a\u0924\u093e \u0939\u094b\u0924\u093e \u0939\u0948 \u0915\u093f \u092f\u0947 \u092c\u094d\u0930\u0939\u094d\u092e\u093e\u0902\u0921 \u0915\u093f\u0924\u0928\u093e \u0905\u0928\u093f\u092f\u092e\u093f\u0924 \u0939\u0948\u0964 \u0914\u0930 \u0915\u093f\u0938 \u0924\u0930\u0939 \u0938\u0947 \u092f\u0939 \u0939\u092e\u0947\u0902 \u0905\u0938\u094d\u0924\u093f\u0924\u094d\u0935 \u0915\u0940 \u0905\u0902\u0924\u0939\u0940\u0928 \u092d\u0942\u0932\u092d\u0941\u0932\u0948\u092f\u093e \u0924\u0915 \u092a\u0939\u0941\u0902\u091a\u093e \u0938\u0915\u0924\u093e \u0939\u0948?",
"target": "\u092b\u094d\u0930\u0947\u0921\u0930\u093f\u0915 \u0928\u0940\u0924\u094d\u0936\u0947 \u0905\u092c \u0924\u0915 \u0915\u0947 \u0938\u092c\u0938\u0947 \u092a\u094d\u0930\u092d\u093e\u0935\u0936\u093e\u0932\u0940 \u0926\u093e\u0930\u094d\u0936\u0928\u093f\u0915\u094b\u0902 \u092e\u0947\u0902 \u0938\u0947 \u090f\u0915 \u0925\u0947\u0964 \u0935\u0939 \u092a\u093e\u0930\u0902\u092a\u0930\u093f\u0915 \u0928\u0948\u0924\u093f\u0915\u0924\u093e \u0914\u0930 \u0927\u0930\u094d\u092e \u0915\u0940 \u0906\u0932\u094b\u091a\u0928\u093e \u0915\u0930\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u092a\u094d\u0930\u0938\u093f\u0926\u094d\u0927 \u0939\u0948\u0902\u0964 \u0932\u0947\u0915\u093f\u0928 \u0905\u092a\u0928\u0947 \u092a\u0942\u0930\u0947 \u091c\u0940\u0935\u0928 \u092e\u0947\u0902 \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u092c\u0939\u0941\u0924 \u0936\u093e\u0930\u0940\u0930\u093f\u0915 \u0914\u0930 \u092e\u093e\u0928\u0938\u093f\u0915 \u092a\u0940\u0921\u093c\u093e \u091d\u0947\u0932\u0940\u0964 \u092a\u0939\u0932\u0947 \u0926\u0930\u094d\u0926 \u0936\u093e\u0930\u0940\u0930\u093f\u0915 \u0925\u093e, \u092b\u093f\u0930 \u092e\u093e\u0928\u0938\u093f\u0915 \u0930\u0942\u092a \u0938\u0947 \u092d\u0940 \u0938\u0924\u093e\u0928\u0947 \u0932\u0917\u093e\u0964 \u0927\u0940\u0930\u0947-\u0927\u0940\u0930\u0947 \u0935\u0939 \u0916\u0941\u0926 \u0915\u094b \u092a\u093e\u0917\u0932\u092a\u0928 \u0915\u0940 \u0913\u0930 \u0932\u0947 \u0917\u092f\u093e\u0964 \u0932\u0947\u0915\u093f\u0928 \u0907\u0938\u0915\u0947 \u092a\u0940\u091b\u0947 \u092e\u0941\u0916\u094d\u092f \u0915\u093e\u0930\u0923 \u0915\u094d\u092f\u093e \u0925\u093e? \u0915\u0941\u091b \u0932\u094b\u0917 \u0936\u093e\u0930\u0940\u0930\u093f\u0915 \u0930\u094b\u0917\u094b\u0902 \u0915\u094b \u091c\u093f\u092e\u094d\u092e\u0947\u0926\u093e\u0930 \u0920\u0939\u0930\u093e\u0924\u0947 \u0939\u0948\u0902, \u0932\u0947\u0915\u093f\u0928 \u0939\u093e\u0932 \u0939\u0940 \u092e\u0947\u0902 \u0915\u0908 \u0935\u093f\u0926\u094d\u0935\u093e\u0928\u094b\u0902 \u0915\u093e \u091d\u0941\u0915\u093e\u0935 \u0938\u094d\u0935\u092a\u094d\u0930\u0947\u0930\u093f\u0924 \u092a\u093e\u0917\u0932\u092a\u0928 \u0915\u0940 \u0938\u0902\u092d\u093e\u0935\u0928\u093e \u0915\u0940 \u0913\u0930 \u0930\u0939\u093e \u0939\u0948\u0964 \u092e\u0924\u0932\u092c \u0939\u094b \u0938\u0915\u0924\u093e \u0939\u0948 \u0915\u093f \u0928\u0940\u0924\u094d\u0936\u0947 \u0905\u092a\u0928\u0940 \u092e\u0930\u094d\u091c\u0940 \u0938\u0947 \u092a\u093e\u0917\u0932 \u0939\u094b \u0917\u092f\u093e \u0939\u094b, \u0909\u0938\u0928\u0947 \u092f\u0939 \u091c\u094b\u0916\u093f\u092e \u0909\u0920\u093e\u092f\u093e\u0964 \u0907\u0938 \u092a\u0949\u0921\u0915\u093e\u0938\u094d\u091f \u0938\u0947\u0917\u092e\u0947\u0902\u091f \u092e\u0947\u0902 \u0939\u092e \u0907\u0938 \u0935\u093f\u091a\u093e\u0930 \u0915\u094b \u0914\u0930 \u0905\u0927\u093f\u0915 \u090f\u0915\u094d\u0938\u092a\u094d\u0932\u094b\u0930 \u0915\u0930\u0947\u0902\u0917\u0947 \u0914\u0930 \u0938\u092e\u091d\u0947\u0902\u0917\u0947 \u0915\u093f \u0928\u0940\u0924\u094d\u0936\u0947 \u0928\u0947 \u091c\u093f\u0938 \u0924\u0930\u0939 \u0938\u0947 \u0915\u093f\u092f\u093e \u0909\u0938\u0915\u093e \u0905\u0902\u0924 \u0915\u094d\u092f\u094b\u0902 \u0939\u0941\u0906\u0964"
},
{
"text": "\u0906\u0938\u093e\u0928 \u0914\u0930 \u092e\u0941\u0936\u094d\u0915\u093f\u0932 \u092f\u0947 \u0936\u092c\u094d\u0926 \u0939\u0930 \u0907\u0902\u0938\u093e\u0928 \u0915\u0947 \u0932\u093f\u090f \u092c\u0939\u0941\u0924 \u0939\u0940 \u0930\u093f\u0932\u0947\u091f\u093f\u0935 \u0939\u094b\u0924\u0947 \u0939\u0948\u0964 \u091c\u0948\u0938\u0947 120 \u0915\u093f\u0932\u094b \u0915\u0947 \u0921\u0902\u092c\u0932 \u0915\u0908 \u0932\u094b\u0917\u094b\u0902 \u0915\u0947 \u0932\u093f\u090f \u092c\u0939\u0941\u0924 \u092d\u093e\u0930\u0940 \u0939\u094b\u0924\u093e \u0939\u0948 \u0914\u0930 \u0926\u094b \u091a\u093e\u0930 \u0930\u0947\u092a\u093f\u091f\u0947\u0936\u0928 \u092e\u0947\u0902 \u0939\u0940 \u0935\u0939 \u0916\u0941\u0926 \u0915\u094b \u091a\u094b\u091f \u092a\u0939\u0941\u0902\u091a\u093e \u0938\u0915\u0924\u0947 \u0939\u0948\u0902, \u091c\u092c\u0915\u093f \u0915\u0908 \u0932\u094b\u0917 \u0907\u0924\u0928\u0947 \u0935\u091c\u0928 \u0938\u0947 \u0935\u0949\u0930\u094d\u092e\u0905\u092a \u0915\u0930\u0924\u0947 \u0939\u0948\u0902\u0964 \u0915\u093f\u0938\u0940 \u092d\u0940 \u0915\u093e\u092e \u0915\u0940 \u0921\u093f\u092b\u093f\u0915\u0932\u094d\u091f\u0940 \u0921\u093f\u092a\u0947\u0902\u0921 \u0915\u0930\u0924\u0940 \u0939\u0948\u0964 \u0909\u0938 \u0915\u093e\u092e \u0915\u094b \u0915\u0930\u0928\u0947 \u0935\u093e\u0932\u0947 \u0907\u0902\u0938\u093e\u0928 \u0915\u0940 \u0915\u093e\u092c\u093f\u0932\u093f\u092f\u0924 \u0915\u0947 \u090a\u092a\u0930 \u0905\u0917\u0930 \u090f\u0915 \u0938\u094d\u091f\u0942\u0921\u0947\u0902\u091f \u0928\u0947 \u090f\u0917\u094d\u095b\u0948\u092e \u0915\u0940 \u092a\u0942\u0930\u0940 \u0924\u0948\u092f\u093e\u0930\u0940 \u0915\u0930\u0940 \u0939\u0948 \u0914\u0930 \u092a\u093f\u091b\u0932\u0947 \u0938\u093e\u0932\u094b\u0902 \u0915\u0947 \u0915\u094d\u0935\u0947\u0936\u094d\u091a\u0928 \u092a\u0947\u092a\u0930\u094d\u0938 \u0915\u094b \u092d\u0940 \u0938\u094d\u091f\u0921\u0940 \u0915\u0930 \u0930\u0939\u093e \u0939\u0948 \u0924\u094b \u0909\u0938\u0915\u0947 \u0932\u093f\u090f \u0935\u094b \u090f\u0917\u094d\u095b\u0948\u092e \u0906\u0938\u093e\u0928 \u0939\u094b\u0917\u093e, \u091c\u092c\u0915\u093f \u092c\u093f\u0928\u093e \u0924\u0948\u092f\u093e\u0930\u0940 \u0935\u093e\u0932\u0947 \u0938\u094d\u091f\u0942\u0921\u0947\u0902\u091f \u0915\u094b \u090f\u0917\u094d\u095b\u0948\u092e \u092e\u0941\u0936\u094d\u0915\u093f\u0932 \u0914\u0930 \u0906\u0909\u091f \u0911\u092b \u0938\u093f\u0932\u0947\u092c\u0938 \u0932\u0917\u0947\u0917\u093e\u0964 \u0938\u093f\u092e\u093f\u0932\u0930\u0932\u0940 \u091c\u093f\u0928 \u0932\u094b\u0917\u094b\u0902 \u0915\u094b \u091c\u093f\u0902\u0926\u0917\u0940 \u0915\u0940 \u092a\u0942\u0930\u0940 \u0938\u092e\u091d \u0939\u094b\u0924\u0940 \u0939\u0948 \u0935\u094b \u0936\u0924\u0930\u0902\u091c \u0915\u0947 \u0916\u0947\u0932 \u0915\u0940 \u0924\u0930\u0939 \u0926\u0941\u0936\u094d\u092e\u0928 \u0915\u0940 \u0939\u0930 \u091a\u093e\u0932 \u0915\u094b \u092a\u0939\u0932\u0947 \u0926\u0947\u0916\u0924\u0947 \u0914\u0930 \u0938\u092e\u091d\u0924\u0947 \u0939\u0948\u0902\u0964 \u092a\u0942\u0930\u0947 \u091a\u0947\u0938 \u092c\u094b\u0930\u094d\u0921 \u0915\u093e \u092e\u0941\u0906\u092f\u0928\u093e \u0915\u0930\u0924\u0947 \u0939\u0948\u0902 \u0914\u0930 \u092b\u093f\u0930 \u0905\u0917\u0932\u0940 \u091a\u093e\u0932 \u091a\u0932\u0924\u0947 \u0939\u0948\u0902 \u0907\u0928 \u0911\u0930\u094d\u0921\u0930 \u091f\u0942 \u0935\u093f\u0928 \u0932\u093e\u0907\u092b \u092f\u0942 \u0928\u0940\u0921 \u091f\u0941 \u092a\u0941\u091f \u092f\u094b\u0930\u0938\u0947\u0932\u094d\u092b \u0907\u0928 \u0926 \u092e\u094b\u0938\u094d\u091f \u090f\u0921\u0935\u093e\u0902\u091f\u0947\u091c \u092f\u0938 \u090f\u0915\u094d\u0938\u091f\u0930\u094d\u0928\u0932 \u0910\u0902\u0921 \u0907\u0902\u091f\u0930\u0928\u0932 \u0938\u093f\u091a\u0941\u090f\u0936\u0928\u094d\u0938 \u0906\u092a \u091a\u093e\u0939\u0947 \u0915\u093f\u0938\u0940 \u092d\u0940 \u092c\u0948\u0915\u0917\u094d\u0930\u093e\u0909\u0902\u0921 \u0938\u0947 \u0939\u094b \u0914\u0930 \u0932\u093e\u0907\u092b \u092e\u0947\u0902 \u0915\u093f\u0924\u0928\u0947 \u092d\u0940 \u092a\u0940\u091b\u0947 \u0939\u094b, \u0906\u092a \u0917\u0947\u092e \u0915\u094b \u091c\u0940\u0924 \u0938\u0915\u0924\u0947 \u0939\u094b\u0964 \u0905\u092a\u0928\u0940 \u0932\u093e\u0907\u092b \u092e\u0947\u0902 \u0907\u0928 \u092a\u093e\u0902\u091a \u0932\u093e\u0907\u0938\u0947\u0902\u0938 \u0915\u094b \u0905\u092a\u0928\u093e\u0915\u0930 \u0928\u0902\u092c\u0930 \u0935\u0928 \u0925\u093f\u0902\u0915 \u0907\u091f\u094d\u0938 \u0938\u094d\u092a\u0940\u0915 \u0905\u092c\u093e\u0909\u091f \u092f\u094b\u0930 \u0917\u094b\u0932\u094d\u0938 \u0906\u0938\u093f\u092b \u0926\u0947 \u0939\u0948\u0935 \u0911\u0932\u0930\u0947\u0921\u0940 \u0939\u0948\u092a\u0928\u094d\u0921\u0964 \u0939\u092e\u093e\u0930\u093e \u092e\u0928 \u0939\u092e\u0947\u0936\u093e \u092c\u0938 \u0926\u094b \u0935\u093f\u0930\u094b\u0927\u0940 \u0916\u094d\u092f\u093e\u0932\u094b\u0902 \u0915\u0947 \u092c\u093e\u0930\u0947 \u092e\u0947\u0902 \u0939\u0940 \u0938\u094b\u091a \u0938\u0915\u0924\u093e \u0939\u0948\u0964 \u091c\u0948\u0938\u0947 \u0905\u091a\u094d\u091b\u093e \u0914\u0930 \u092c\u0941\u0930\u093e \u092f\u093e \u092b\u093f\u0930 \u092a\u0949\u095b\u093f\u091f\u093f\u0935 \u0914\u0930 \u0928\u0947\u0917\u0947\u091f\u093f\u0935\u0964 \u0906\u092a \u091c\u0940\u0938 \u092d\u0940 \u0916\u094d\u092f\u093e\u0932 \u092a\u0930 \u0905\u092a\u0928\u0940 \u0905\u091f\u0947\u0928\u094d\u0936\u0928 \u0926\u0947\u0924\u0947 \u0939\u094b \u0914\u0930 \u091c\u0940\u0938 \u092d\u0940 \u0924\u0930\u0939 \u0915\u0940 \u092c\u093e\u0924\u0947\u0902 \u0915\u0930\u0924\u0947 \u0939\u094b\u0964 \u0906\u092a\u0915\u093e \u092e\u0928 \u0906\u092a\u0915\u0947 \u092a\u0942\u0930\u0947 \u0938\u093f\u0938\u094d\u091f\u092e \u0915\u094b \u0909\u0938\u0940 \u092b\u094d\u0932\u0947\u0935\u0930 \u0915\u093e \u092c\u0928\u093e \u0926\u0947\u0924\u093e \u0939\u0948\u0964 \u092e\u0928 \u0938\u093f\u0930\u094d\u092b \u0910\u0921\u093f\u0936\u0928 \u0914\u0930 \u092e\u0932\u094d\u091f\u093f\u092a\u094d\u0932\u093f\u0915\u0947\u0936\u0928 \u0915\u094b \u091c\u093e\u0928\u0924\u093e \u0939\u0948 \u0914\u0930 \u0909\u0938\u0947 \u0938\u092c \u091f\u094d\u0930\u0948\u0915\u094d\u0936\u0928 \u0914\u0930 \u0921\u093f\u0935\u093f\u095b\u0928 \u0915\u0930\u0928\u093e \u0928\u0939\u0940\u0902 \u0906\u0924\u093e\u0964 \u092f\u093e\u0928\u0940 \u0906\u092a \u0915\u093f\u0938\u0940 \u092d\u0940 \u0916\u094d\u092f\u093e\u0932 \u0915\u094b \u0905\u092a\u0928\u0947 \u092e\u0928 \u0938\u0947 \u092b\u094b\u0930\u094d\u0938\u094d\u092b\u0941\u0932\u0940 \u092c\u093e\u0939\u0930 \u0928\u0939\u0940\u0902 \u0928\u093f\u0915\u093e\u0932 \u0938\u0915\u0924\u0947 \u0939\u0948\u0902\u0964 \u0906\u092a \u092c\u0938 \u0928\u090f \u0916\u094d\u092f\u093e\u0932\u094b\u0902 \u0915\u094b \u0910\u0921 \u0915\u0930 \u0938\u0915\u0924\u0947 \u0939\u094b\u0964 \u0926\u0948\u091f \u0938\u094d\u0935\u093e\u0907 \u0916\u0941\u0926 \u0915\u094b \u0930\u093f\u092a\u0940\u091f\u0947\u0921 \u0932\u0940 \u092a\u0949\u095b\u093f\u091f\u093f\u0935, \u092b\u0949\u0930\u094d\u092e\u0947\u0902\u0936\u0928\u094d\u0938 \u0914\u0930 \u0916\u0941\u0926 \u0915\u0947 \u0917\u094b\u0932 \u0938\u0947 \u092b\u0940\u0921 \u0915\u0930\u0928\u093e \u092c\u0939\u0941\u0924 \u095b\u0930\u0942\u0930\u0940 \u0939\u094b\u0924\u093e \u0939\u0948\u0964 \u0916\u0941\u0926 \u0915\u094b \u092a\u0949\u095b\u093f\u091f\u093f\u0935 \u0932\u0940 \u091f\u094d\u0930\u093e\u0902\u0938\u092b\u0949\u0930\u094d\u092e \u0915\u0930\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u091c\u092c \u092d\u0940 \u0906\u092a \u0905\u092a\u0928\u0947 \u0917\u094b\u0932\u094d\u0938 \u0915\u0947 \u092c\u093e\u0930\u0947 \u092e\u0947\u0902 \u0938\u094b\u091a\u094b \u092f\u093e \u092c\u094b\u0932\u094b \u0924\u092c \u0910\u0938\u093e \u092b\u0940\u0932 \u0915\u0930\u094b \u091c\u0948\u0938\u0947 \u0906\u092a\u0928\u0947 \u0909\u0938 \u0917\u094b\u0932 \u0915\u094b \u0911\u0932\u0930\u0947\u0921\u0940 \u0905\u091a\u0940\u0935 \u0915\u0930 \u0932\u093f\u092f\u093e \u0939\u0948\u0964 \u092b\u0949\u0930 \u090f\u0917\u094d\u095b\u0948\u092e\u094d\u092a\u0932 \u092f\u0947 \u092c\u093f\u0932\u0940\u0935 \u0915\u0930\u0928\u0947 \u0938\u0947 \u0915\u093f \u0906\u092a \u0911\u0932\u0930\u0947\u0921\u0940 \u090f\u0915 \u0905\u0930\u092c\u092a\u0924\u093f \u0939\u094b, \u0906\u092a\u0915\u093e \u092c\u0939\u0941\u0924 \u092c\u095c\u093e \u092c\u093f\u095b\u0928\u0947\u0938 \u0939\u0948 \u0906\u092a\u0915\u0947 \u0905\u0902\u0926\u0930 \u092c\u0939\u0941\u0924 \u0938\u093e\u0930\u0947 \u090f\u092e\u094d\u092a\u0932\u0949\u0908\u0938 \u0915\u093e\u092e \u0915\u0930\u0924\u0947 \u0939\u0948\u0902 \u0914\u0930 \u0906\u092a \u0926\u0947\u0936 \u0915\u0940 \u0917\u094d\u0930\u094b\u0925 \u092e\u0947\u0902 \u092c\u0939\u0941\u0924 \u091c\u094d\u092f\u093e\u0926\u093e \u0915\u0949\u0928\u094d\u091f\u094d\u0930\u093f\u092c\u094d\u092f\u0942\u091f \u0915\u0930 \u0930\u0939\u0947 \u0939\u094b\u0964 \u092f\u0942 \u0938\u094d\u091f\u093e\u0930\u094d\u091f \u092c\u093f\u0939\u0947\u0935 \u0907\u0928 \u0932\u093e\u0907\u0915 \u0930\u093f\u091a \u0930\u093f\u0938\u094d\u092a\u0949\u0928\u094d\u0938\u093f\u092c\u0932 \u0910\u0902\u0921 \u092e\u0948\u091a\u094d\u092f\u094b\u0930 \u092a\u0930\u094d\u0938\u0928 \u0906\u092a \u090f\u0915 \u0932\u0940\u0921\u0930 \u0915\u0940 \u0924\u0930\u0939 \u0916\u095c\u0947 \u0939\u094b\u0928\u0947 \u0914\u0930 \u092c\u093e\u0924 \u0915\u0930\u0928\u0947 \u0932\u0917\u0924\u0947 \u0939\u094b\u0964 \u0925\u093f\u0938 \u0907\u0938 \u0928\u0949\u091f \u092b\u0947\u0915 \u0907\u091f \u091f\u093f\u0932 \u092f\u0942 \u092e\u0947\u0915 \u0907\u091f \u0925\u093f\u0938 \u0907\u0938 \u092e\u0947\u0915\u093f\u0902\u0917 \u0907\u091f \u0907\u0902\u091f\u0930\u0928\u0932\u0940 \u092c\u093f\u092b\u094b\u0930 \u092f\u0942 \u092e\u0947\u0915 \u0907\u091f \u090f\u0915\u094d\u0938\u091f\u0930\u094d\u0928\u0932 \u0932\u0940 \u0907\u0902\u091f\u0930\u092a\u094d\u0930\u093f\u091f\u0930 \u092e\u093e\u0907\u0902\u0921\u0938\u0947\u091f \u092c\u093f\u092b\u094b\u0930 \u092f\u0942 \u0910\u0915\u094d\u091a\u0942\u0905\u0932\u0940 \u092c\u093f\u0915\u092e \u0930\u093f\u091c \u0907\u0928\u094d\u091f\u0930\u092a\u094d\u0930\u0947\u091f\u0930 \u0935\u093e\u0907\u091c \u092e\u093e\u0907\u0902\u0921\u0938\u0947\u091f \u092c\u093f\u092b\u094b\u0930 \u092f\u0942 \u090f\u0915\u094d\u091a\u0941\u0905\u0932\u0940 \u092c\u093f\u0915\u092e \u0935\u093e\u0907\u095b \u091c\u094b \u0905\u092c\u0948\u0928\u094d\u0921\u0928 \u0938\u0947 \u092f\u093e \u092b\u093f\u0930 \u0938\u0915\u094d\u0938\u0947\u0938 \u0924\u0941\u092e \u0905\u092a\u0928\u0947 \u092c\u093e\u0939\u0930 \u0926\u0947\u0916\u0928\u093e \u091a\u093e\u0939\u0924\u0947 \u0939\u094b? \u092a\u0939\u0932\u0947 \u0909\u0938\u0947 \u0905\u092a\u0928\u0947 \u0905\u0902\u0926\u0930 \u090f\u0915 \u092a\u0930\u092e\u093e\u0928\u0947\u0902\u091f \u091c\u0917\u0939 \u0926\u094b \u0928\u0902\u092c\u0930 \u091f\u0942 \u091f\u0947\u0915 \u0930\u093f\u0938\u094d\u092a\u0949\u0928\u094d\u0938\u093f\u092c\u093f\u0932\u093f\u091f\u0940 \u0917\u094b\u0932\u094d\u0938 \u0914\u0930 \u090f\u0915\u094d\u0938\u094d\u092a\u094d\u0930\u0947\u0936\u0928\u094d\u0938 \u0924\u094b \u0938\u092c\u0915\u0947 \u0939\u094b\u0924\u0947 \u0939\u0948\u0902 \u0932\u0947\u0915\u093f\u0928 \u0909\u0928 \u0917\u094b\u0932\u094d\u0938 \u092a\u0930 \u0910\u0915\u094d\u0936\u0928 \u0932\u0947\u0928\u093e \u0914\u0930 \u0905\u092a\u0928\u0940 \u0932\u093e\u0907\u092b \u0915\u094b \u0905\u092a\u0928\u0947 \u0917\u094b\u0932\u094d\u0938 \u0915\u0947 \u0905\u0930\u093e\u0909\u0902\u0921 \u0911\u0930\u094d\u0917\u0928\u093e\u0907\u091c \u0915\u0930\u0928\u093e \u092f\u0947 \u092c\u0924\u093e\u0924\u093e \u0939\u0948 \u0915\u093f \u0906\u092a \u0905\u092a\u0928\u0940 \u0932\u093e\u0907\u092b \u0915\u0940 \u0915\u093f\u0924\u0928\u0940 \u0930\u093f\u0938\u094d\u092a\u0949\u0928\u094d\u0938\u093f\u092c\u093f\u0932\u093f\u091f\u0940 \u0932\u0947 \u0930\u0939\u0947 \u0939\u094b\u0964 \u092a\u0930\u094d\u0938\u0928\u0932 \u0930\u093f\u0938\u094d\u092a\u0949\u0928\u094d\u0938\u093f\u092c\u093f\u0932\u093f\u091f\u0940 \u0932\u093e\u0907\u092b \u0915\u094b \u092c\u0939\u0941\u0924 \u0939\u0940 \u0906\u0938\u093e\u0928 \u0914\u0930 \u0938\u093f\u0902\u092a\u0932 \u092c\u0928\u093e \u0926\u0947\u0924\u0940 \u0939\u0948, \u0915\u094d\u092f\u094b\u0902\u0915\u093f \u090f\u0915 \u0930\u093f\u0938\u094d\u092a\u0949\u0928\u094d\u0938\u093f\u092c\u0932 \u0907\u0902\u0938\u093e\u0928 \u0915\u094b \u092a\u0924\u093e \u0939\u094b\u0924\u093e \u0939\u0948\u0964 \u0915\u093f \u0909\u0938\u0915\u0940 \u0932\u093e\u0907\u092b \u0909\u0938\u0915\u0947 \u0915\u0902\u091f\u094d\u0930\u094b\u0932 \u092e\u0947\u0902 \u0939\u0948 \u0914\u0930 \u0935\u094b \u091a\u093e\u0939\u0947 \u091c\u093f\u0924\u0928\u0940 \u092c\u093e\u0930 \u092d\u0940 \u0939\u093e\u0930\u0947 \u092f\u093e \u092b\u093f\u0930 \u0915\u093f\u0924\u0928\u0940 \u092d\u0940 \u092c\u095c\u0940 \u092e\u0941\u0936\u094d\u0915\u093f\u0932 \u092e\u0947\u0902 \u092b\u0902\u0938 \u091c\u093e\u090f, \u0935\u094b \u092b\u093f\u0930 \u092d\u0940 \u0915\u094b\u0908 \u0928\u093e \u0915\u094b\u0908 \u091c\u0941\u0917\u093e\u095c \u0932\u0917\u093e\u0915\u0930 \u091c\u0940\u0924 \u0938\u0915\u0924\u093e \u0939\u0948\u0964 \u0930\u093f\u0938\u094d\u092a\u0949\u0928\u094d\u0938\u093f\u092c\u093f\u0932\u093f\u091f\u0940 \u0915\u094b\u0908 \u092c\u094b\u091d \u0928\u0939\u0940\u0902 \u0939\u0948, \u092c\u0932\u094d\u0915\u093f \u092f\u0939 \u090f\u0915 \u0924\u0930\u0940\u0915\u093e \u0939\u0948 \u091c\u093f\u0938\u0938\u0947 \u0906\u092a \u0926\u0942\u0938\u0930\u094b\u0902 \u0915\u0947 \u090a\u092a\u0930 \u092c\u094b\u091d \u0928\u0939\u0940\u0902 \u092c\u0928\u0924\u0947 \u0939\u0948\u0902 \u0914\u0930 \u0905\u092a\u0928\u0940 \u091c\u093f\u0902\u0926\u0917\u0940 \u0915\u094b \u0915\u0932\u0947 \u0915\u0940 \u0924\u0930\u0939 \u0905\u092a\u0928\u093e \u092e\u0928\u091a\u093e\u0939\u093e \u0906\u0915\u093e\u0930 \u0926\u0947 \u092a\u093e\u0924\u0947 \u0939\u094b\u0964 \u0928\u0902\u092c\u0930 \u0925\u094d\u0930\u0940 \u092b\u093e\u0907\u0928\u094d\u0921 \u092e\u0940\u0928\u093f\u0902\u0917 \u0907\u0928 \u0932\u093e\u0907\u092b \u091c\u093f\u0902\u0926\u0917\u0940 \u092e\u0940\u0928\u093f\u0902\u0917 \u0915\u0947 \u092c\u093f\u0928\u093e \u0905\u0927\u0942\u0930\u0940 \u0914\u0930 \u092c\u0939\u0941\u0924 \u092e\u0941\u0936\u094d\u0915\u093f\u0932 \u0932\u0917\u0924\u0940 \u0939\u0948\u0964 \u092e\u0940\u0928\u093f\u0902\u0917 \u0935\u094b \u091a\u092e\u0915 \u0939\u094b\u0924\u0940 \u0939\u0948 \u091c\u094b \u0906\u092a\u0915\u094b \u091c\u093f\u0902\u0926\u0917\u0940 \u0915\u0947 \u0905\u0902\u0927\u0947\u0930\u0947 \u092e\u0947\u0902 \u0930\u093e\u0938\u094d\u0924\u093e \u0926\u093f\u0916\u093e\u0924\u0940 \u0939\u0948\u0964 \u0906\u092a\u0915\u094b \u092a\u0924\u093e \u0939\u094b\u0924\u093e \u0939\u0948 \u0915\u093f \u0920\u0940\u0915 \u0939\u0948, \u092e\u0948\u0902 \u0907\u0924\u0928\u0940 \u091c\u094d\u092f\u093e\u0926\u093e \u092e\u0941\u0936\u094d\u0915\u093f\u0932\u094b\u0902 \u0915\u094b \u091d\u0947\u0932 \u0930\u0939\u093e \u0939\u0942\u0901\u0964 \u0938\u092b\u0930 \u0915\u0930 \u0930\u0939\u093e \u0939\u0942\u0901 \u092c\u091f \u092f\u0947 \u0938\u092c \u090f\u0915 \u092c\u095c\u0947 \u0915\u0949\u0938\u094d\u091f \u0915\u0947 \u0932\u093f\u090f \u0939\u0948\u0964 \u0907\u0902\u0938\u093e\u0928 \u0915\u094b \u0915\u0937\u094d\u091f \u0909\u0924\u0928\u0947 \u0924\u0902\u0917 \u0928\u0939\u0940\u0902 \u0915\u0930\u0924\u0947 \u091c\u093f\u0924\u0928\u093e \u0909\u0928\u094d\u0939\u0947\u0902 \u0930\u093f\u092f\u0932\u093e\u0907\u091c\u0947\u0936\u0928 \u0924\u0902\u0917 \u0915\u0930\u0924\u0940 \u0939\u0948 \u0915\u093f \u0935\u094b \u092c\u093f\u0928\u093e \u0915\u093f\u0938\u0940 \u0935\u091c\u0939 \u0915\u0947 \u0907\u0928 \u0915\u0937\u094d\u091f\u094b\u0902 \u0915\u094b \u091d\u0947\u0932 \u0930\u0939\u0947 \u0939\u0948\u0902\u0964 \u0907\u0938\u0932\u093f\u090f \u0905\u092a\u0928\u0947 \u091c\u0940\u0928\u0947 \u0915\u0940 \u0935\u091c\u0939 \u0922\u0942\u0902\u0922\u094b\u0964 \u0906\u092a\u0915\u093e \u0932\u093e\u0907\u092b \u092a\u0930\u094d\u092a\u0938 \u0939\u0948 \u0905\u092a\u0928\u0940 \u0939\u093e\u0907\u090f\u0938\u094d\u091f \u092a\u094b\u091f\u0947\u0902\u0936\u093f\u0905\u0932 \u0924\u0915 \u092a\u0939\u0941\u0902\u091a\u0928\u093e, \u0932\u0947\u0915\u093f\u0928 \u0906\u092a \u092f\u0947 \u0915\u093e\u092e \u0915\u0948\u0938\u0947 \u0915\u0930\u094b\u0917\u0947? \u092f\u0939 \u095e\u093f\u0917\u0930 \u0906\u0909\u091f \u0915\u0930\u094b \u0928\u0902\u092c\u0930 \u095e\u094b\u0930 \u091f\u0947\u0915 \u092f\u0941\u0905\u0930 \u092b\u094d\u0930\u0947\u0928\u094d\u0921\u0938 \u090f\u0902\u0921 \u092b\u0948\u092e\u093f\u0932\u0940 \u091f\u0942 \u0926 \u092c\u0948\u091f\u0932 \u0935\u093f\u0926 \u092f\u0942 \u0906\u092a \u092e\u0947\u0902 \u0938\u0947 \u092c\u0939\u0941\u0924 \u0938\u0947 \u0932\u094b\u0917 \u0905\u092a\u0928\u0940 \u092b\u0948\u092e\u093f\u0932\u0940 \u0914\u0930 \u0907\u0935\u0928 \u0905\u092a\u0928\u0947 \u092b\u094d\u0930\u0947\u0902\u0921\u094d\u0938 \u0915\u0947 \u091c\u094d\u092f\u093e\u0926\u093e \u0915\u094d\u0932\u094b\u095b \u0928\u0939\u0940\u0902 \u0939\u0948, \u091c\u093f\u0938\u0915\u0940 \u0935\u091c\u0939 \u0938\u0947 \u0939\u0930 \u0907\u0902\u0938\u093e\u0928 \u0915\u0940 \u0936\u0915\u094d\u0924\u093f \u090f\u0915 \u0939\u0940 \u091c\u0917\u0939 \u092a\u0930 \u0915\u0949\u0928\u094d\u0938\u0928\u094d\u091f\u094d\u0930\u0947\u091f \u0939\u094b\u0928\u0947 \u0915\u0947 \u092c\u091c\u093e\u092f \u0906\u092a \u0932\u094b\u0917\u094b\u0902 \u0915\u0940 \u0936\u0915\u094d\u0924\u093f \u0907\u0927\u0930 \u0909\u0927\u0930 \u092c\u093f\u0916\u0930\u0940 \u0939\u0941\u0908 \u0939\u0948 \u0914\u0930 \u0906\u092a \u0915\u093f\u0938\u0940 \u092c\u095c\u0947 \u0917\u094b\u0932 \u092a\u0930 \u0915\u093e\u092e \u0928\u0939\u0940\u0902 \u0915\u0930 \u0938\u0915\u0924\u0947 \u0939\u0948\u0902\u0964 \u091f\u0940\u092e \u090f\u092b\u0930\u094d\u091f \u092e\u0947\u0902 \u0907\u0924\u0928\u0940 \u0924\u093e\u0915\u0924 \u0939\u094b\u0924\u0940 \u0939\u0948 \u0915\u093f \u0935\u0939 \u090f\u0915 \u092a\u0939\u093e\u095c \u0915\u094b \u092d\u0940 \u0939\u093f\u0932\u093e \u0926\u0947\u0964 \u091c\u093f\u0924\u0928\u093e \u092c\u095c\u093e \u0917\u094b\u0932 \u0939\u094b\u0917\u093e \u0909\u0924\u0928\u093e \u0939\u0940 \u0906\u092a\u0915\u094b \u0926\u0942\u0938\u0930\u094b\u0902 \u0915\u093e \u0938\u093e\u0925 \u091a\u093e\u0939\u093f\u090f \u0939\u094b\u0917\u093e, \u091c\u092c\u0915\u093f \u091b\u094b\u091f\u0947 \u0917\u094b\u095b \u0924\u094b \u0906\u092a \u0905\u0915\u0947\u0932\u0947 \u0939\u0940 \u0905\u091a\u0940\u0935 \u0915\u0930 \u0932\u094b\u0917\u0947\u0964 \u0932\u0947\u0915\u093f\u0928 \u091b\u094b\u091f\u0947 \u0917\u094b\u0932\u094d\u0938 \u0915\u0947 \u0938\u093e\u0925 \u092a\u094d\u0930\u0949\u092c\u094d\u0932\u092e \u092f\u0947 \u0939\u094b\u0924\u0940 \u0939\u0948 \u0915\u093f \u0935\u094b \u0906\u092a\u0915\u094b \u0939\u092e\u0947\u0936\u093e \u0939\u0940 \u092e\u0940\u0921\u093f\u092f\u093e \u0939\u094b\u0915\u0930 \u092c\u0928\u093e\u090f \u0930\u0916\u0924\u0947 \u0939\u0948\u0902\u0964 \u0906\u092a\u0915\u094b \u092d\u0940 \u0932\u093e\u0907\u092b \u0915\u093e \u092e\u095b\u093e \u0938\u093f\u0930\u094d\u092b \u0924\u092d\u0940 \u0906\u090f\u0917\u093e \u091c\u092c \u0906\u092a \u090f\u0915 \u0905\u0928\u0930\u093f\u092f\u0932\u093f\u0938\u094d\u091f\u093f\u0915 \u0914\u0930 \u092a\u093e\u0917\u0932\u094b\u0902 \u0935\u093e\u0932\u0940 \u091a\u0940\u095b \u0915\u093e \u092a\u0940\u091b\u093e \u0915\u0930\u094b\u0917\u0947\u0964 \u091c\u092c \u0924\u0915 \u0906\u092a \u0916\u0941\u0926 \u0915\u094b \u0905\u092a\u0928\u0940 \u0932\u093f\u092e\u093f\u091f\u094d\u0938 \u0915\u0947 \u092a\u093e\u0938 \u0938\u094d\u091f\u094d\u0930\u0947\u091a \u0928\u0939\u0940\u0902 \u0915\u0930\u094b\u0917\u0947, \u0924\u092c \u0924\u0915 \u0906\u092a\u0915\u094b \u092f\u0939 \u092b\u0940\u0932 \u0939\u0940 \u0928\u0939\u0940\u0902 \u0939\u094b\u0917\u093e\u0964 \u0915\u094d\u092f\u093e \u0906\u092a \u091c\u093f\u0928\u094d\u0926\u093e \u0939\u094b \u0914\u0930 \u0906\u092a \u0905\u092a\u0928\u0940 \u0938\u094b\u091a \u0938\u0947 \u091c\u094d\u092f\u093e\u0926\u093e \u092a\u0949\u0935\u0930\u092b\u0941\u0932 \u0939\u094b? \u0907\u0938\u0932\u093f\u090f \u0905\u092a\u0928\u0947 \u092b\u094d\u0930\u0947\u0902\u0921\u094d\u0938 \u0914\u0930 \u092b\u0948\u092e\u093f\u0932\u0940 \u0915\u094b\u0938\u092e\u0947 \u092a\u0947\u091c \u092a\u0930 \u0932\u093e\u0913 \u0909\u0928\u0938\u0947 \u0917\u094b\u0932\u094d\u0938, \u092a\u0930\u094d\u092a\u0938 \u0914\u0930 \u092e\u0940\u0928\u093f\u0902\u0917 \u0915\u0940 \u092c\u093e\u0924\u0947\u0902 \u0915\u0930\u094b\u0964 \u092a\u0939\u0932\u0947 \u0916\u0941\u0926 \u0915\u0940 \u0914\u0930 \u092b\u093f\u0930 \u0909\u0928\u0915\u0940 \u091f\u094d\u0930\u093e\u0902\u0938\u092b\u0949\u0930\u094d\u092e\u0947\u0936\u0928 \u092a\u0930 \u0927\u094d\u092f\u093e\u0928 \u0926\u094b \u0914\u0930 \u0938\u093e\u0925 \u092e\u0947\u0902 \u0917\u094d\u0930\u094b \u0915\u0930\u0928\u0947 \u0915\u093e \u092a\u094d\u0932\u093e\u0928 \u092c\u0928\u093e \u0939\u094b\u0964 \u0928\u0902\u092c\u0930 \u092b\u093e\u0907\u0935 \u092c\u093f\u0915\u092e \u092c\u093f\u0917\u0930 \u0926\u0947\u0928 \u092f\u0942 \u0911\u092b\u093f\u0930\u094d\u0938 \u0924\u0941\u092e\u094d\u0939\u0947 \u0932\u093e\u0907\u092b \u0915\u093e\u092b\u0940 \u092e\u0948\u0928\u0947\u091c\u0947\u092c\u0932 \u0914\u0930 \u0906\u0938\u093e\u0928 \u0932\u0917\u0928\u0947 \u0932\u0917\u0947 \u0917\u0940 \u091c\u092c \u0924\u0941\u092e \u092f\u0947 \u0915\u0930\u094b\u0917\u0947 \u0915\u093f \u0924\u0941\u092e \u0905\u092a\u0928\u0947 \u0921\u0930\u094b \u0938\u0947 \u091c\u094d\u092f\u093e\u0926\u093e \u092c\u095c\u0947 \u0939\u094b, \u0939\u0930 \u0921\u0930 \u092e\u094c\u0924 \u0938\u0947 \u0939\u0940 \u091c\u0941\u095c\u093e \u0939\u094b\u0924\u093e \u0939\u0948 \u0914\u0930 \u0921\u0930 \u0915\u094b \u0913\u0935\u0930 \u0915\u092e \u0915\u0930\u0928\u0947 \u0915\u093e \u092e\u0924\u0932\u092c \u0939\u094b\u0924\u093e \u0939\u0948 \u092e\u094c\u0924 \u0915\u094b \u0913\u0935\u0930 \u0915\u092e \u0915\u0930\u0928\u093e\u0964 \u0905\u092c \u0939\u092e \u0905\u092a\u0928\u0947 \u0936\u0930\u0940\u0930 \u0915\u094b \u0924\u094b \u092e\u0930\u0928\u0947 \u0938\u0947 \u0930\u094b\u0915 \u0928\u0939\u0940\u0902 \u0938\u0915\u0924\u0947 \u0932\u0947\u0915\u093f\u0928 \u0939\u092e \u092f\u0947 \u091c\u0930\u0942\u0930 \u0938\u092e\u091d \u0938\u0915\u0924\u0947 \u0939\u0948\u0902 \u0915\u093f \u091c\u0940\u0938 \u0907\u0928\u094d\u091f\u0947\u0932\u093f\u091c\u0947\u0928\u094d\u0938 \u0928\u0947 \u0939\u092e\u0947\u0902 \u092c\u0928\u093e\u092f\u093e \u0939\u0948\u0964 \u0935\u094b \u0915\u092d\u0940 \u092e\u0930 \u092f\u093e \u092b\u093f\u0930 \u092a\u0948\u0926\u093e \u0939\u094b \u0939\u0940 \u0928\u0939\u0940\u0902 \u0938\u0915\u0924\u0940\u0964 \u0935\u094b \u0905\u0928\u0928\u094d\u0924 \u0914\u0930 \u0936\u093e\u0936\u094d\u0935\u0924 \u0939\u0948\u0964 \u0905\u0917\u0930 \u0906\u092a \u0907\u0938 \u092c\u093e\u0930\u0947 \u092e\u0947\u0902 \u0905\u092a\u0928\u0940 \u0908\u0917\u094b \u0928\u093e\u092e \u0915\u0947 \u0924\u093f\u0928\u0915\u0947 \u0915\u093e \u0938\u0939\u093e\u0930\u093e \u0932\u0947\u0928\u0947 \u0915\u0940 \u0915\u094b\u0936\u093f\u0936 \u0915\u0930\u094b\u0917\u0947 \u0924\u094b \u0906\u092a\u0915\u093e \u0921\u0942\u092c\u0928\u093e \u0924\u092f \u0939\u0948\u0964 \u0907\u0938\u0932\u093f\u090f \u0915\u092d\u0940 \u092d\u0940 \u0916\u0941\u0926 \u0915\u094b \u0905\u092a\u0928\u0947 \u0936\u0930\u0940\u0930 \u092f\u093e \u092b\u093f\u0930 \u092e\u0928\u0938\u0947 \u0906\u0907\u0921\u0947\u0902\u091f\u093f\u092b\u093e\u0907 \u092e\u0924 \u0915\u0930\u094b\u0964 \u0916\u0941\u0926 \u0915\u0940 \u0906\u0907\u0921\u0947\u0902\u091f\u093f\u091f\u0940 \u092e\u0947\u0902 \u091b\u0941\u092a\u0947 \u091d\u0942\u0920 \u0915\u094b \u0926\u0947\u0916\u094b\u0964 \u091c\u093f\u0924\u0928\u093e \u0906\u092a \u0905\u092a\u0928\u0940 \u0908\u0917\u094b \u0915\u094b \u0938\u0940\u0930\u093f\u092f\u0938\u0932\u0940 \u0932\u094b\u0917\u0947, \u0909\u0924\u0928\u093e \u0939\u0940 \u0906\u092a \u0921\u0930, \u0932\u093e\u0932\u091a, \u092e\u094b\u0939, \u092a\u0940\u095c\u093e \u0914\u0930 \u0917\u0941\u0938\u094d\u0938\u0947 \u091c\u0948\u0938\u0940 \u091a\u0940\u091c\u094b\u0902 \u092e\u0947\u0902 \u0930\u0939\u094b\u0917\u0947 \u0913\u0935\u0930 \u0915\u092e\u093f\u0902\u0917 \u092b\u093f\u0932\u094d\u0938 \u0907\u0938\u0915\u093e \u0938\u094d\u092a\u093f\u0930\u093f\u091a\u0941\u0905\u0932 \u091f\u093e\u0938\u094d\u0915 \u0928\u0949\u091f \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u0906\u092a\u0915\u093e \u0926\u093f\u092e\u093e\u0917 \u0914\u0930 \u0936\u0930\u0940\u0930 \u092a\u094d\u0930\u094b\u0917\u094d\u0930\u093e\u092e\u094d\u0921 \u0939\u0948, \u0921\u0930\u094b \u0938\u0947 \u0926\u0942\u0930 \u092d\u093e\u0917\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u0939\u0948 \u0915\u094d\u092f\u094b\u0902\u0915\u093f \u092e\u094c\u0924 \u0906\u092a\u0915\u0947 \u0936\u0930\u0940\u0930 \u0914\u0930 \u092e\u0928 \u0938\u0947 \u092c\u095d\u0940 \u0939\u0948\u0964 \u0905\u092a\u0928\u0947 \u0905\u0902\u0926\u0930 \u0909\u0938 \u0924\u093e\u0915\u0924 \u0915\u094b \u0922\u0942\u0902\u0922\u094b \u091c\u094b \u092e\u094c\u0924 \u0938\u0947 \u092d\u0940 \u092c\u095c\u0940 \u0939\u0948 \u0914\u0930 \u0924\u092c \u0924\u0941\u092e \u092c\u094b\u0932\u094b\u0917\u0947 \u0915\u093f \u091c\u093f\u0902\u0926\u0917\u0940 \u0909\u0924\u0928\u0940 \u092e\u0941\u0936\u094d\u0915\u093f\u0932 \u0914\u0930 \u0926\u0941\u0916\u0926\u093e\u092f\u0940 \u0928\u0939\u0940\u0902 \u0939\u0948 \u091c\u093f\u0924\u0928\u093e \u0939\u092e \u0909\u0938\u0947 \u092c\u0928\u093e \u0926\u0947\u0924\u0947 \u0939\u0948\u0902\u0964 \u0938\u094d\u091f\u0949\u092a \u092e\u0947\u0915\u093f\u0902\u0917 \u092f\u094b\u0930 \u0932\u093e\u0907\u092b \u0905\u0928\u0928\u0947\u0938\u0947\u0938\u0930\u0940 \u0932\u0940 \u0939\u093e\u0930\u094d\u0921 \u0924\u0941\u092e\u094d\u0939\u093e\u0930\u0940 \u091c\u093f\u0902\u0926\u0917\u0940 \u0915\u092d\u0940 \u0928\u093e \u0905\u0902\u0924 \u0939\u094b\u0928\u0947 \u0935\u093e\u0932\u0940 \u0932\u0940\u0932\u093e \u0915\u093e \u0939\u093f\u0938\u094d\u0938\u093e \u0939\u0948\u0964 \u0907\u0938\u0915\u093e \u0930\u0938 \u0932\u094b \u0914\u0930 \u091c\u093f\u0924\u0928\u093e \u0939\u094b \u0938\u0915\u0947 \u0909\u0924\u0928\u093e \u0907\u0938 \u0926\u0941\u0928\u093f\u092f\u093e \u092e\u0947\u0902 \u092c\u0926\u0932\u093e\u0935 \u0932\u093e\u0913\u0964 \u092b\u0941\u091f \u092a\u0930 \u0914\u0930 \u0916\u0941\u0926 \u0915\u0940 \u092a\u094b\u091f\u0947\u0902\u0936\u093f\u0905\u0932 \u092a\u0930 \u0915\u093e\u092e \u0915\u0930\u0915\u0947 \u0932\u093e\u0907\u092b \u0907\u095b \u0930\u093f\u092f\u0932\u0940 \u0939\u093e\u0930\u094d\u0921 \u092b\u0949\u0930 \u092f\u0942 \u092c\u091f \u0907\u091f \u0907\u0938 \u0935\u0947\u0907\u091f\u094d \u092b\u0949\u0930 \u0926\u094b\u095b \u0939\u0942\u0901 \u0914\u0930 \u0932\u093f\u0935\u093f\u0902\u0917 \u0905\u092a \u091f\u0941 \u0926\u0947\u0930 \u092a\u094b\u091f\u0947\u0902\u0936\u093f\u0905\u0932\u0964",
"target": "\u0906\u0938\u093e\u0928 \u0939\u094b \u092f\u093e \u092e\u0941\u0936\u094d\u0915\u093f\u0932, \u092f\u0947 \u0936\u092c\u094d\u0926 \u091c\u094d\u092f\u093e\u0926\u093e\u0924\u0930 \u0932\u094b\u0917\u094b\u0902 \u0915\u0947 \u0932\u093f\u090f \u0938\u093e\u092a\u0947\u0915\u094d\u0937 \u0939\u094b\u0924\u0947 \u0939\u0948\u0902\u0964 \u091c\u094b \u0906\u092a\u0915\u0947 \u0932\u093f\u090f \u0906\u0938\u093e\u0928 \u0939\u0948 \u0935\u0939 \u0915\u093f\u0938\u0940 \u0914\u0930 \u0915\u0947 \u0932\u093f\u090f \u092e\u0941\u0936\u094d\u0915\u093f\u0932 \u0939\u094b \u0938\u0915\u0924\u093e \u0939\u0948 \u0914\u0930 \u0907\u0938\u0915\u0947 \u0935\u093f\u092a\u0930\u0940\u0924\u0964 \u0915\u093f\u0938\u0940 \u0915\u093e\u0930\u094d\u092f \u0915\u0940 \u0915\u0920\u093f\u0928\u093e\u0908 \u0909\u0938\u0947 \u0915\u0930\u0928\u0947 \u0935\u093e\u0932\u0947 \u0935\u094d\u092f\u0915\u094d\u0924\u093f \u0915\u0940 \u0915\u094d\u0937\u092e\u0924\u093e \u092a\u0930 \u0928\u093f\u0930\u094d\u092d\u0930 \u0915\u0930\u0924\u0940 \u0939\u0948\u0964 \u090f\u0915 \u0916\u093e\u0938 \u092e\u093e\u0928\u0938\u093f\u0915\u0924\u093e \u0930\u0916\u0928\u0947 \u0935\u093e\u0932\u094b\u0902 \u0915\u0947 \u0932\u093f\u090f \u091c\u0940\u0935\u0928 \u0906\u0938\u093e\u0928 \u0939\u094b\u0924\u093e \u0939\u0948 \u0914\u0930 \u0907\u0938 \u092a\u0949\u0921\u0915\u093e\u0938\u094d\u091f \u0938\u0947\u0917\u092e\u0947\u0902\u091f \u092e\u0947\u0902 \u0939\u092e \u0910\u0938\u0940 \u0939\u0940 \u092e\u093e\u0928\u0938\u093f\u0915\u0924\u093e \u0915\u0947 \u092c\u093e\u0930\u0947 \u092e\u0947\u0902 \u092c\u093e\u0924 \u0915\u0930\u0947\u0902\u0917\u0947\u0964"
}
]
```
### Dataset Fields
The dataset has the following fields (also called "features"):
```json
{
"text": "Value(dtype='string', id=None)",
"target": "Value(dtype='string', id=None)"
}
```
### Dataset Splits
This dataset is split into a train and validation split. The split sizes are as follow:
| Split name | Num samples |
| ------------ | ------------------- |
| train | 451 |
| valid | 113 |
|
viditsorg/autotrain-data-summarization-xlsum
|
[
"task_categories:summarization",
"region:us"
] |
2023-10-24T09:03:59+00:00
|
{"task_categories": ["summarization"]}
|
2023-10-24T09:06:15+00:00
|
[] |
[] |
TAGS
#task_categories-summarization #region-us
|
AutoTrain Dataset for project: summarization-xlsum
==================================================
Dataset Description
-------------------
This dataset has been automatically processed by AutoTrain for project summarization-xlsum.
### Languages
The BCP-47 code for the dataset's language is unk.
Dataset Structure
-----------------
### Data Instances
A sample from this dataset looks as follows:
### Dataset Fields
The dataset has the following fields (also called "features"):
### Dataset Splits
This dataset is split into a train and validation split. The split sizes are as follow:
|
[
"### Languages\n\n\nThe BCP-47 code for the dataset's language is unk.\n\n\nDataset Structure\n-----------------",
"### Data Instances\n\n\nA sample from this dataset looks as follows:",
"### Dataset Fields\n\n\nThe dataset has the following fields (also called \"features\"):",
"### Dataset Splits\n\n\nThis dataset is split into a train and validation split. The split sizes are as follow:"
] |
[
"TAGS\n#task_categories-summarization #region-us \n",
"### Languages\n\n\nThe BCP-47 code for the dataset's language is unk.\n\n\nDataset Structure\n-----------------",
"### Data Instances\n\n\nA sample from this dataset looks as follows:",
"### Dataset Fields\n\n\nThe dataset has the following fields (also called \"features\"):",
"### Dataset Splits\n\n\nThis dataset is split into a train and validation split. The split sizes are as follow:"
] |
[
16,
27,
17,
23,
27
] |
[
"passage: TAGS\n#task_categories-summarization #region-us \n### Languages\n\n\nThe BCP-47 code for the dataset's language is unk.\n\n\nDataset Structure\n-----------------### Data Instances\n\n\nA sample from this dataset looks as follows:### Dataset Fields\n\n\nThe dataset has the following fields (also called \"features\"):### Dataset Splits\n\n\nThis dataset is split into a train and validation split. The split sizes are as follow:"
] |
233d58166b89d40f085c4c957a0598c2972615d6
|
*TR Wikipedia DDO QA Dataseti*
Bu dataset, Türkçe Wikipedia'dan çeşitli konularda soru-cevap verilerini içermektedir. Dataset, aşağıdaki kategorilere ayrılmıştır:
Edebiyat: edebiyat.json dosyasında bulabilirsiniz. Dosya boyutu: 9.31 MB.
Siyaset: siyaset.json dosyasında bulabilirsiniz. Dosya boyutu: 6.15 MB.
Televizyon: televizyon.json dosyasında bulabilirsiniz. Dosya boyutu: 6.67 MB.
Kullanım
Bu dataseti, doğal dil işleme uygulamalarınızda, özellikle soru-cevap sistemlerinizde kullanabilirsiniz. Veriler, JSON formatında sağlanmaktadır ve her bir kategori için ayrı dosyalarda bulunmaktadır.
|
aerdincdal/tr_wikipedia_ddo_qa
|
[
"license:apache-2.0",
"region:us"
] |
2023-10-24T09:08:27+00:00
|
{"license": "apache-2.0"}
|
2023-10-30T10:29:01+00:00
|
[] |
[] |
TAGS
#license-apache-2.0 #region-us
|
*TR Wikipedia DDO QA Dataseti*
Bu dataset, Türkçe Wikipedia'dan çeşitli konularda soru-cevap verilerini içermektedir. Dataset, aşağıdaki kategorilere ayrılmıştır:
Edebiyat: URL dosyasında bulabilirsiniz. Dosya boyutu: 9.31 MB.
Siyaset: URL dosyasında bulabilirsiniz. Dosya boyutu: 6.15 MB.
Televizyon: URL dosyasında bulabilirsiniz. Dosya boyutu: 6.67 MB.
Kullanım
Bu dataseti, doğal dil işleme uygulamalarınızda, özellikle soru-cevap sistemlerinizde kullanabilirsiniz. Veriler, JSON formatında sağlanmaktadır ve her bir kategori için ayrı dosyalarda bulunmaktadır.
|
[] |
[
"TAGS\n#license-apache-2.0 #region-us \n"
] |
[
14
] |
[
"passage: TAGS\n#license-apache-2.0 #region-us \n"
] |
cbca5380fecd2b0fc46364370e3e1d05bddd183a
|
# Dataset Card for "order_dataset"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
awaisakhtar/order_dataset
|
[
"task_categories:question-answering",
"task_categories:conversational",
"size_categories:1K<n<10K",
"language:en",
"order",
"region:us"
] |
2023-10-24T09:46:17+00:00
|
{"language": ["en"], "size_categories": ["1K<n<10K"], "task_categories": ["question-answering", "conversational"], "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}], "dataset_info": {"features": [{"name": "System_Prompt", "dtype": "string"}, {"name": "Instruction", "dtype": "string"}, {"name": "Context", "dtype": "string"}, {"name": "Menu", "dtype": "string"}, {"name": "Conversation_History", "dtype": "string"}, {"name": "Response", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 25430540, "num_examples": 5140}], "download_size": 1277262, "dataset_size": 25430540}, "tags": ["order"]}
|
2023-11-06T13:37:44+00:00
|
[] |
[
"en"
] |
TAGS
#task_categories-question-answering #task_categories-conversational #size_categories-1K<n<10K #language-English #order #region-us
|
# Dataset Card for "order_dataset"
More Information needed
|
[
"# Dataset Card for \"order_dataset\"\n\nMore Information needed"
] |
[
"TAGS\n#task_categories-question-answering #task_categories-conversational #size_categories-1K<n<10K #language-English #order #region-us \n",
"# Dataset Card for \"order_dataset\"\n\nMore Information needed"
] |
[
46,
14
] |
[
"passage: TAGS\n#task_categories-question-answering #task_categories-conversational #size_categories-1K<n<10K #language-English #order #region-us \n# Dataset Card for \"order_dataset\"\n\nMore Information needed"
] |
23efc67aab600e58c14167b6a5a27ec90e2e025f
|
# Spontanous speech QA
This dataset contains QA pairs from the spontaneous speech subsection of the Danish Gigaword.
The dataset is created from the [DDSC dataset](DDSC/partial-danish-gigaword-no-twitter) and
filtered to only include QA pairs where the question is less than 20 tokens and the answer is
at least 4 tokens long.
To find out more about the creation see the accompanying script.
|
KennethEnevoldsen/spontanous-speech-qa
|
[
"task_categories:question-answering",
"annotations_creators:found",
"size_categories:n<1K",
"source_datasets:DDSC/partial-danish-gigaword-no-twitter",
"language:da",
"conversational",
"region:us"
] |
2023-10-24T10:17:48+00:00
|
{"annotations_creators": ["found"], "language": ["da"], "size_categories": ["n<1K"], "source_datasets": ["DDSC/partial-danish-gigaword-no-twitter"], "task_categories": ["question-answering"], "pretty_name": "Spontanous speech QA", "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "test", "path": "data/test-*"}]}], "dataset_info": {"features": [{"name": "question", "dtype": "string"}, {"name": "answer", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 44345.110764430574, "num_examples": 512}, {"name": "test", "num_bytes": 11172.889235569422, "num_examples": 129}], "download_size": 37996, "dataset_size": 55518}, "tags": ["conversational"]}
|
2023-10-24T10:23:18+00:00
|
[] |
[
"da"
] |
TAGS
#task_categories-question-answering #annotations_creators-found #size_categories-n<1K #source_datasets-DDSC/partial-danish-gigaword-no-twitter #language-Danish #conversational #region-us
|
# Spontanous speech QA
This dataset contains QA pairs from the spontaneous speech subsection of the Danish Gigaword.
The dataset is created from the DDSC dataset and
filtered to only include QA pairs where the question is less than 20 tokens and the answer is
at least 4 tokens long.
To find out more about the creation see the accompanying script.
|
[
"# Spontanous speech QA\n\nThis dataset contains QA pairs from the spontaneous speech subsection of the Danish Gigaword. \nThe dataset is created from the DDSC dataset and \nfiltered to only include QA pairs where the question is less than 20 tokens and the answer is\nat least 4 tokens long.\n\nTo find out more about the creation see the accompanying script."
] |
[
"TAGS\n#task_categories-question-answering #annotations_creators-found #size_categories-n<1K #source_datasets-DDSC/partial-danish-gigaword-no-twitter #language-Danish #conversational #region-us \n",
"# Spontanous speech QA\n\nThis dataset contains QA pairs from the spontaneous speech subsection of the Danish Gigaword. \nThe dataset is created from the DDSC dataset and \nfiltered to only include QA pairs where the question is less than 20 tokens and the answer is\nat least 4 tokens long.\n\nTo find out more about the creation see the accompanying script."
] |
[
70,
87
] |
[
"passage: TAGS\n#task_categories-question-answering #annotations_creators-found #size_categories-n<1K #source_datasets-DDSC/partial-danish-gigaword-no-twitter #language-Danish #conversational #region-us \n# Spontanous speech QA\n\nThis dataset contains QA pairs from the spontaneous speech subsection of the Danish Gigaword. \nThe dataset is created from the DDSC dataset and \nfiltered to only include QA pairs where the question is less than 20 tokens and the answer is\nat least 4 tokens long.\n\nTo find out more about the creation see the accompanying script."
] |
1b81962ba993a0c5a9ce484c4c250f334be231af
|
Toxicity contail three types of data. 1. from realtoxicty prompt .2 response from gpt3.5 generation as prompt 3. same as 2 but it comes from gpt4
|
lzy337/attack_data_hf
|
[
"region:us"
] |
2023-10-24T10:35:51+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": ["toxicity/toxic.jsonl.gpt3.n=25.out1.split.annotated.jsonl.filtered_train.jsonl"]}, {"split": "test", "path": ["toxicity/toxic.jsonl.gpt3.n=25.out1.split.annotated.jsonl.filtered_test.jsonl"]}, {"split": "dev", "path": ["toxicity/toxic.jsonl.gpt3.n=25.out1.split.annotated.jsonl.filtered_dev.jsonl"]}]}]}
|
2023-10-24T11:24:10+00:00
|
[] |
[] |
TAGS
#region-us
|
Toxicity contail three types of data. 1. from realtoxicty prompt .2 response from gpt3.5 generation as prompt 3. same as 2 but it comes from gpt4
|
[] |
[
"TAGS\n#region-us \n"
] |
[
6
] |
[
"passage: TAGS\n#region-us \n"
] |
50b68b0b9e4b3abe024bf793ab024c4add53ffd2
|
# Multimodal Concept Conjunction 250
In our paper [MultiFusion: Fusing Pre-Trained Models for
Multi-Lingual, Multi-Modal Image Generation](https://arxiv.org/abs/2305.15296) we propose the MCC-250 benchmark to evaluate generative image composition capablities for multimodal inputs.
MCC-250 is built on a subset of [CC-500](https://arxiv.org/abs/2212.05032) which contains 500 text-only prompts of the pattern "a red apple and a yellow banana", textually
describing two objects with respective attributes.
With MCC-250, we provide a set of reference images for each object and attribute combination, enabling multimodal applications.
## Attribution
All images where source from these four stock imagery providers:
- [Pixabay](https://pixabay.com/)
- [Unsplash](https://unsplash.com/)
- [Pexels](https://www.pexels.com/)
- [Freepik](https://www.freepik.com/)
|
AIML-TUDA/MCC-250
|
[
"task_categories:text-to-image",
"task_categories:image-to-image",
"size_categories:n<1K",
"language:en",
"license:mit",
"arxiv:2305.15296",
"arxiv:2212.05032",
"region:us"
] |
2023-10-24T10:42:03+00:00
|
{"language": ["en"], "license": "mit", "size_categories": ["n<1K"], "task_categories": ["text-to-image", "image-to-image"]}
|
2023-10-25T08:31:42+00:00
|
[
"2305.15296",
"2212.05032"
] |
[
"en"
] |
TAGS
#task_categories-text-to-image #task_categories-image-to-image #size_categories-n<1K #language-English #license-mit #arxiv-2305.15296 #arxiv-2212.05032 #region-us
|
# Multimodal Concept Conjunction 250
In our paper MultiFusion: Fusing Pre-Trained Models for
Multi-Lingual, Multi-Modal Image Generation we propose the MCC-250 benchmark to evaluate generative image composition capablities for multimodal inputs.
MCC-250 is built on a subset of CC-500 which contains 500 text-only prompts of the pattern "a red apple and a yellow banana", textually
describing two objects with respective attributes.
With MCC-250, we provide a set of reference images for each object and attribute combination, enabling multimodal applications.
## Attribution
All images where source from these four stock imagery providers:
- Pixabay
- Unsplash
- Pexels
- Freepik
|
[
"# Multimodal Concept Conjunction 250\n\nIn our paper MultiFusion: Fusing Pre-Trained Models for\nMulti-Lingual, Multi-Modal Image Generation we propose the MCC-250 benchmark to evaluate generative image composition capablities for multimodal inputs. \nMCC-250 is built on a subset of CC-500 which contains 500 text-only prompts of the pattern \"a red apple and a yellow banana\", textually\ndescribing two objects with respective attributes.\n\nWith MCC-250, we provide a set of reference images for each object and attribute combination, enabling multimodal applications.",
"## Attribution\n\nAll images where source from these four stock imagery providers:\n- Pixabay\n- Unsplash\n- Pexels\n- Freepik"
] |
[
"TAGS\n#task_categories-text-to-image #task_categories-image-to-image #size_categories-n<1K #language-English #license-mit #arxiv-2305.15296 #arxiv-2212.05032 #region-us \n",
"# Multimodal Concept Conjunction 250\n\nIn our paper MultiFusion: Fusing Pre-Trained Models for\nMulti-Lingual, Multi-Modal Image Generation we propose the MCC-250 benchmark to evaluate generative image composition capablities for multimodal inputs. \nMCC-250 is built on a subset of CC-500 which contains 500 text-only prompts of the pattern \"a red apple and a yellow banana\", textually\ndescribing two objects with respective attributes.\n\nWith MCC-250, we provide a set of reference images for each object and attribute combination, enabling multimodal applications.",
"## Attribution\n\nAll images where source from these four stock imagery providers:\n- Pixabay\n- Unsplash\n- Pexels\n- Freepik"
] |
[
67,
138,
29
] |
[
"passage: TAGS\n#task_categories-text-to-image #task_categories-image-to-image #size_categories-n<1K #language-English #license-mit #arxiv-2305.15296 #arxiv-2212.05032 #region-us \n# Multimodal Concept Conjunction 250\n\nIn our paper MultiFusion: Fusing Pre-Trained Models for\nMulti-Lingual, Multi-Modal Image Generation we propose the MCC-250 benchmark to evaluate generative image composition capablities for multimodal inputs. \nMCC-250 is built on a subset of CC-500 which contains 500 text-only prompts of the pattern \"a red apple and a yellow banana\", textually\ndescribing two objects with respective attributes.\n\nWith MCC-250, we provide a set of reference images for each object and attribute combination, enabling multimodal applications.## Attribution\n\nAll images where source from these four stock imagery providers:\n- Pixabay\n- Unsplash\n- Pexels\n- Freepik"
] |
5b826a49f0f7bd1a6ef038db551458266bd09eb5
|
# HSR <br>
HSR is a compilation of 5605 reddit posts scraped from the following subreddits:
- r/ScaryStories
- r/LetsNotMeet
- r/TwoSentenceHorror
- r/freehorrorstories
- r/TrueScaryStories
- r/NoSleep
- r/Ruleshorror
# HSR Credits
If you are using HSR, you must cite us for your project. This dataset can be used for Translation, Generative or Conversational models. <br>
Here are a few ideas that you can use HSR for: <br>
- Title-to-story
- Text Generation
- Spooky chats
|
intone/horror_stories_reddit
|
[
"task_categories:text-generation",
"task_categories:translation",
"size_categories:1K<n<10K",
"language:en",
"region:us"
] |
2023-10-24T11:00:59+00:00
|
{"language": ["en"], "size_categories": ["1K<n<10K"], "task_categories": ["text-generation", "translation"]}
|
2023-10-24T15:16:22+00:00
|
[] |
[
"en"
] |
TAGS
#task_categories-text-generation #task_categories-translation #size_categories-1K<n<10K #language-English #region-us
|
# HSR <br>
HSR is a compilation of 5605 reddit posts scraped from the following subreddits:
- r/ScaryStories
- r/LetsNotMeet
- r/TwoSentenceHorror
- r/freehorrorstories
- r/TrueScaryStories
- r/NoSleep
- r/Ruleshorror
# HSR Credits
If you are using HSR, you must cite us for your project. This dataset can be used for Translation, Generative or Conversational models. <br>
Here are a few ideas that you can use HSR for: <br>
- Title-to-story
- Text Generation
- Spooky chats
|
[
"# HSR <br>\nHSR is a compilation of 5605 reddit posts scraped from the following subreddits:\n- r/ScaryStories\n- r/LetsNotMeet\n- r/TwoSentenceHorror\n- r/freehorrorstories\n- r/TrueScaryStories\n- r/NoSleep\n- r/Ruleshorror",
"# HSR Credits\nIf you are using HSR, you must cite us for your project. This dataset can be used for Translation, Generative or Conversational models. <br>\nHere are a few ideas that you can use HSR for: <br>\n- Title-to-story\n- Text Generation\n- Spooky chats"
] |
[
"TAGS\n#task_categories-text-generation #task_categories-translation #size_categories-1K<n<10K #language-English #region-us \n",
"# HSR <br>\nHSR is a compilation of 5605 reddit posts scraped from the following subreddits:\n- r/ScaryStories\n- r/LetsNotMeet\n- r/TwoSentenceHorror\n- r/freehorrorstories\n- r/TrueScaryStories\n- r/NoSleep\n- r/Ruleshorror",
"# HSR Credits\nIf you are using HSR, you must cite us for your project. This dataset can be used for Translation, Generative or Conversational models. <br>\nHere are a few ideas that you can use HSR for: <br>\n- Title-to-story\n- Text Generation\n- Spooky chats"
] |
[
42,
85,
71
] |
[
"passage: TAGS\n#task_categories-text-generation #task_categories-translation #size_categories-1K<n<10K #language-English #region-us \n# HSR <br>\nHSR is a compilation of 5605 reddit posts scraped from the following subreddits:\n- r/ScaryStories\n- r/LetsNotMeet\n- r/TwoSentenceHorror\n- r/freehorrorstories\n- r/TrueScaryStories\n- r/NoSleep\n- r/Ruleshorror# HSR Credits\nIf you are using HSR, you must cite us for your project. This dataset can be used for Translation, Generative or Conversational models. <br>\nHere are a few ideas that you can use HSR for: <br>\n- Title-to-story\n- Text Generation\n- Spooky chats"
] |
4a8472949103d65164a4dd08bff41bf1953a976e
|
# Dataset Card for "models-metadata-snapshot"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
davanstrien/models-metadata-snapshot
|
[
"region:us"
] |
2023-10-24T11:24:45+00:00
|
{"dataset_info": {"features": [{"name": "id", "dtype": "string"}, {"name": "date_checked", "dtype": "date32"}, {"name": "created", "dtype": "timestamp[us]"}, {"name": "last_repo_commit", "dtype": "timestamp[us, tz=UTC]"}, {"name": "tags", "sequence": "string"}, {"name": "pipeline_tag", "dtype": "string"}, {"name": "author", "dtype": "string"}, {"name": "likes", "dtype": "int64"}, {"name": "downloads", "dtype": "int64"}, {"name": "library_name", "dtype": "string"}, {"name": "license", "dtype": "string"}, {"name": "language", "sequence": "null"}, {"name": "datasets", "sequence": "string"}, {"name": "number_authors", "dtype": "int64"}, {"name": "readme_length", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 529260, "num_examples": 1998}], "download_size": 101185, "dataset_size": 529260}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-27T11:44:02+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "models-metadata-snapshot"
More Information needed
|
[
"# Dataset Card for \"models-metadata-snapshot\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"models-metadata-snapshot\"\n\nMore Information needed"
] |
[
6,
19
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"models-metadata-snapshot\"\n\nMore Information needed"
] |
e921d309f32531c0cfdbd38ea4183e8a6943ded2
|
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_examples: 6
|
TruongDiem/test4
|
[
"region:us"
] |
2023-10-24T11:34:07+00:00
|
{}
|
2023-10-24T11:37:10+00:00
|
[] |
[] |
TAGS
#region-us
|
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_examples: 6
|
[] |
[
"TAGS\n#region-us \n"
] |
[
6
] |
[
"passage: TAGS\n#region-us \n"
] |
55bc5c78a8ce2647f64ab1b477ff9f6ba241797e
|
# Dataset Card for "guikal"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
Arsture/guikal
|
[
"region:us"
] |
2023-10-24T11:34:37+00:00
|
{"dataset_info": {"features": [{"name": "Line", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 196000, "num_examples": 7378}], "download_size": 136705, "dataset_size": 196000}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T11:34:39+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "guikal"
More Information needed
|
[
"# Dataset Card for \"guikal\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"guikal\"\n\nMore Information needed"
] |
[
6,
12
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"guikal\"\n\nMore Information needed"
] |
8d0b3db91d471245ada50aa49a6ed167db47bfdb
|
## Schema description:
The manually curated dataset of open-closed monomers is included here as `benchmarking_monomeric_open_closed_conformers.csv`.
Column descriptions:
## Schema description:
The manually curated dataset of open-closed monomers is included here as `benchmarking_monomeric_open_closed_conformers.csv`.
Column descriptions:
- **`UNP_ACC`** | UniProt accession code
- **`UNP_START`** | Start of UniProt sequence for given PDBe entries
- **`UNP_END`** | End of UniProt sequence for given PDBe entries
- **`PDBe_ID`** | Protein Data Bank code
- **`CHAIN_ID`** | Author declared chain ID (`char`)
- **`label_asym_id`** | Programmatically assigned chain ID (`char`)
- **`CONFORMER_ID`** | Unique code for PDBe entries with distinct conformation, corresponding to a given UniProt accession
- **`CONFORMER_DESCR`** | Short description of conformation, based on depositor's assessment of the protein/conformation
- **`LIT_CONFIRMED`** | True/false value based on whether a publication (scientific literature) was available for manually curating clusters. NB: Clusters with 0 in this field should be used with caution.
- **`ALT_CONFORMER_ID`** | Where the publication for a structure is currently outstanding, an executive decision on the conformation classification is made. Where the literature is not explicit on the features of a given conformation, the second most suitable `CONFORMER_ID` is provided in this column. Blank cells have no other likely conformation assignmnt and are therefore the same as in `CONFORMER_ID`.
- **`ALT_CONFORMER_DESCR`** | Description for conformation in alternative conformation ID.
## Curation process
As of 09 Mar 2022, a manually curated dataset of monomeric protein conformations has been collated, containing 'open'-'closed' pairs as well as intermediary states defined by the authors of the entry.
1. The PDBe was queried, through its Oracle DB, to find PDBe entries with 100 % sequence identity for a UniProt segment in both 'open' and 'closed' conformations, as stated in the entry's `TITLE` field. The query used:
```
select b.accession, b.unp_start, b.unp_end, a.id, a.title, d.id, d.title
from entry a, unp_entity b, unp_entity c, entry d, pdb_assembly e
where a.title like ‘%open%’ and d.title like ‘%close%’
and a.id = b.entry_id and d.id = c.entry_id and a.id != d.id
and b.accession = c.accession
and b.unp_start = c.unp_start
and b.unp_end = c.unp_end
and a.id = e.entry_id
and e.type = ‘homo’
and e.name = ‘monomer’
```
was written by Dr Sameer Velankar.
2. These results were cleaned to remove entries with 'open' or 'close' substrings in their `TITLE` fields that did not refer to conformation. The 'open' substring often appeared in ligand names in the entries' `TITLE` field, such as in *dichlorido(1,3-dimethylbenzimidaz ol-2-ylidene)(eta5-pentamethylcycl**open**tadienyl)rhodium(III)* and 'close' in terms like *dis**close**s*.
3. All remaining entries were then manually curated by investigating each PDBe entry's corresponding publication, where available.
1. Additional PDBe entries submitted by the authors, which were missed in the original search due to a lack of 'open' or 'close' substrings in their `TITLE` field but stated as fitting one of the states in the publication, were added.
2. For some UniProt accessions, intermediary conformations were reported by the authors and these were noted in the dataset under the `CONFORMER_DESCR` column.
3. Entries deposited in monomeric form but solved as a multimeric complex were also removed.
4. PDBe entries, now clustered by author-stated conformation, were cross-referenced against the PDBe-KB's existing clustering algorithm (available on the [Aggregate Views of Proteins](https://www.ebi.ac.uk/pdbe/pdbe-kb/protein) page) to assess current conformer clustering success. These results are currently awaiting publication.
### Curation process outline
<img src="http://ftp.ebi.ac.uk/pub/databases/pdbe-kb/benchmarking/distinct-monomer-conformers/work_progress_flowdiagram_200pc.png" alt="Curation flow diagram">
### Dataset summary
<img src="http://ftp.ebi.ac.uk/pub/databases/pdbe-kb/benchmarking/distinct-monomer-conformers/summary_data_visualisation.png" alt="Benchmark dataset summary graphs">
|
PDBEurope/protein_chain_conformational_states
|
[
"task_categories:feature-extraction",
"size_categories:10K<n<100K",
"language:en",
"license:cc-by-4.0",
"Structural biology",
"Bioinformatics",
"Machine learning",
"Conformation",
"Conformational state",
"Monomeric",
"Training data",
"Benchmark",
"Manually curated",
"region:us"
] |
2023-10-24T11:36:32+00:00
|
{"language": ["en"], "license": "cc-by-4.0", "size_categories": ["10K<n<100K"], "task_categories": ["feature-extraction"], "pretty_name": "Curated dataset of protein chain conformational states", "tags": ["Structural biology", "Bioinformatics", "Machine learning", "Conformation", "Conformational state", "Monomeric", "Training data", "Benchmark", "Manually curated"]}
|
2023-10-24T11:43:32+00:00
|
[] |
[
"en"
] |
TAGS
#task_categories-feature-extraction #size_categories-10K<n<100K #language-English #license-cc-by-4.0 #Structural biology #Bioinformatics #Machine learning #Conformation #Conformational state #Monomeric #Training data #Benchmark #Manually curated #region-us
|
## Schema description:
The manually curated dataset of open-closed monomers is included here as 'benchmarking_monomeric_open_closed_conformers.csv'.
Column descriptions:
## Schema description:
The manually curated dataset of open-closed monomers is included here as 'benchmarking_monomeric_open_closed_conformers.csv'.
Column descriptions:
- 'UNP_ACC' | UniProt accession code
- 'UNP_START' | Start of UniProt sequence for given PDBe entries
- 'UNP_END' | End of UniProt sequence for given PDBe entries
- 'PDBe_ID' | Protein Data Bank code
- 'CHAIN_ID' | Author declared chain ID ('char')
- 'label_asym_id' | Programmatically assigned chain ID ('char')
- 'CONFORMER_ID' | Unique code for PDBe entries with distinct conformation, corresponding to a given UniProt accession
- 'CONFORMER_DESCR' | Short description of conformation, based on depositor's assessment of the protein/conformation
- 'LIT_CONFIRMED' | True/false value based on whether a publication (scientific literature) was available for manually curating clusters. NB: Clusters with 0 in this field should be used with caution.
- 'ALT_CONFORMER_ID' | Where the publication for a structure is currently outstanding, an executive decision on the conformation classification is made. Where the literature is not explicit on the features of a given conformation, the second most suitable 'CONFORMER_ID' is provided in this column. Blank cells have no other likely conformation assignmnt and are therefore the same as in 'CONFORMER_ID'.
- 'ALT_CONFORMER_DESCR' | Description for conformation in alternative conformation ID.
## Curation process
As of 09 Mar 2022, a manually curated dataset of monomeric protein conformations has been collated, containing 'open'-'closed' pairs as well as intermediary states defined by the authors of the entry.
1. The PDBe was queried, through its Oracle DB, to find PDBe entries with 100 % sequence identity for a UniProt segment in both 'open' and 'closed' conformations, as stated in the entry's 'TITLE' field. The query used:
was written by Dr Sameer Velankar.
2. These results were cleaned to remove entries with 'open' or 'close' substrings in their 'TITLE' fields that did not refer to conformation. The 'open' substring often appeared in ligand names in the entries' 'TITLE' field, such as in *dichlorido(1,3-dimethylbenzimidaz ol-2-ylidene)(eta5-pentamethylcyclopentadienyl)rhodium(III)* and 'close' in terms like *discloses*.
3. All remaining entries were then manually curated by investigating each PDBe entry's corresponding publication, where available.
1. Additional PDBe entries submitted by the authors, which were missed in the original search due to a lack of 'open' or 'close' substrings in their 'TITLE' field but stated as fitting one of the states in the publication, were added.
2. For some UniProt accessions, intermediary conformations were reported by the authors and these were noted in the dataset under the 'CONFORMER_DESCR' column.
3. Entries deposited in monomeric form but solved as a multimeric complex were also removed.
4. PDBe entries, now clustered by author-stated conformation, were cross-referenced against the PDBe-KB's existing clustering algorithm (available on the Aggregate Views of Proteins page) to assess current conformer clustering success. These results are currently awaiting publication.
### Curation process outline
<img src="URL alt="Curation flow diagram">
### Dataset summary
<img src="URL alt="Benchmark dataset summary graphs">
|
[
"## Schema description:\nThe manually curated dataset of open-closed monomers is included here as 'benchmarking_monomeric_open_closed_conformers.csv'. \n\nColumn descriptions:",
"## Schema description:\nThe manually curated dataset of open-closed monomers is included here as 'benchmarking_monomeric_open_closed_conformers.csv'. \n\nColumn descriptions:\n\n- 'UNP_ACC' | UniProt accession code\n- 'UNP_START' | Start of UniProt sequence for given PDBe entries\n- 'UNP_END' | End of UniProt sequence for given PDBe entries\n- 'PDBe_ID' | Protein Data Bank code\n- 'CHAIN_ID' | Author declared chain ID ('char')\n- 'label_asym_id' | Programmatically assigned chain ID ('char')\n- 'CONFORMER_ID' | Unique code for PDBe entries with distinct conformation, corresponding to a given UniProt accession\n- 'CONFORMER_DESCR' | Short description of conformation, based on depositor's assessment of the protein/conformation\n- 'LIT_CONFIRMED' | True/false value based on whether a publication (scientific literature) was available for manually curating clusters. NB: Clusters with 0 in this field should be used with caution. \n- 'ALT_CONFORMER_ID' | Where the publication for a structure is currently outstanding, an executive decision on the conformation classification is made. Where the literature is not explicit on the features of a given conformation, the second most suitable 'CONFORMER_ID' is provided in this column. Blank cells have no other likely conformation assignmnt and are therefore the same as in 'CONFORMER_ID'. \n- 'ALT_CONFORMER_DESCR' | Description for conformation in alternative conformation ID.",
"## Curation process\nAs of 09 Mar 2022, a manually curated dataset of monomeric protein conformations has been collated, containing 'open'-'closed' pairs as well as intermediary states defined by the authors of the entry. \n\n1. The PDBe was queried, through its Oracle DB, to find PDBe entries with 100 % sequence identity for a UniProt segment in both 'open' and 'closed' conformations, as stated in the entry's 'TITLE' field. The query used: \n\nwas written by Dr Sameer Velankar. \n\n2. These results were cleaned to remove entries with 'open' or 'close' substrings in their 'TITLE' fields that did not refer to conformation. The 'open' substring often appeared in ligand names in the entries' 'TITLE' field, such as in *dichlorido(1,3-dimethylbenzimidaz ol-2-ylidene)(eta5-pentamethylcyclopentadienyl)rhodium(III)* and 'close' in terms like *discloses*.\n\n3. All remaining entries were then manually curated by investigating each PDBe entry's corresponding publication, where available.\n 1. Additional PDBe entries submitted by the authors, which were missed in the original search due to a lack of 'open' or 'close' substrings in their 'TITLE' field but stated as fitting one of the states in the publication, were added.\n 2. For some UniProt accessions, intermediary conformations were reported by the authors and these were noted in the dataset under the 'CONFORMER_DESCR' column. \n 3. Entries deposited in monomeric form but solved as a multimeric complex were also removed. \n\n4. PDBe entries, now clustered by author-stated conformation, were cross-referenced against the PDBe-KB's existing clustering algorithm (available on the Aggregate Views of Proteins page) to assess current conformer clustering success. These results are currently awaiting publication.",
"### Curation process outline\n<img src=\"URL alt=\"Curation flow diagram\">",
"### Dataset summary\n<img src=\"URL alt=\"Benchmark dataset summary graphs\">"
] |
[
"TAGS\n#task_categories-feature-extraction #size_categories-10K<n<100K #language-English #license-cc-by-4.0 #Structural biology #Bioinformatics #Machine learning #Conformation #Conformational state #Monomeric #Training data #Benchmark #Manually curated #region-us \n",
"## Schema description:\nThe manually curated dataset of open-closed monomers is included here as 'benchmarking_monomeric_open_closed_conformers.csv'. \n\nColumn descriptions:",
"## Schema description:\nThe manually curated dataset of open-closed monomers is included here as 'benchmarking_monomeric_open_closed_conformers.csv'. \n\nColumn descriptions:\n\n- 'UNP_ACC' | UniProt accession code\n- 'UNP_START' | Start of UniProt sequence for given PDBe entries\n- 'UNP_END' | End of UniProt sequence for given PDBe entries\n- 'PDBe_ID' | Protein Data Bank code\n- 'CHAIN_ID' | Author declared chain ID ('char')\n- 'label_asym_id' | Programmatically assigned chain ID ('char')\n- 'CONFORMER_ID' | Unique code for PDBe entries with distinct conformation, corresponding to a given UniProt accession\n- 'CONFORMER_DESCR' | Short description of conformation, based on depositor's assessment of the protein/conformation\n- 'LIT_CONFIRMED' | True/false value based on whether a publication (scientific literature) was available for manually curating clusters. NB: Clusters with 0 in this field should be used with caution. \n- 'ALT_CONFORMER_ID' | Where the publication for a structure is currently outstanding, an executive decision on the conformation classification is made. Where the literature is not explicit on the features of a given conformation, the second most suitable 'CONFORMER_ID' is provided in this column. Blank cells have no other likely conformation assignmnt and are therefore the same as in 'CONFORMER_ID'. \n- 'ALT_CONFORMER_DESCR' | Description for conformation in alternative conformation ID.",
"## Curation process\nAs of 09 Mar 2022, a manually curated dataset of monomeric protein conformations has been collated, containing 'open'-'closed' pairs as well as intermediary states defined by the authors of the entry. \n\n1. The PDBe was queried, through its Oracle DB, to find PDBe entries with 100 % sequence identity for a UniProt segment in both 'open' and 'closed' conformations, as stated in the entry's 'TITLE' field. The query used: \n\nwas written by Dr Sameer Velankar. \n\n2. These results were cleaned to remove entries with 'open' or 'close' substrings in their 'TITLE' fields that did not refer to conformation. The 'open' substring often appeared in ligand names in the entries' 'TITLE' field, such as in *dichlorido(1,3-dimethylbenzimidaz ol-2-ylidene)(eta5-pentamethylcyclopentadienyl)rhodium(III)* and 'close' in terms like *discloses*.\n\n3. All remaining entries were then manually curated by investigating each PDBe entry's corresponding publication, where available.\n 1. Additional PDBe entries submitted by the authors, which were missed in the original search due to a lack of 'open' or 'close' substrings in their 'TITLE' field but stated as fitting one of the states in the publication, were added.\n 2. For some UniProt accessions, intermediary conformations were reported by the authors and these were noted in the dataset under the 'CONFORMER_DESCR' column. \n 3. Entries deposited in monomeric form but solved as a multimeric complex were also removed. \n\n4. PDBe entries, now clustered by author-stated conformation, were cross-referenced against the PDBe-KB's existing clustering algorithm (available on the Aggregate Views of Proteins page) to assess current conformer clustering success. These results are currently awaiting publication.",
"### Curation process outline\n<img src=\"URL alt=\"Curation flow diagram\">",
"### Dataset summary\n<img src=\"URL alt=\"Benchmark dataset summary graphs\">"
] |
[
85,
50,
405,
468,
21,
24
] |
[
"passage: TAGS\n#task_categories-feature-extraction #size_categories-10K<n<100K #language-English #license-cc-by-4.0 #Structural biology #Bioinformatics #Machine learning #Conformation #Conformational state #Monomeric #Training data #Benchmark #Manually curated #region-us \n## Schema description:\nThe manually curated dataset of open-closed monomers is included here as 'benchmarking_monomeric_open_closed_conformers.csv'. \n\nColumn descriptions:",
"passage: ## Schema description:\nThe manually curated dataset of open-closed monomers is included here as 'benchmarking_monomeric_open_closed_conformers.csv'. \n\nColumn descriptions:\n\n- 'UNP_ACC' | UniProt accession code\n- 'UNP_START' | Start of UniProt sequence for given PDBe entries\n- 'UNP_END' | End of UniProt sequence for given PDBe entries\n- 'PDBe_ID' | Protein Data Bank code\n- 'CHAIN_ID' | Author declared chain ID ('char')\n- 'label_asym_id' | Programmatically assigned chain ID ('char')\n- 'CONFORMER_ID' | Unique code for PDBe entries with distinct conformation, corresponding to a given UniProt accession\n- 'CONFORMER_DESCR' | Short description of conformation, based on depositor's assessment of the protein/conformation\n- 'LIT_CONFIRMED' | True/false value based on whether a publication (scientific literature) was available for manually curating clusters. NB: Clusters with 0 in this field should be used with caution. \n- 'ALT_CONFORMER_ID' | Where the publication for a structure is currently outstanding, an executive decision on the conformation classification is made. Where the literature is not explicit on the features of a given conformation, the second most suitable 'CONFORMER_ID' is provided in this column. Blank cells have no other likely conformation assignmnt and are therefore the same as in 'CONFORMER_ID'. \n- 'ALT_CONFORMER_DESCR' | Description for conformation in alternative conformation ID.## Curation process\nAs of 09 Mar 2022, a manually curated dataset of monomeric protein conformations has been collated, containing 'open'-'closed' pairs as well as intermediary states defined by the authors of the entry. \n\n1. The PDBe was queried, through its Oracle DB, to find PDBe entries with 100 % sequence identity for a UniProt segment in both 'open' and 'closed' conformations, as stated in the entry's 'TITLE' field. The query used: \n\nwas written by Dr Sameer Velankar. \n\n2. These results were cleaned to remove entries with 'open' or 'close' substrings in their 'TITLE' fields that did not refer to conformation. The 'open' substring often appeared in ligand names in the entries' 'TITLE' field, such as in *dichlorido(1,3-dimethylbenzimidaz ol-2-ylidene)(eta5-pentamethylcyclopentadienyl)rhodium(III)* and 'close' in terms like *discloses*.\n\n3. All remaining entries were then manually curated by investigating each PDBe entry's corresponding publication, where available.\n 1. Additional PDBe entries submitted by the authors, which were missed in the original search due to a lack of 'open' or 'close' substrings in their 'TITLE' field but stated as fitting one of the states in the publication, were added.\n 2. For some UniProt accessions, intermediary conformations were reported by the authors and these were noted in the dataset under the 'CONFORMER_DESCR' column. \n 3. Entries deposited in monomeric form but solved as a multimeric complex were also removed. \n\n4. PDBe entries, now clustered by author-stated conformation, were cross-referenced against the PDBe-KB's existing clustering algorithm (available on the Aggregate Views of Proteins page) to assess current conformer clustering success. These results are currently awaiting publication.### Curation process outline\n<img src=\"URL alt=\"Curation flow diagram\">"
] |
4d15a850369342eec0b77b267d1a59ace6750621
|
# Dataset Card for "arxiv_cs_papers"
This dataset contains the subset of ArXiv papers with the "cs.LG" tag to indicate the paper is about Machine Learning.
The core dataset is filtered from the full ArXiv dataset hosted on Kaggle: https://www.kaggle.com/datasets/Cornell-University/arxiv. The original dataset contains roughly 2 million papers. This dataset contains roughly 100,000 papers following the category filtering.
The dataset is maintained with requests to the ArXiv API.
The ArXiv dataset contains features:
<ul>
<li> title </li>
<li> abstract </li>
<li> authors </li>
<li> published </li>
<li> url </li>
<li> pdf_url </li>
<li> arxiv_id </li>
</ul>
|
ashish-chouhan/arxiv_cs_papers
|
[
"region:us"
] |
2023-10-24T11:39:22+00:00
|
{"dataset_info": {"features": [{"name": "title", "dtype": "string"}, {"name": "abstract", "dtype": "string"}, {"name": "authors", "sequence": "string"}, {"name": "published", "dtype": "string"}, {"name": "url", "dtype": "string"}, {"name": "pdf_url", "dtype": "string"}, {"name": "arxiv_id", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 7726383, "num_examples": 5000}], "download_size": 4366827, "dataset_size": 7726383}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T12:31:08+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "arxiv_cs_papers"
This dataset contains the subset of ArXiv papers with the "cs.LG" tag to indicate the paper is about Machine Learning.
The core dataset is filtered from the full ArXiv dataset hosted on Kaggle: URL The original dataset contains roughly 2 million papers. This dataset contains roughly 100,000 papers following the category filtering.
The dataset is maintained with requests to the ArXiv API.
The ArXiv dataset contains features:
<ul>
<li> title </li>
<li> abstract </li>
<li> authors </li>
<li> published </li>
<li> url </li>
<li> pdf_url </li>
<li> arxiv_id </li>
</ul>
|
[
"# Dataset Card for \"arxiv_cs_papers\"\n\nThis dataset contains the subset of ArXiv papers with the \"cs.LG\" tag to indicate the paper is about Machine Learning.\n\nThe core dataset is filtered from the full ArXiv dataset hosted on Kaggle: URL The original dataset contains roughly 2 million papers. This dataset contains roughly 100,000 papers following the category filtering.\n\nThe dataset is maintained with requests to the ArXiv API.\n\nThe ArXiv dataset contains features:\n<ul>\n<li> title </li>\n<li> abstract </li>\n<li> authors </li>\n<li> published </li>\n<li> url </li>\n<li> pdf_url </li>\n<li> arxiv_id </li>\n</ul>"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"arxiv_cs_papers\"\n\nThis dataset contains the subset of ArXiv papers with the \"cs.LG\" tag to indicate the paper is about Machine Learning.\n\nThe core dataset is filtered from the full ArXiv dataset hosted on Kaggle: URL The original dataset contains roughly 2 million papers. This dataset contains roughly 100,000 papers following the category filtering.\n\nThe dataset is maintained with requests to the ArXiv API.\n\nThe ArXiv dataset contains features:\n<ul>\n<li> title </li>\n<li> abstract </li>\n<li> authors </li>\n<li> published </li>\n<li> url </li>\n<li> pdf_url </li>\n<li> arxiv_id </li>\n</ul>"
] |
[
6,
194
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"arxiv_cs_papers\"\n\nThis dataset contains the subset of ArXiv papers with the \"cs.LG\" tag to indicate the paper is about Machine Learning.\n\nThe core dataset is filtered from the full ArXiv dataset hosted on Kaggle: URL The original dataset contains roughly 2 million papers. This dataset contains roughly 100,000 papers following the category filtering.\n\nThe dataset is maintained with requests to the ArXiv API.\n\nThe ArXiv dataset contains features:\n<ul>\n<li> title </li>\n<li> abstract </li>\n<li> authors </li>\n<li> published </li>\n<li> url </li>\n<li> pdf_url </li>\n<li> arxiv_id </li>\n</ul>"
] |
f0401dfab5a8eee1fd45e4e7e3ff147dadd36431
|
# Dataset Card for "recipes"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
Eitanli/recipes
|
[
"region:us"
] |
2023-10-24T11:40:15+00:00
|
{"dataset_info": {"features": [{"name": "recipe", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 105767040, "num_examples": 74465}], "download_size": 53711472, "dataset_size": 105767040}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T11:40:20+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "recipes"
More Information needed
|
[
"# Dataset Card for \"recipes\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"recipes\"\n\nMore Information needed"
] |
[
6,
13
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"recipes\"\n\nMore Information needed"
] |
977208a05ebef0edc01b86ae141b2ac8d3d45e6e
|
# Dataset Card for "llama-2-clinc-oos-test"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
Luciya/llama-2-clinc-oos-test
|
[
"region:us"
] |
2023-10-24T11:52:24+00:00
|
{"dataset_info": {"features": [{"name": "text", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 12858700, "num_examples": 5492}], "download_size": 1217377, "dataset_size": 12858700}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T11:52:25+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "llama-2-clinc-oos-test"
More Information needed
|
[
"# Dataset Card for \"llama-2-clinc-oos-test\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"llama-2-clinc-oos-test\"\n\nMore Information needed"
] |
[
6,
21
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"llama-2-clinc-oos-test\"\n\nMore Information needed"
] |
aee90d12cc7b2aa7dc38734f3a1f1254e2d4ab33
|
# Dataset Card for "llama-2-clinc-oos-train"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
Luciya/llama-2-clinc-oos-train
|
[
"region:us"
] |
2023-10-24T11:52:47+00:00
|
{"dataset_info": {"features": [{"name": "text", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 6593005, "num_examples": 15100}], "download_size": 977625, "dataset_size": 6593005}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T11:52:48+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "llama-2-clinc-oos-train"
More Information needed
|
[
"# Dataset Card for \"llama-2-clinc-oos-train\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"llama-2-clinc-oos-train\"\n\nMore Information needed"
] |
[
6,
22
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"llama-2-clinc-oos-train\"\n\nMore Information needed"
] |
fea7626ce734e8c8ccb7a25fa2ab694be1a9b23c
|
# AutoTrain Dataset for project: mbart-finetune-hindi
## Dataset Description
This dataset has been automatically processed by AutoTrain for project mbart-finetune-hindi.
### Languages
The BCP-47 code for the dataset's language is unk.
## Dataset Structure
### Data Instances
A sample from this dataset looks as follows:
```json
[
{
"text": "\u092e\u0928 \u0915\u0940 \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u092e\u0947\u0902 \u092e\u094c\u091c\u0942\u0926 \u0905\u0902\u0927\u0947\u0930\u093e \u092f\u093e \u0924\u094b \u0939\u092e\u0947\u0902 \u0916\u0941\u0926 \u0930\u094c\u0936\u0928\u0940 \u0915\u093e \u0938\u094d\u0930\u094b\u0924 \u092c\u0928\u0928\u093e \u0938\u0940\u0916\u093e \u0938\u0915\u0924\u093e \u0939\u0948 \u092f\u093e \u092b\u093f\u0930 \u0935\u0939 \u0939\u092e\u093e\u0930\u0940 \u092c\u091a\u094d\u091a\u0940 \u0915\u0940 \u091a\u092e\u0915 \u0915\u094b \u092d\u0940 \u0916\u0924\u094d\u092e \u0915\u0930 \u0938\u0915\u0924\u093e \u0939\u0948\u0964 \u0915\u0941\u091b \u0910\u0938\u093e \u0939\u0940 \u0939\u0941\u0906 \u0925\u093e \u091c\u0930\u094d\u092e\u0928\u0940 \u092b\u093f\u0932\u0949\u0938\u092b\u0930 \u092b\u094d\u0930\u0947\u0921\u094d\u0930\u093f\u0915 \u0928\u0940\u091a\u0947 \u0915\u0947 \u0938\u093e\u0925\u0964 \u0907\u0902\u0938\u093e\u0928 \u092a\u0943\u0925\u094d\u0935\u0940 \u0915\u0947 \u0939\u0930 \u091b\u094b\u0930 \u0924\u0915 \u092f\u093e\u0924\u094d\u0930\u093e \u0915\u0930 \u091a\u0941\u0915\u093e \u0939\u0948, \u0938\u092e\u0941\u0926\u094d\u0930 \u0915\u0940 \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u092e\u0947\u0902 \u0921\u0941\u092c\u0915\u0940 \u0932\u0917\u093e \u091a\u0941\u0915\u093e \u0939\u0948 \u0914\u0930 \u0916\u0941\u0926 \u0915\u094b \u0905\u0902\u0924\u0930\u093f\u0915\u094d\u0937 \u0915\u093e \u0938\u092b\u0930 \u0915\u0930\u093e \u091a\u0941\u0915\u093e \u0939\u0948\u0964 \u0932\u0947\u0915\u093f\u0928 \u0939\u092e\u093e\u0930\u093e \u092e\u0928 \u0905\u092d\u0940 \u092d\u0940 \u090f\u0915 \u0910\u0938\u0940 \u091a\u0940\u095b \u0939\u0948 \u091c\u0939\u093e\u0901 \u091c\u093e\u0928\u0947 \u0914\u0930 \u091c\u0948\u0938\u0947 \u090f\u0915\u094d\u0938\u094d\u092a\u094d\u0932\u094b\u0930 \u0915\u0930\u0928\u0947 \u0915\u0940 \u0939\u093f\u092e\u094d\u092e\u0924 \u0938\u093f\u0930\u094d\u092b \u0915\u0908 \u0917\u093f\u0928\u0947 \u091a\u0941\u0928\u0947 \u0932\u094b\u0917\u094b\u0902 \u092e\u0947\u0902 \u0939\u0940 \u0939\u0948\u0964 \u092a\u093e\u0917\u0932 \u0939\u094b \u091c\u093e\u0928\u0947 \u0914\u0930 \u092c\u0939\u0941\u0924 \u0907\u0928 10 \u0938\u094d\u092a\u0947\u0928 \u0915\u0947 \u0921\u0930 \u0938\u0947 \u091c\u094d\u092f\u093e\u0926\u093e\u0924\u0930 \u0939\u092e \u0932\u094b\u0917 \u0905\u092a\u0928\u0947 \u092e\u0928 \u0915\u0940 \u0938\u0924\u0939 \u092a\u0930 \u0939\u0940 \u091c\u0940\u0924\u0947 \u0939\u0948\u0902 \u0914\u0930 \u092c\u093f\u0928\u093e \u0916\u0941\u0926 \u0915\u0947 \u0938\u091a \u0915\u094b \u091c\u093e\u0928\u0947 \u0939\u0940 \u092e\u0930 \u092d\u0940 \u091c\u093e\u0924\u0947 \u0939\u0948\u0902\u0964 \u0907\u0938\u0915\u0947 \u0938\u093e\u0925 \u0939\u0940 \u092e\u0928 \u090f\u0915 \u0910\u0938\u0940 \u091c\u0917\u0939 \u0939\u0948 \u091c\u093f\u0938\u092e\u0947\u0902 \u0928\u0947\u0935\u093f\u0917\u0947\u091f \u0915\u0930\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u0939\u092e\u093e\u0930\u0947 \u092a\u093e\u0938 \u0915\u094b\u0908 \u092e\u0948\u092a, \u0915\u0902\u092a\u093e\u0938 \u092f\u093e \u0917\u093e\u0907\u0921 \u092d\u0940 \u0928\u0939\u0940\u0902 \u0939\u0948\u0964 \u0939\u092e\u0947\u0936\u093e \u091c\u0917\u0939 \u092a\u0930 \u0905\u0915\u0947\u0932\u0947 \u091c\u093e\u0928\u093e \u0939\u094b\u0924\u093e \u0939\u0948 \u0914\u0930 \u0916\u0941\u0926 \u0939\u0940 \u0909\u0938\u0915\u0940 \u091a\u0941\u0928\u094c\u0924\u093f\u092f\u094b\u0902 \u0915\u093e \u0938\u093e\u092e\u0928\u093e \u0915\u0930\u0928\u093e \u092a\u095c\u0924\u093e \u0939\u0948\u0964 \u0928\u0940\u091a\u093e \u090f\u0915 \u0910\u0938\u0947 \u0938\u093e\u0939\u0938\u0940 \u0910\u0921\u0935\u0947\u0902\u091a\u0930 \u0925\u0947 \u091c\u094b \u092e\u0928 \u0915\u0940 \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u092e\u0947\u0902 \u0909\u0924\u0930\u0928\u0947 \u0938\u0947 \u091c\u0941\u095c\u0947 \u091c\u094b\u0916\u093f\u092e \u0915\u094b \u0909\u0920\u093e\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u0924\u0948\u092f\u093e\u0930 \u0925\u0947\u0964 \u0935\u094b \u0905\u092a\u0928\u0940 \u092c\u0941\u0915 \u0926 \u0921\u0949\u0928 \u0911\u092b \u0921\u0947 \u092e\u0947\u0902 \u0932\u093f\u0916\u0924\u0947 \u0939\u0948\u0902 \u0915\u093f \u092e\u0948\u0902\u0928\u0947 \u090f\u0915 \u0910\u0938\u093e \u0915\u0926\u092e \u0909\u0920\u093e \u0932\u093f\u092f\u093e \u0939\u0948 \u091c\u094b \u0939\u0930 \u0915\u093f\u0938\u0940 \u0915\u094b \u0928\u0939\u0940\u0902 \u0909\u0920\u093e\u0928\u093e \u091a\u093e\u0939\u093f\u090f\u0964 \u092e\u0948\u0902 \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u092e\u0947\u0902 \u0909\u0924\u0930 \u0917\u092f\u093e \u0939\u0942\u0901\u0964 \u092e\u0948\u0902\u0928\u0947 \u092c\u0941\u0928\u093f\u092f\u093e\u0926 \u0915\u094b \u0939\u0940 \u0916\u094b\u0926\u0928\u093e \u0936\u0941\u0930\u0942 \u0915\u0930 \u0926\u093f\u092f\u093e \u0939\u0948\u0964 \u0928\u0940\u091a\u0947 \u0915\u0940 \u0907\u0928\u0930 \u090f\u0915\u094d\u0938\u092a\u094d\u0932\u094b\u0930\u0947\u0936\u0928 \u0915\u0940 \u0935\u091c\u0939 \u0938\u0947 \u0939\u092e\u0947\u0902 \u0909\u0928\u0915\u0940 \u0915\u0908 \u0905\u0928\u094b\u0916\u0940 \u0930\u093e\u0907\u091f\u093f\u0902\u0917\u094d\u0938 \u092e\u093f\u0932\u0940 \u0939\u0948\u0964 \u0932\u0947\u0915\u093f\u0928 \u091c\u092c \u0935\u094b 45 \u0938\u093e\u0932 \u0915\u0947 \u0925\u0947 \u0924\u092c \u0909\u0928\u0915\u0940 \u092f\u0939\u0940 \u0907\u0928\u0930 \u090f\u0915\u094d\u0938\u092a\u094d\u0932\u094b\u0930\u0947\u0936\u0928 \u092a\u093e\u0917\u0932\u092a\u0928 \u092a\u0930 \u091c\u093e\u0915\u0930 \u0930\u0941\u0915\u0940 \u0914\u0930 \u092f\u0939 \u091a\u0940\u095b \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u0907\u0938 \u0938\u092e\u092f \u0938\u0947 8 \u0938\u093e\u0932 \u092a\u0939\u0932\u0947 \u0939\u0940 \u092a\u094d\u0930\u0940 \u0911\u0921\u093f\u091f \u0915\u0930 \u0926\u0940 \u0925\u0940\u0964 \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u0905\u092a\u0928\u0947 \u090f\u0915 \u0932\u0947\u091f\u0930 \u092e\u0947\u0902 \u0932\u093f\u0916\u093e \u0915\u093f \u0915\u0908 \u092c\u093e\u0930 \u092e\u0947\u0930\u0947 \u092e\u0928 \u092e\u0947\u0902 \u090f\u0915 \u091a\u0947\u0924\u093e\u0935\u0928\u0940 \u0906\u0924\u0940 \u0939\u0948 \u091c\u094b \u092c\u094b\u0932\u0924\u0940 \u0939\u0948 \u0915\u093f \u092e\u0948\u0902 \u090f\u0915 \u092c\u0939\u0941\u0924 \u0939\u0940 \u0916\u0924\u0930\u0928\u093e\u0915 \u091c\u093f\u0902\u0926\u0917\u0940 \u091c\u0940 \u0930\u0939\u093e \u0939\u0942\u0901 \u0915\u094d\u092f\u094b\u0902\u0915\u093f \u092e\u0948\u0902 \u0909\u0928 \u092e\u0936\u0940\u0928\u094b\u0902 \u092e\u0947\u0902 \u0938\u0947 \u0939\u0942\u0901 \u091c\u094b \u092b\u091f \u0938\u0915\u0924\u0940 \u0939\u0948\u0964 \u0905\u092a\u0928\u0947 \u092a\u093e\u0917\u0932\u092a\u0928 \u0938\u0947 1 \u0938\u093e\u0932 \u092a\u0939\u0932\u0947 \u092f\u093e\u0928\u0940 1888 \u092e\u0947\u0902 \u0928\u0940\u091a\u0947 \u0905\u092a\u0928\u0947 \u0906\u092a\u0915\u094b \u092d\u0917\u0935\u093e\u0928 \u0914\u0930 \u092e\u0939\u093e\u0928 \u0930\u093e\u091c\u093e\u0913\u0902 \u0915\u0947 \u0928\u093e\u092e \u0938\u0947 \u090f\u0921\u094d\u0930\u0947\u0938 \u0915\u0930\u0928\u0947 \u0932\u0917\u0947 \u0925\u0947\u0964 \u0909\u0928\u0915\u0947 \u0906\u0938\u092a\u093e\u0938 \u0930\u0939\u0928\u0947 \u0935\u093e\u0932\u0947 \u0932\u094b\u0917 \u092c\u0924\u093e\u0924\u0947 \u0939\u0948\u0902 \u0915\u093f \u0935\u094b \u0905\u092a\u0928\u0947 \u0915\u092e\u0930\u0947 \u0938\u0947 \u0915\u0908 \u0926\u093f\u0928\u094b\u0902 \u0924\u0915 \u092c\u093e\u0939\u0930 \u0928\u0939\u0940\u0902 \u0928\u093f\u0915\u0932\u0924\u0947 \u0925\u0947, \u0932\u0947\u0915\u093f\u0928 \u091c\u092c \u092c\u093e\u0939\u0930 \u0906\u0924\u0947 \u0924\u092c \u090f\u0915\u0926\u092e \u0910\u0938\u0947 \u0910\u0915\u094d\u091f\u0930 \u0915\u0930\u0924\u0947 \u0914\u0930 \u0928\u093e\u091a\u0924\u0947 \u0917\u093e\u0924\u0947 \u091c\u0948\u0938\u0947 \u0909\u0928 \u092a\u0930 \u0915\u094b\u0908 \u092d\u0942\u0924 \u0938\u0935\u093e\u0930 \u0939\u094b \u0917\u092f\u093e \u0939\u094b \u0914\u0930 \u0935\u094b \u091c\u0948\u0928\u094d\u092f\u0942\u0905\u0930\u0940 \u0925\u0930\u094d\u0921 1889 \u0915\u093e \u0926\u093f\u0928 \u0925\u093e\u0964 \u091c\u092c \u0928\u0940\u091a\u093e \u0938\u095c\u0915 \u092a\u0930 \u091a\u0932\u0924\u0947 \u0939\u0941\u090f \u090f\u0915 \u0918\u094b\u095c\u0947 \u0915\u094b \u092a\u0940\u091f\u0924\u0947 \u0926\u0947\u0916 \u092c\u0939\u0941\u0924 \u0939\u0940 \u0907\u092e\u094b\u0936\u0928\u0932 \u0939\u094b \u0917\u090f \u0914\u0930 \u0909\u0938 \u0918\u094b\u095c\u0947 \u0915\u094b \u0917\u0932\u0947 \u0932\u0917\u093e\u0928\u0947 \u0932\u0917\u0947\u0964 \u0935\u094b \u0935\u0939\u0940 \u092a\u0930 \u0916\u0941\u0926 \u0915\u093e \u090f\u0915 \u0924\u092e\u093e\u0936\u093e \u092c\u0928\u093e \u0915\u0930 \u092c\u0947\u0939\u094b\u0936 \u0939\u094b \u0917\u090f \u0914\u0930 \u0935\u0939\u0940\u0902 \u0938\u0947 \u0909\u0928\u0915\u093e \u092e\u0947\u0902\u091f\u0932 \u092c\u094d\u0930\u0947\u0915 \u0921\u093e\u0909\u0928 \u0936\u0941\u0930\u0942 \u0939\u094b \u0917\u092f\u093e\u0964 \u0928\u0940\u091a\u0947 \u0915\u0947 \u0907\u0938 \u0915\u0947\u0938 \u0915\u094b \u0938\u092e\u091d\u0928\u0947 \u0938\u0947 \u0939\u092e\u0947\u0902 \u0939\u094d\u092f\u0942\u092e\u0928 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u0940 \u0915\u0940 \u090f\u0915 \u092c\u0939\u0941\u0924 \u0921\u0940\u092a \u0905\u0928\u094d\u0921\u0930\u0938\u094d\u091f\u0948\u0928\u094d\u0921\u093f\u0902\u0917 \u092e\u093f\u0932 \u0938\u0915\u0924\u0940 \u0939\u0948\u0964 \u0939\u093e\u0909\u090f\u0935\u0930 \u0906\u092e \u0924\u094c\u0930 \u092a\u0930 \u0928\u0940\u091a\u093e \u0915\u0947 \u0907\u0938 \u092a\u093e\u0917\u0932\u092a\u0928 \u0915\u094b \u0928\u094d\u092f\u0942\u0930\u094b\u0938\u093f\u092b\u0932\u093f\u0938 \u091c\u0948\u0938\u0940 \u092c\u093f\u092e\u093e\u0930\u0940 \u0915\u093e \u0928\u0924\u0940\u091c\u093e \u092c\u0924\u093e\u092f\u093e \u091c\u093e\u0924\u093e \u0939\u0948\u0964 \u0932\u0947\u0915\u093f\u0928 \u090f\u0915 \u091c\u0930\u094d\u092e\u0928 \u0938\u094d\u092a\u0947\u0936\u0932\u093f\u0938\u094d\u091f \u090f\u0930\u093f\u0915 \u092a\u094d\u0930\u094b\u0921\u0915\u094d \u0928\u0940\u091a\u0947 \u0915\u0947 \u092e\u0947\u0921\u093f\u0915\u0932 \u0930\u093f\u0915\u0949\u0930\u094d\u0921\u094d\u0938 \u0915\u094b \u090f\u0917\u094d\u095b\u0948\u092e \u0907\u0928 \u0915\u0930\u0928\u0947 \u0915\u0947 \u092c\u093e\u0926 \u0905\u092a\u0928\u0940 \u092c\u0941\u0915 \u0915\u0947 \u092e\u0948\u0921\u0928\u0947\u0938 \u0911\u092b \u0928\u0940\u091a\u093e \u092e\u0947\u0902 \u0932\u093f\u0916\u0924\u0947 \u0939\u0948\u0902 \u0915\u093f \u0938\u093f\u092b\u0932\u093f\u0938 \u0915\u0947 \u0910\u0938\u0947 \u0915\u093f\u0938\u0940 \u0907\u0928\u094d\u092b\u0947\u0915\u094d\u0936\u0928 \u0915\u093e \u0915\u094b\u0908 \u0938\u092c\u0942\u0924 \u0928\u0939\u0940\u0902 \u0939\u0948\u0964 \u0914\u0930 \u0928\u093f\u091a\u093e \u0915\u0947 \u0938\u093f\u092e\u094d\u092a\u091f\u092e\u094d\u0938 \u0938\u093f\u092b\u0932\u093f\u0938 \u0915\u0940 \u092c\u093f\u092e\u093e\u0930\u0940 \u0915\u0947 \u0938\u093e\u0925 \u0915\u0949\u0928\u094d\u0938\u094d\u091f\u0947\u0902\u091f \u092d\u0940 \u0928\u0939\u0940\u0902 \u0939\u0948\u0964 \u091c\u0940\u0935\u0928 \u0905\u092e\u0947\u0930\u093f\u0915\u0928 \u092b\u093f\u0932\u0949\u0938\u092b\u0930 \u091c\u0942\u0932\u093f\u092f\u0928 \u092f\u0902\u0917 \u092d\u0940 \u0905\u092a\u0928\u0940 \u092c\u0941\u0915 \u0928\u0940\u091a\u093e \u0905\u092c \u092c\u093e\u092f\u094b\u0917\u094d\u0930\u093e\u092b\u0940 \u092e\u0947\u0902 \u092c\u094b\u0932\u0924\u0947 \u0939\u0948\u0902\u0964 \u0915\u093f \u0928\u0940\u091a\u0947 \u0915\u0947 \u092a\u093e\u0917\u0932\u092a\u0928 \u0915\u093e \u0938\u092c\u0938\u0947 \u092c\u095c\u093e \u0914\u0930 \u0911\u092c\u094d\u0935\u093f\u092f\u0938 \u0915\u093e\u0930\u0923 \u092b\u093f\u091c\u093f\u0915\u0932 \u092f\u093e \u092b\u093f\u0930 \u091c\u0947\u0928\u0947\u091f\u093f\u0915 \u0928\u0939\u0940\u0902 \u092c\u0932\u094d\u0915\u093f \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u0928\u0947\u091a\u0930 \u0915\u093e \u0932\u0917\u0924\u093e \u0939\u0948\u0964 \u0932\u0947\u0915\u093f\u0928 \u0905\u0917\u0930 \u0928\u0940\u091a\u0947 \u0915\u094b \u092f\u0939 \u092e\u093e\u0932\u0942\u092e \u0925\u093e \u0915\u093f \u0935\u0939 \u0916\u0941\u0926 \u0915\u0947 \u0905\u0928\u0915\u0949\u0928\u094d\u0936\u0938 \u092e\u093e\u0907\u0902\u0921 \u0915\u0940 \u0921\u0947\u0925 \u092e\u0947\u0902 \u0917\u0941\u092e \u0939\u094b \u0938\u0915\u0924\u0947 \u0939\u0948\u0902 \u0924\u094b \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u092b\u093f\u0930 \u092d\u0940 \u0907\u0924\u0928\u093e \u092c\u095c\u093e \u0916\u0924\u0930\u093e \u0915\u094d\u092f\u094b\u0902 \u092e\u094b\u0932 \u0932\u093f\u092f\u093e? \u0907\u0938\u0915\u0947 \u091c\u092c \u0906\u092a\u0915\u094b \u0939\u092e \u0909\u0928\u0915\u0940 \u092c\u0941\u0915 \u0917\u0948\u0938 \u0938\u093e\u0907\u0902\u0938 \u0915\u0947 \u090f\u0915 \u092a\u0948\u0938\u0947\u091c \u092e\u0947\u0902 \u0922\u0942\u0902\u0922 \u0938\u0915\u0924\u0947 \u0939\u0948\u0902, \u091c\u0939\u093e\u0901 \u0935\u094b \u092c\u094b\u0932\u0924\u0947 \u0939\u0948 \u0915\u093f \u0935\u094b \u0905\u0938\u093e\u092e\u093e\u0928\u094d\u092f \u0926\u0930\u094d\u0926 \u0914\u0930 \u0932\u0902\u092c\u0940 \u0914\u0930 \u0938\u094d\u0932\u094b \u0938\u092b\u0930\u093f\u0902\u0917 \u0939\u0940 \u0939\u094b\u0924\u0940 \u0939\u0948\u0902 \u091c\u093f\u0928\u0915\u0940 \u0906\u0917 \u092e\u0947\u0902 \u0939\u092e \u091c\u0932\u0915\u0930 \u0905\u0902\u0924\u093f\u092e \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u0924\u0915 \u092a\u0939\u0941\u0901\u091a \u092a\u093e\u0924\u0947 \u0939\u0948\u0902\u0964 \u092f\u093e\u0928\u0940 \u0928\u0940\u091a\u0947 \u0928\u0947 \u0916\u0941\u0926 \u0915\u094b \u091c\u093e\u0928\u092c\u0942\u091d\u0915\u0930 \u0926\u0930\u094d\u0926 \u0915\u0947 \u0915\u0941\u090f\u0902 \u092e\u0947\u0902 \u092b\u0947\u0902\u0915\u093e\u0964 \u0914\u0930 \u0909\u0928\u0915\u0940 \u0907\u0938\u0940 \u0938\u092b\u0930\u093f\u0902\u0917 \u0915\u0940 \u0935\u091c\u0939 \u0938\u0947 \u0909\u0928\u0915\u0940 \u092b\u093f\u0932\u0949\u0938\u094b\u092b\u093f\u0915\u0932 \u0907\u0928\u0938\u093e\u0907\u091f\u094d\u0938 \u0907\u0924\u0928\u0940 \u0921\u0940\u092a \u0925\u0940\u0964 \u0907\u0938\u0915\u093e \u092e\u0924\u0932\u092c \u090f\u0915 \u0924\u0930\u0939 \u0938\u0947 \u0909\u0928\u0915\u093e \u0907\u0938 \u0924\u0930\u0939 \u092e\u0928 \u0915\u0940 \u0905\u0928\u091c\u093e\u0928 \u0917\u0941\u092b\u093e\u0913\u0902 \u092e\u0947\u0902 \u091c\u093e\u0928\u093e \u0909\u0928\u0915\u0947 \u0932\u093f\u090f \u0935\u093f\u095b\u094d\u0921\u092e \u0939\u093e\u0938\u093f\u0932 \u0915\u0930\u0928\u0947 \u0915\u0940 \u091c\u0930\u0942\u0930\u0924 \u0925\u0940 \u0928\u093e \u0915\u093f \u0915\u094b\u0908 \u0909\u0928\u0915\u0940 \u091a\u0949\u0907\u0938 \u091c\u093f\u0938\u0938\u0947 \u0935\u0939 \u092c\u0926\u0932 \u0938\u0915\u0924\u0947 \u0939\u0948\u0902\u0964 \u0928\u0940\u091a\u093e \u091c\u093e\u0928\u0924\u0947 \u0925\u0947 \u0915\u093f \u0909\u0928\u0915\u0947 \u0907\u0938 \u0915\u093e\u092e \u0914\u0930 \u0938\u0948\u0915\u094d\u0930\u093f\u095e\u093e\u0907\u0938 \u0915\u0940 \u0935\u0948\u0932\u094d\u092f\u0942 \u091a\u093e\u0939\u0947 \u0909\u0928\u0915\u0947 \u091c\u0940\u0924\u0947 \u091c\u0940 \u0928\u093e \u0939\u094b \u0914\u0930 \u0915\u094b\u0908 \u092c\u093e\u0924 \u0928\u0939\u0940\u0902 \u0905\u0917\u0930 \u091c\u094d\u092f\u093e\u0926\u093e \u0932\u094b\u0917 \u0909\u0928\u0915\u0940 \u092c\u0941\u0915\u094d\u0938 \u0928\u0939\u0940\u0902 \u0916\u0930\u0940\u0926 \u0930\u0939\u0947 \u0939\u0948\u0902\u0964 \u0932\u0947\u0915\u093f\u0928 \u092b\u093f\u0930 \u092d\u0940 \u0935\u094b \u092f\u0947 \u092e\u093e\u0928\u0924\u0947 \u0925\u0947 \u0915\u093f \u0909\u0928\u0915\u0940 \u0939\u0930 \u092c\u0941\u0915 \u092f\u0942\u0928\u093f\u091f\u0940 \u0915\u0947 \u0932\u093f\u090f \u090f\u0915 \u0917\u093f\u092b\u094d\u091f \u0939\u0948\u0964 \u090f\u0915 \u0932\u0947\u091f\u0930 \u092e\u0947\u0902 \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u0905\u092a\u0928\u093e \u0926\u0930\u094d\u0926 \u091c\u0924\u093e\u0924\u0947 \u0939\u0941\u090f \u0932\u093f\u0916\u093e \u0915\u093f \u092e\u0948\u0902 45 \u0938\u093e\u0932 \u0915\u093e \u0939\u094b \u091a\u0941\u0915\u093e \u0939\u0942\u0901\u0964 \u0914\u0930 15 \u0915\u093f\u0924\u093e\u092c\u0947\u0902 \u092a\u092c\u094d\u0932\u093f\u0936 \u0915\u0930 \u091a\u0941\u0915\u093e \u0939\u0942\u0901\u0964 \u0932\u0947\u0915\u093f\u0928 \u0905\u092d\u0940 \u0924\u0915 \u092e\u0947\u0930\u0940 \u0915\u093f\u0938\u0940 \u092d\u0940 \u0915\u093f\u0924\u093e\u092c \u0915\u093e \u090f\u0915 \u0905\u0915\u0947\u0932\u093e \u0905\u091a\u094d\u091b\u093e \u0930\u093f\u0935\u094d\u092f\u0941 \u092d\u0940 \u0928\u0939\u0940\u0902 \u0939\u0948\u0964 \u0932\u094b\u0917 \u092e\u0947\u0930\u0947 \u0915\u093e\u092e \u0915\u094b \u0905\u091c\u0940\u092c, \u0916\u0924\u0930\u0928\u093e\u0915 \u0914\u0930 \u092a\u093e\u0917\u0932\u092a\u0928 \u092c\u094b\u0932\u0924\u0947 \u0939\u0948\u0902\u0964 \u092e\u0941\u091d\u0947 \u092f\u0939 \u0938\u094b\u091a\u0915\u0930 \u092c\u0939\u0941\u0924 \u0926\u0941\u0916 \u0939\u094b\u0924\u093e \u0939\u0948 \u0915\u093f \u0907\u0924\u0928\u0947 \u0938\u093e\u0932\u094b\u0902 \u092e\u0947\u0902 \u0915\u093f\u0938\u0940 \u090f\u0915 \u0907\u0902\u0938\u093e\u0928 \u0928\u0947 \u092d\u0940 \u092e\u0947\u0930\u0947 \u0915\u093e\u092e \u0915\u094b \u0921\u093f\u0938\u094d\u0915\u0935\u0930\u094d\u0921 \u0928\u0939\u0940\u0902 \u0915\u0930 \u0930\u0939\u093e, \u0928\u093e \u092e\u0947\u0930\u0940 \u0915\u093f\u0938\u0940 \u0915\u094b \u091c\u0930\u0942\u0930\u0924 \u092a\u095c\u0940 \u0914\u0930 \u0928\u093e \u0939\u0940 \u092e\u0941\u091d\u0947 \u0915\u093f\u0938\u0940\u0928\u0947 \u092a\u094d\u092f\u093e\u0930 \u0915\u0930\u093e\u0902 \u0928\u0940\u091a\u0947 \u0905\u0915\u094d\u0938\u0930 \u0905\u092a\u0928\u0947 \u0905\u0915\u0947\u0932\u0947 \u092a\u0928 \u0914\u0930 \u0909\u0938\u0938\u0947 \u091c\u0941\u095c\u0947 \u0926\u0930\u094d\u0926 \u0915\u0947 \u092c\u093e\u0930\u0947 \u092e\u0947\u0902 \u092c\u0924\u093e\u0924\u0947 \u0925\u0947\u0964 \u091c\u0940\u0935\u0928 \u090f\u0915 \u0926\u0942\u0938\u0930\u0940 \u092b\u093f\u0932\u0949\u0938\u092b\u0930 \u0932\u0942 \u0938\u0932\u094b\u0928 \u0915\u0947 \u0938\u093e\u0925 \u090f\u0915 \u091b\u094b\u091f\u0947 \u0905\u092b\u0947\u092f\u0930 \u0915\u0947 \u092c\u093e\u0926 \u092d\u0940 \u0928\u0940\u091a\u0947 \u0915\u0947 \u092a\u094d\u0930\u092a\u094b\u091c\u0932 \u0915\u094b \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u0924\u0940\u0928 \u092c\u093e\u0930 \u0930\u093f\u091c\u0947\u0915\u094d\u091f \u0915\u0930 \u0926\u093f\u092f\u093e\u0964 \u092c\u093f\u0928\u093e \u0915\u093f\u0938\u0940 \u0939\u094d\u092f\u0942\u092e\u0928 \u0907\u0902\u091f\u0930\u0948\u0915\u094d\u0936\u0928 \u0914\u0930 \u0915\u0928\u0947\u0915\u094d\u0936\u0928 \u0915\u0947 \u0928\u0940\u091a\u0947 \u0916\u0941\u0926 \u0915\u094b \u090f\u0915 \u0910\u0938\u0940 \u091c\u092e\u0940\u0928 \u0938\u0947 \u0915\u0902\u092a\u0947\u0930 \u0915\u0930\u0924\u0947 \u0925\u0947 \u091c\u0939\u093e\u0901 \u092a\u0930 \u0915\u092d\u0940 \u092c\u093e\u0930\u093f\u0936 \u0928\u0939\u0940\u0902 \u0939\u094b\u0924\u0940 \u0914\u0930 \u091c\u0939\u093e \u0916\u0941\u0936 \u0939\u094b\u0928\u0947 \u0915\u0940 \u0915\u094b\u0908 \u0935\u091c\u0939 \u0928\u0939\u0940\u0902 \u0939\u094b\u0924\u0940\u0964 \u0909\u0928\u0915\u0940 \u092c\u093f\u0917\u095c\u0924\u0940 \u0939\u0947\u0932\u094d\u0925, \u0909\u0928\u0915\u0940 \u0915\u093f\u0924\u093e\u092c\u094b\u0902 \u0915\u093e \u0928 \u092c\u093f\u0915\u0928\u093e \u0914\u0930 \u0909\u0928\u0915\u093e \u0905\u0915\u0947\u0932\u093e\u092a\u0928 \u0909\u0928\u0915\u0940 \u092e\u0947\u0902\u091f\u0932 \u0939\u0947\u0932\u094d\u0925 \u0915\u094b \u0916\u0930\u093e\u092c \u0915\u0930\u0924\u093e \u091c\u093e \u0930\u0939\u093e \u0925\u093e \u0914\u0930 \u0907\u0938\u0940 \u0926\u0930\u094d\u0926 \u0928\u0947 \u0909\u0928\u094d\u0939\u0947\u0902 \u0916\u0941\u0926 \u0915\u0947 \u0905\u0902\u0926\u0930 \u091c\u093e\u0915\u0930 \u0909\u0938 \u0916\u091c\u093e\u0928\u0947 \u0915\u094b \u0922\u0942\u0902\u0922\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u092b\u094b\u0930\u094d\u0938 \u0915\u0930\u093e \u091c\u094b \u0915\u093f\u0938\u0940 \u092d\u0940 \u0907\u0902\u0938\u093e\u0928 \u0915\u094b \u092c\u0939\u0941\u0924 \u091c\u094d\u092f\u093e\u0926\u093e \u0928\u0949\u0932\u0947\u091c\u0947\u092c\u0932 \u0914\u0930 \u092a\u0949\u0935\u0930\u092b\u0941\u0932 \u092c\u0928\u093e \u0938\u0915\u0924\u093e \u0939\u0948\u0964 \u0938\u094d\u0935\u093f\u0938 \u0938\u093e\u0907\u0915\u093f\u092f\u093e\u091f\u094d\u0930\u093f\u0938\u094d\u091f \u0905\u092a\u0928\u0940 \u092c\u0941\u0915 \u0938\u093f\u0902\u092c\u0932\u094d\u0938 \u0911\u092b \u091f\u094d\u0930\u093e\u0902\u0938\u092b\u0949\u0930\u094d\u092e\u0947\u0936\u0928 \u092e\u0947\u0902 \u0932\u093f\u0916\u0924\u0947 \u0939\u0948\u0902 \u0915\u093f \u0939\u092e\u093e\u0930\u0947 \u092e\u0928 \u0915\u0940 \u0938\u092c\u0938\u0947 \u0917\u0939\u0930\u0940 \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u092e\u0947\u0902 \u092f\u093e \u092b\u093f\u0930 \u0939\u092e\u093e\u0930\u0947 \u0905\u0928\u0915\u0949\u0928\u094d\u0936\u0938 \u092e\u093e\u0907\u0902\u0921 \u0915\u0947 \u0938\u092e\u0941\u0926\u094d\u0930 \u0915\u0947 \u0924\u0932\u0947 \u092a\u0930 \u090f\u0915 \u0910\u0938\u093e \u0916\u091c\u093e\u0928\u093e \u0939\u094b\u0924\u093e \u0939\u0948 \u091c\u093f\u0938\u0947 \u0924\u0915 \u0938\u093f\u0930\u094d\u092b \u0938\u092c\u0938\u0947 \u0938\u093e\u0939\u0938\u0940 \u0932\u094b\u0917 \u0939\u0940 \u092a\u0939\u0941\u0901\u091a \u0938\u0915\u0924\u0947 \u0939\u0948\u0902\u0964 \u092f\u0947 \u0916\u091c\u093e\u0928\u093e \u090f\u0915 \u0938\u093f\u0902\u092c\u0932 \u0939\u094b\u0924\u093e \u0939\u0948 \u091c\u093f\u0902\u0926\u0917\u0940 \u0915\u0947 \u0938\u092c\u0938\u0947 \u092c\u095c\u0947 \u0938\u0940\u0915\u094d\u0930\u0947\u091f\u094d\u0938 \u0915\u093e, \u091c\u093f\u0938\u0947 \u0905\u0928\u0917\u093f\u0928\u0924 \u0924\u0930\u0940\u0915\u094b\u0902 \u0938\u0947 \u092e\u093f\u0925\u094b\u0932\u0949\u091c\u0940 \u0915\u0947 \u0925\u094d\u0930\u0942 \u090f\u0915\u094d\u0938\u092a\u094d\u0930\u0947\u0938 \u0915\u0939\u093e \u091c\u093e\u0924\u093e \u0939\u0948\u0964 \u0907\u0902\u0938\u093e\u0928\u094b\u0902 \u0928\u0947 \u0939\u092e\u0947\u0936\u093e \u0938\u0947 \u0939\u0940 \u0907\u0938 \u0916\u091c\u093e\u0928\u0947 \u0915\u0940 \u0932\u093e\u0932\u0938\u093e \u0915\u0930\u0940 \u0939\u0948\u0964 \u0928\u0940\u091a\u0947 \u0905\u092a\u0928\u0947 \u0905\u0928\u0915\u0949\u0928\u094d\u0936\u0938 \u092e\u093e\u0907\u0902\u0921 \u092e\u0947\u0902 \u0928\u0940\u091a\u0947 \u091c\u093e\u0924\u0947 \u0917\u090f \u0924\u093e\u0915\u093f \u0935\u094b \u0907\u0938 \u0916\u091c\u093e\u0928\u0947 \u0915\u094b \u092a\u093e\u0915\u0930 \u0905\u092a\u0928\u0947 \u0926\u0930\u094d\u0926 \u0915\u094b \u0938\u094b\u0928\u0947 \u092e\u0947\u0902 \u092c\u0926\u0932 \u0938\u0915\u0947\u0964 \u092f\u0947 \u090f\u0932 \u0915\u092e\u0940 \u0915\u093e \u0905\u0932\u094d\u091f\u0940\u092e\u0947\u091f \u0917\u094b\u0932 \u0925\u093e \u0915\u093f \u0915\u093f\u0938\u0940 \u0924\u0930\u0939 \u0938\u0947 \u092a\u094d\u0930\u093e\u091a\u0940\u0928 \u0915\u0947\u092e\u093f\u0938\u094d\u091f\u094d\u0930\u0940 \u0915\u094b \u092f\u0942\u095b \u0915\u0930\u093e \u091c\u093e \u0938\u0915\u0947\u0964 \u0915\u093f\u0938\u0940 \u0906\u092e \u0927\u093e\u0924\u0941 \u0915\u094b \u0938\u094b\u0928\u0947 \u092e\u0947\u0902 \u092c\u0926\u0932\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u0914\u0930 \u092f\u0939\u0940 \u0935\u093f\u095b\u094d\u0921\u092e \u0932\u093e\u0907\u091f\u092e\u0948\u0928 \u0915\u0947 \u0915\u0947\u0938 \u092e\u0947\u0902 \u092d\u0940 \u0905\u092a\u094d\u0932\u093e\u0908 \u0939\u094b\u0924\u0940 \u0939\u0948\u0964 \u0915\u094d\u092f\u094b\u0902\u0915\u093f \u0939\u0930 \u0907\u0902\u0938\u093e\u0928 \u092f\u0939 \u092c\u093e\u0924 \u091c\u093e\u0928\u0924\u093e \u0939\u0948 \u0915\u093f \u0906\u092a\u0915\u0940 \u092a\u0930\u0938\u0947\u092a\u094d\u0936\u0928 \u0914\u0930 \u0915\u0949\u0928\u094d\u0936\u0938\u0928\u0947\u0938 \u0924\u092c \u0924\u0915 \u0928\u0939\u0940\u0902 \u092c\u095d \u0938\u0915\u0924\u0940 \u091c\u092c \u0924\u0915 \u0906\u092a \u0926\u0941\u0928\u093f\u092f\u093e \u0915\u0940 \u0921\u0942\u0910\u0932\u093f\u091f\u0940 \u0915\u0947 \u092a\u093e\u0930\u094d\u091f \u0926\u0947\u0916\u0928\u093e \u0928\u0939\u0940\u0902 \u0936\u0941\u0930\u0942 \u0915\u0930\u0924\u0947 \u0939\u0948\u0902 \u0914\u0930 \u0915\u0940\u091a\u095c \u092e\u0947\u0902 \u092d\u0940 \u091b\u0941\u092a\u0947 \u0939\u0941\u090f \u0916\u091c\u093e\u0928\u0947 \u0915\u094b \u0928\u0939\u0940\u0902 \u0922\u0942\u0902\u0922 \u092a\u093e\u0924\u0947 \u0939\u0948\u0902\u0964 \u0907\u0938 \u092c\u093e\u0924 \u0915\u094b \u0932\u093f\u091f\u0930\u0932\u094d\u0932\u0940 \u0928\u0939\u0940\u0902 \u092e\u0947\u091f\u093e\u092b\u094b\u0930 \u0928\u093f\u0915\u0932\u0940 \u0938\u092e\u091d\u094b \u0915\u093f \u0939\u0930 \u0935\u094b \u091a\u0940\u095b \u091c\u093f\u0938\u0938\u0947 \u0906\u092a \u0917\u0902\u0926\u093e \u0914\u0930 \u0921\u093f\u0938\u094d\u0917\u0938\u094d\u091f\u093f\u0902\u0917 \u0938\u092e\u091d\u0924\u0947 \u0939\u094b, \u0909\u0938\u092e\u0947\u0902 \u0939\u0940 \u0906\u092a\u0915\u094b \u0917\u094d\u0930\u094b \u0915\u0930\u0928\u0947 \u0915\u0940 \u0938\u092c\u0938\u0947 \u091c\u094d\u092f\u093e\u0926\u093e \u0924\u093e\u0915\u0924 \u0939\u094b\u0924\u0940 \u0939\u0948\u0964 \u0928\u0940\u091a\u0947 \u0928\u0947 \u092d\u0940 \u0916\u0941\u0926 \u0905\u092a\u0928\u0940 \u0907\u0928 \u0905\u0930 \u091c\u0930\u094d\u0928\u0940 \u092a\u0930 \u090f\u0915 \u0932\u0947\u091f\u0930 \u092e\u0947\u0902 \u092f\u0947 \u0932\u093f\u0916\u093e \u0925\u093e \u0915\u093f \u0905\u0928\u0932\u0947\u0938 \u092e\u0941\u091d\u0947 \u0915\u093f\u0938\u0940 \u0924\u0930\u0939 \u0938\u0947 \u090f\u0932 \u0915\u092e\u0940 \u0915\u0947 \u0915\u0942\u095c\u0947 \u0915\u0947 \u0922\u0947\u0930 \u0915\u094b \u0938\u094b\u0928\u0947 \u092e\u0947\u0902 \u0915\u0928\u094d\u0935\u0930\u094d\u091f \u0915\u0930\u0928\u0947 \u0915\u0940 \u091f\u094d\u0930\u093f\u0915 \u092a\u0924\u093e \u0932\u0917 \u091c\u093e\u090f\u0964 \u092e\u0948\u0902 \u0917\u0941\u092e \u0939\u0942\u0901 \u0914\u0930 \u092e\u0941\u091d\u0947 \u0928\u0939\u0940\u0902 \u092e\u093e\u0932\u0942\u092e\u0964 \u0915\u093f \u092e\u0948\u0902 \u0905\u092a\u0928\u0947 \u092e\u0928 \u0915\u0940 \u0907\u0938 \u092d\u0942\u0932\u092d\u0941\u0932\u0948\u092f\u093e \u0938\u0947 \u0915\u0948\u0938\u0947 \u092c\u093e\u0939\u0930 \u0928\u093f\u0915\u0932\u0942\u0902\u0917\u093e\u0964 \u0939\u0930 \u0907\u0902\u0938\u093e\u0928 \u0926\u0941\u0928\u093f\u092f\u093e \u0915\u094b \u0905\u0932\u0917 \u0905\u0932\u0917 \u0932\u0947\u0935\u0932 \u092a\u0930 \u090f\u0915\u094d\u0938\u092a\u0940\u0930\u093f\u092f\u0902\u0938 \u0915\u0930\u0924\u093e \u0939\u0948 \u091c\u0948\u0938\u0947 \u092b\u093f\u091c\u093f\u0915\u0932 \u0932\u0947\u0935\u0932, \u0907\u092e\u094b\u0936\u0928\u0932 \u0932\u0947\u0935\u0932 \u0914\u0930 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u0932\u0947\u0935\u0932\u0964 \u0932\u0947\u0915\u093f\u0928 \u0905\u0917\u0930 \u0939\u092e \u0915\u093f\u0938\u0940 \u0935\u091c\u0939 \u0938\u0947 \u0938\u093f\u0930\u094d\u092b \u090f\u0915 \u0932\u0947\u0935\u0932 \u0915\u0947 \u090f\u0915\u094d\u0938\u094d\u092a\u093f\u0930\u093f\u092f\u0902\u0938 \u092e\u0947\u0902 \u0939\u0940 \u0905\u091f\u0915 \u0917\u090f \u0924\u094b \u0939\u092e\u093e\u0930\u0940 \u0938\u092b\u0930\u093f\u0902\u0917 \u0915\u093e \u0915\u094b\u0908 \u0905\u0902\u0924 \u0939\u0940 \u0928\u0939\u0940\u0902 \u0939\u094b\u0917\u093e\u0964 \u090f\u0915 \u0921\u093f\u0938\u0947\u092c\u0932\u094d\u0921 \u0907\u0902\u0938\u093e\u0928 \u0915\u093e \u090f\u0915\u094d\u0938\u092a\u093f\u0930\u093f\u092f\u0902\u0938 \u092b\u093f\u091c\u093f\u0915\u0932 \u092a\u0947\u0928 \u0914\u0930 \u0921\u093f\u0938\u0947\u092c\u093f\u0932\u093f\u091f\u0940 \u0915\u0940 \u0935\u091c\u0939 \u0938\u0947 \u092c\u0949\u0921\u0940 \u0915\u0947 \u0932\u0947\u0935\u0932 \u092a\u0930 \u0939\u0940 \u0930\u0939 \u091c\u093e\u0924\u093e \u0939\u0948\u0964 \u090f\u0915 \u0907\u092e\u094b\u0936\u0928\u0932 \u0907\u0902\u0938\u093e\u0928, \u091c\u094b \u092a\u0930\u094d\u0938\u0928\u0948\u0932\u093f\u091f\u0940 \u0935\u093e\u0907\u091c \u0915\u093e\u092b\u0940 \u0938\u0947\u0902\u091f\u0940\u092e\u0947\u0902\u091f\u0932 \u0939\u0948, \u0909\u0938\u0915\u093e \u090f\u0915\u094d\u0938\u092a\u093f\u0930\u093f\u092f\u0902\u0938 \u0907\u092e\u094b\u0936\u0928\u094d\u0938 \u0915\u0947 \u092c\u0947\u0938\u093f\u0938 \u092a\u0930 \u0939\u094b\u0924\u093e \u0939\u0948\u0964 \u0939\u093e\u0909\u090f\u0935\u0930 \u0928\u0940\u091a\u093e \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u0932\u0947\u0935\u0932 \u092a\u0930 \u0925\u0915 \u0917\u090f \u0925\u0947 \u0914\u0930 \u0909\u0928\u0915\u0940 \u0930\u093f\u0910\u0932\u093f\u091f\u0940 \u0914\u0930 \u0926\u0941\u0928\u093f\u092f\u093e \u0915\u094b \u0926\u0947\u0916\u0928\u0947 \u0915\u093e \u0928\u091c\u0930\u093f\u092f\u093e \u0909\u0928\u0915\u0940 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u0938\u092b\u0930\u093f\u0902\u0917 \u0915\u0940 \u0935\u091c\u0939 \u0938\u0947 22 \u0925\u093e\u0964 \u091c\u0940\u0935\u0928 \u092d\u0940 \u091c\u092c \u0905\u092a\u0928\u0947 \u0938\u092a\u0928\u094b\u0902 \u0914\u0930 \u0905\u0928\u0915\u0949\u0928\u094d\u0936\u0938 \u092e\u093e\u0907\u0902\u0921 \u0915\u0947 \u0938\u092c\u0938\u0947 \u092b\u0902\u0921\u093e\u092e\u0947\u0902\u091f\u0932 \u0938\u094d\u091f\u094d\u0930\u0915\u094d\u091a\u0930 \u0915\u094b \u091c\u093e\u0928\u0928\u0947 \u0915\u0940 \u0915\u094b\u0936\u093f\u0936 \u0915\u0930 \u0930\u0939\u0947 \u0925\u0947, \u0924\u092c \u0909\u0928\u094d\u0939\u0947\u0902 \u092c\u0939\u0941\u0924 \u091c\u094d\u092f\u093e\u0926\u093e \u0939\u0948\u0932\u094b \u0938\u0947 \u0928\u0947\u0936\u0928\u094d\u0938 \u0914\u0930 \u0921\u0930\u093e\u0935\u0928\u0940 \u0935\u093f\u095b\u0928 \u0938\u093e\u092e\u0928\u0947 \u0932\u0917\u0947 \u0925\u0947, \u091c\u094b \u0915\u0908 \u0918\u0902\u091f\u094b\u0902 \u0924\u0915 \u0909\u0928\u094d\u0939\u0947\u0902 \u091f\u0949\u0930\u094d\u091a\u0930 \u0915\u0930\u0924\u0947 \u0925\u0947 \u0914\u0930 \u0909\u0928\u0915\u093e \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u092a\u0947\u0928 \u092d\u0940 \u0907\u0924\u0928\u093e \u091c\u094d\u092f\u093e\u0926\u093e \u092c\u095d \u0917\u092f\u093e \u0925\u093e \u0915\u093f \u0935\u094b \u0939\u0930 \u0930\u093e\u0924 \u0905\u092a\u0928\u0947 \u0924\u0915\u093f\u090f \u0915\u0947 \u092a\u093e\u0938 \u0917\u0928 \u0930\u0916\u0915\u0930 \u0938\u094b\u0928\u0947 \u0932\u0917\u0947 \u0925\u0947 \u0924\u093e\u0915\u093f \u091c\u0940\u0938 \u092e\u0942\u0935\u092e\u0947\u0902\u091f \u092e\u0947\u0902 \u092d\u0940 \u0909\u0928\u0915\u0940 \u0938\u092b\u0930\u093f\u0902\u0917 \u0905\u0928\u092c\u0947\u092f\u0930\u0947\u092c\u0932 \u092c\u0928 \u091c\u093e\u090f\u0964 \u0924\u092c \u0935\u094b \u0909\u0938\u0940 \u0938\u092e\u092f \u0916\u0941\u0926 \u0915\u094b \u0916\u0924\u094d\u092e \u0915\u0930 \u0938\u0915\u0924\u0947 \u0939\u0948\u0902\u0964 \u0906\u092a \u092c\u0938 \u0905\u092d\u0940 \u0905\u092a\u0928\u0947 \u0907\u0928\u0915\u0940 \u0938\u0924\u0939 \u0914\u0930 \u0909\u0938\u0915\u0940 \u0926\u094b \u0924\u0940\u0928 \u0905\u091a\u094d\u091b\u0940 \u0915\u094d\u0935\u093e\u0932\u093f\u091f\u0940\u091c \u0938\u0947 \u0935\u093e\u0915\u093f\u092b \u0939\u094b\u0902\u0964 \u0932\u0947\u0915\u093f\u0928 \u091c\u094b \u0915\u093e\u0932\u0940 \u0914\u0930 \u0930\u0939\u0938\u094d\u092f\u092e\u092f\u0940 \u091a\u0940\u091c\u0947\u0902 \u0906\u092a \u0915\u0947 \u0905\u0938\u094d\u0924\u093f\u0924\u094d\u0935 \u0915\u0940 \u092c\u0941\u0928\u093f\u092f\u093e\u0926 \u0939\u0948, \u0906\u092a\u0915\u094b \u0909\u0928\u0915\u093e \u0915\u094b\u0908 \u091c\u094d\u091e\u093e\u0928 \u0939\u0940 \u0928\u0939\u0940\u0902 \u0939\u0948\u0964 \u0939\u0930 \u0938\u093e\u0907\u091f \u0915\u0947 \u090f\u0935\u0932\u094d\u092f\u0942\u0936\u0928 \u090f\u0915 \u092c\u095c\u0947 \u092c\u094d\u0930\u0947\u0915 \u0921\u093e\u0909\u0928 \u0938\u0947 \u0939\u0940 \u0936\u0941\u0930\u0942 \u0939\u094b\u0924\u093e \u0939\u0948 \u0914\u0930 \u092f\u0947 \u091a\u0940\u095b \u0906\u092a \u091c\u093f\u0924\u0928\u093e \u0938\u094b\u091a\u0924\u0947 \u0939\u094b \u0909\u0938\u0938\u0947 \u092c\u0939\u0941\u0924 \u091c\u094d\u092f\u093e\u0926\u093e \u0915\u0949\u092e\u0928 \u0939\u0948 \u0939\u093e\u0909\u090f\u0935\u0930 \u090f\u0915 \u0907\u0902\u0938\u093e\u0928 \u0907\u0938 \u091f\u094d\u0930\u093e\u0902\u0938\u092b\u0949\u0930\u094d\u092e\u0947\u0936\u0928 \u0914\u0930 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u0911\u092c\u094d\u0938\u094d\u091f\u0947\u0915\u0932 \u0915\u0940 \u0926\u0942\u0938\u0930\u0940 \u0938\u093e\u0907\u091f \u092a\u0930 \u0938\u0947\u092b\u094d\u0932\u0940 \u092a\u0939\u0941\u0901\u091a \u092a\u093e\u0924\u093e \u0939\u0948 \u092f\u093e \u092b\u093f\u0930 \u0928\u0939\u0940\u0902 \u092f\u0947 \u0921\u093f\u092a\u0947\u0902\u0921 \u0915\u0930\u0924\u093e \u0939\u0948 \u0915\u093f \u0909\u0938\u0915\u0940 \u090f\u0915\u094d\u0938\u091f\u0930\u094d\u0928\u0932 \u0932\u093e\u0907\u092b \u0915\u093f\u0924\u0928\u0940 \u0938\u094d\u091f\u0947\u092c\u0932 \u0939\u0948\u0964 \u091c\u0948\u0938\u0947 \u092f\u0941\u0917 \u0915\u0947 \u092d\u0940 \u0928\u0940\u091a\u0947 \u0915\u0940 \u0924\u0930\u0939 \u0939\u092e\u0947\u0936\u093e \u0915\u0947 \u0932\u093f\u090f \u092a\u093e\u0917\u0932 \u0939\u094b \u091c\u093e\u0928\u0947 \u0915\u0947 \u092a\u0942\u0930\u0947 \u091a\u093e\u0928\u094d\u0938\u0947\u0938 \u0925\u0947\u0964 \u0914\u0930 \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u092d\u0940 \u0905\u092a\u0928\u0940 \u0915\u0940 \u092c\u0939\u0941\u0924 \u0938\u0947 \u0917\u0939\u0930\u0940 \u0907\u0928\u0938\u093e\u0907\u091f\u094d\u0938 \u0915\u0940 \u092e\u0926\u0926 \u0938\u0947 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u0940 \u0915\u0940 \u092b\u0940\u0932\u094d\u0921 \u0915\u094b \u0907\u0924\u0928\u093e \u0930\u093f\u0932\u093e\u090f\u092c\u0932 \u0914\u0930 \u0907\u0902\u091f\u094d\u0930\u0947\u0938\u094d\u091f\u093f\u0902\u0917 \u092c\u0928\u093e\u092f\u093e\u0964 \u0932\u0947\u0915\u093f\u0928 \u091c\u093f\u0928 \u091a\u0940\u091c\u094b\u0902 \u0928\u0947 \u092f\u094b\u0917 \u0915\u094b \u0905\u0928\u0915\u0949\u0928\u094d\u0936\u0938 \u092e\u093e\u0907\u0902\u0921 \u0915\u0940 \u092a\u093e\u0935\u0930 \u0915\u0940 \u0906\u0917\u0947 \u092d\u0940 \u0938\u0932\u093e\u092e\u0924 \u0930\u0916\u093e, \u0935\u0939 \u0925\u0940 \u0909\u0928\u0915\u0940 \u092b\u0948\u092e\u093f\u0932\u0940\u0964 \u0909\u0928\u0915\u0947 \u092a\u0947\u0936\u0947\u0902\u091f\u094d\u0938, \u0909\u0928\u0915\u0947 \u0921\u0947\u0932\u0940 \u0930\u0942\u091f\u0940\u0928 \u0938\u094d \u0909\u0928\u0915\u0947 \u092c\u091a\u094d\u091a\u0947 \u0914\u0930 \u0909\u0928\u0915\u0940 \u0935\u093f\u095b\u094d\u0921\u092e \u091c\u094b \u0909\u0928\u094d\u0939\u0947\u0902 \u092c\u094b\u0932\u0924\u0940 \u0925\u0940 \u0915\u093f \u0935\u094b \u0905\u092a\u0928\u0947 \u0935\u093f\u095b\u0928 \u0938\u0947 \u092e\u0947\u0902 \u0926\u093f\u0916\u0940, \u0915\u093f\u0938\u0940 \u092d\u0940 \u091a\u0940\u095b \u0938\u0947 \u0921\u093f\u091f\u0948\u091a \u0930\u0939\u0947, \u092f\u0939\u0940 \u0924\u094b \u0930\u0940\u095b\u0928 \u0939\u0948 \u0915\u093f \u0915\u094d\u092f\u094b\u0902 \u0939\u093e\u0907\u0932\u0940 \u0910\u0921\u0935\u093e\u0928\u094d\u0938 \u0921\u0940 \u0939\u094b\u0917\u0940\u0964 \u0907\u0938 \u092d\u0940 \u0905\u092a\u0928\u0947 \u092e\u0928 \u0915\u0940 \u0917\u0939\u0930\u093e\u0907\u092f\u094b\u0902 \u0938\u0947 \u0935\u093e\u0915\u093f\u092b \u0924\u094b \u0939\u094b\u0924\u0947 \u0939\u0948\u0902 \u0932\u0947\u0915\u093f\u0928 \u091c\u0938\u094d\u091f \u092c\u093f\u0915\u0949\u095b \u0935\u094b \u0905\u092a\u0928\u0940 \u0938\u093e\u0907\u091f \u0915\u0947 \u0915\u0949\u0928\u094d\u091f\u0947\u0902\u091f \u0938\u0947 \u0915\u092d\u0940 \u092d\u0940 \u0915\u094b\u0908 \u0905\u091f\u0948\u091a\u092e\u0947\u0902\u091f \u0939\u0940 \u0928\u0939\u0940\u0902 \u092c\u0928\u093e\u0924\u0947\u0964 \u0935\u094b \u0905\u0928\u0915\u0949\u0928\u094d\u0936\u0938 \u092e\u093e\u0907\u0902\u0921 \u0915\u0940 \u0917\u093f\u0930\u092b\u094d\u0924 \u0938\u0947 \u0938\u0947\u092b \u0930\u0939\u0924\u0947 \u0939\u0948\u0902\u0964 \u0932\u0947\u0915\u093f\u0928 \u0928\u0940\u091a\u0947 \u0915\u0940 \u0938\u092c\u0938\u0947 \u092c\u095c\u0940 \u0917\u0932\u0924\u0940 \u092f\u0939 \u0925\u0940 \u0915\u093f \u0935\u094b \u0916\u0941\u0926 \u0915\u094b \u091c\u0940\u0938\u0938, \u092c\u0941\u0926\u094d\u0927, \u0910\u0932\u0948\u0915\u094d\u095b\u0948\u0923\u094d\u0921\u0930 \u0915\u0947 \u0917\u094d\u0930\u0947\u091f \u0914\u0930 \u0939\u0930 \u092c\u095c\u0947 \u092d\u0917\u0935\u093e\u0928 \u092f\u093e \u092b\u093f\u0930 \u0935\u094b \u0930\u093f\u0935\u0930\u094d\u0938 \u0906\u0907\u0921\u0947\u0902\u091f\u093f\u092b\u093e\u0907 \u0915\u0930\u0928\u0947 \u0932\u0917\u0947 \u0925\u0947\u0964 \u0928\u0940\u091a\u0947 \u0905\u092a\u0928\u0947 \u092a\u0948\u0930\u094b\u0902 \u0915\u0947 \u0928\u0940\u091a\u0947 \u0938\u0947 \u091c\u092e\u0940\u0928 \u0916\u094b\u0928\u0947 \u0932\u0917\u0947 \u0925\u0947\u0964 \u091c\u0938\u094d\u091f \u092c\u093f\u0915\u0949\u095b \u0909\u0928\u0915\u0940 \u090f\u0915 \u0930\u093f\u0910\u0932\u093f\u091f\u0940 \u0915\u094b \u092c\u0948\u0932\u0947\u0902\u0938 \u0915\u0930\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u0909\u0928\u0915\u0940 \u0906\u0909\u091f\u0930 \u0930\u093f\u0910\u0932\u093f\u091f\u0940 \u092e\u0947\u0902 \u0915\u0941\u091b \u092d\u0940 \u0928\u0949\u0930\u094d\u092e\u0932 \u092f\u093e \u092b\u093f\u0930 \u0938\u094d\u091f\u0947\u092c\u0932 \u0928\u0939\u0940\u0902 \u0925\u093e\u0964 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u0940 \u092e\u0947\u0902 \u0907\u0938 \u0915\u0902\u0921\u093f\u0936\u0928 \u0915\u094b \u092c\u094b\u0932\u0924\u0947 \u0939\u0948\u0902 \u0938\u093e\u0907\u0915\u093f\u0915 \u0907\u0928\u094d\u092b\u094d\u0932\u0947\u0936\u0928, \u091c\u0939\u093e\u0901 \u092a\u0930 \u090f\u0915 \u0907\u0902\u0938\u093e\u0928 \u0915\u0940 \u0938\u094d\u092a\u094d\u0930\u0947 \u0916\u0941\u0926 \u0909\u0938 \u0907\u0902\u0938\u093e\u0928 \u0915\u094b \u092e\u0928 \u0914\u0930 \u0936\u0930\u0940\u0930 \u0915\u0947 \u092a\u093e\u0930 \u091c\u093e\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u092a\u0941\u0936 \u0915\u0930\u0928\u0947 \u0915\u0947 \u092c\u091c\u093e\u092f \u0915\u0930\u092a\u094d\u091f \u092c\u0928\u0915\u0930 \u092c\u0938 \u0916\u0941\u0926 \u0938\u0947 \u092c\u095c\u0940 \u091a\u0940\u091c\u094b\u0902 \u0915\u0947 \u0906\u0907\u0921\u093f\u092f\u093e \u0906\u0907\u0921\u0947\u0902\u091f\u093f\u092b\u093e\u0907 \u0915\u0930\u0928\u0947 \u0932\u0917\u0924\u0940 \u0939\u0948 \u0914\u0930 \u0907\u0938\u0940 \u0938\u093e\u0907\u0915\u093f\u0915 \u0915\u0930\u092a\u094d\u0936\u0928 \u0938\u0947 \u0939\u0940 \u0938\u0941\u092a\u093f\u0930\u093f\u092f\u094b\u0930\u093f\u091f\u0940 \u0914\u0930 \u0917\u0949\u0921 \u0915\u0949\u092e\u094d\u092a\u094d\u0932\u0947\u0915\u094d\u0938 \u091c\u0948\u0938\u0940 \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u092a\u094d\u0930\u0949\u092c\u094d\u0932\u092e\u094d\u0938 \u0915\u092d\u0940 \u091c\u0928\u092e \u0939\u094b\u0924\u093e \u0939\u0948\u0964 \u0928\u0940\u091a\u0947 \u0915\u093e \u092a\u093e\u0917\u0932\u092a\u0928 \u0905\u0938\u0932\u0940 \u0938\u094d\u092a\u093f\u0930\u093f\u091a\u0942\u0910\u0932\u093f\u091f\u0940 \u0938\u0947 \u091c\u094d\u092f\u093e\u0926\u093e \u0926\u0942\u0930 \u0928\u0939\u0940\u0902 \u0939\u0948\u0964 \u092c\u0938 \u092b\u0930\u094d\u0915 \u0907\u0924\u0928\u093e \u0939\u0948 \u0915\u093f \u0938\u094d\u092a\u093f\u0930\u093f\u091a\u0941\u0905\u0932 \u0932\u094b\u0917 \u092c\u094b\u0932\u0924\u0947 \u0939\u0948 \u0915\u093f \u0939\u0930 \u0915\u093f\u0938\u0940 \u092e\u0947\u0902 \u092d\u0917\u0935\u093e\u0928 \u0939\u0948\u0902, \u0932\u0947\u0915\u093f\u0928 \u0928\u0940\u091a\u093e \u0905\u092a\u0928\u0947 \u092e\u0928 \u0938\u0947 \u0906\u0907\u0921\u0947\u0902\u091f\u093f\u092b\u093e\u0907 \u0915\u0930\u0928\u0947 \u0915\u0940 \u0935\u091c\u0939 \u0938\u0947 \u092f\u0947 \u092c\u094b\u0932\u0924\u0947 \u0925\u0947 \u0915\u093f \u092e\u0948\u0902 \u0939\u0940 \u092d\u0917\u0935\u093e\u0928 \u0939\u0942\u0901 \u0914\u0930 \u091c\u0948\u0938\u093e \u0915\u093f \u0928\u0940\u091a\u093e \u0928\u0947 \u0905\u092a\u0928\u0940 \u092c\u0941\u0915 \u0926\u0947\u0917\u093e \u0938\u093e\u0907\u0902\u0938 \u092e\u0947\u0902 \u092c\u094b\u0932\u093e \u0925\u093e \u0915\u093f \u091c\u094b \u092d\u0940 \u0905\u092a\u0928\u0947 \u0905\u0902\u0926\u0930 \u091d\u093e\u0902\u0915\u0915\u0930 \u0916\u0941\u0926 \u092e\u0947\u0902 \u0938\u092e\u093e\u090f \u092c\u094d\u0930\u0939\u094d\u092e\u093e\u0902\u0921 \u0915\u094b \u0926\u0947\u0916 \u0932\u0947\u0924\u093e \u0939\u0948, \u0909\u0938\u0947 \u092a\u0924\u093e \u0939\u094b\u0924\u093e \u0939\u0948 \u0915\u093f \u092f\u0947 \u092c\u094d\u0930\u0939\u094d\u092e\u093e\u0902\u0921 \u0915\u093f\u0924\u0928\u093e \u0905\u0928\u093f\u092f\u092e\u093f\u0924 \u0939\u0948\u0964 \u0914\u0930 \u0915\u093f\u0938 \u0924\u0930\u0939 \u0938\u0947 \u092f\u0939 \u0939\u092e\u0947\u0902 \u0905\u0938\u094d\u0924\u093f\u0924\u094d\u0935 \u0915\u0940 \u0905\u0902\u0924\u0939\u0940\u0928 \u092d\u0942\u0932\u092d\u0941\u0932\u0948\u092f\u093e \u0924\u0915 \u092a\u0939\u0941\u0902\u091a\u093e \u0938\u0915\u0924\u093e \u0939\u0948?",
"target": "\u092b\u094d\u0930\u0947\u0921\u0930\u093f\u0915 \u0928\u0940\u0924\u094d\u0936\u0947 \u0905\u092c \u0924\u0915 \u0915\u0947 \u0938\u092c\u0938\u0947 \u092a\u094d\u0930\u092d\u093e\u0935\u0936\u093e\u0932\u0940 \u0926\u093e\u0930\u094d\u0936\u0928\u093f\u0915\u094b\u0902 \u092e\u0947\u0902 \u0938\u0947 \u090f\u0915 \u0925\u0947\u0964 \u0935\u0939 \u092a\u093e\u0930\u0902\u092a\u0930\u093f\u0915 \u0928\u0948\u0924\u093f\u0915\u0924\u093e \u0914\u0930 \u0927\u0930\u094d\u092e \u0915\u0940 \u0906\u0932\u094b\u091a\u0928\u093e \u0915\u0930\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u092a\u094d\u0930\u0938\u093f\u0926\u094d\u0927 \u0939\u0948\u0902\u0964 \u0932\u0947\u0915\u093f\u0928 \u0905\u092a\u0928\u0947 \u092a\u0942\u0930\u0947 \u091c\u0940\u0935\u0928 \u092e\u0947\u0902 \u0909\u0928\u094d\u0939\u094b\u0902\u0928\u0947 \u092c\u0939\u0941\u0924 \u0936\u093e\u0930\u0940\u0930\u093f\u0915 \u0914\u0930 \u092e\u093e\u0928\u0938\u093f\u0915 \u092a\u0940\u0921\u093c\u093e \u091d\u0947\u0932\u0940\u0964 \u092a\u0939\u0932\u0947 \u0926\u0930\u094d\u0926 \u0936\u093e\u0930\u0940\u0930\u093f\u0915 \u0925\u093e, \u092b\u093f\u0930 \u092e\u093e\u0928\u0938\u093f\u0915 \u0930\u0942\u092a \u0938\u0947 \u092d\u0940 \u0938\u0924\u093e\u0928\u0947 \u0932\u0917\u093e\u0964 \u0927\u0940\u0930\u0947-\u0927\u0940\u0930\u0947 \u0935\u0939 \u0916\u0941\u0926 \u0915\u094b \u092a\u093e\u0917\u0932\u092a\u0928 \u0915\u0940 \u0913\u0930 \u0932\u0947 \u0917\u092f\u093e\u0964 \u0932\u0947\u0915\u093f\u0928 \u0907\u0938\u0915\u0947 \u092a\u0940\u091b\u0947 \u092e\u0941\u0916\u094d\u092f \u0915\u093e\u0930\u0923 \u0915\u094d\u092f\u093e \u0925\u093e? \u0915\u0941\u091b \u0932\u094b\u0917 \u0936\u093e\u0930\u0940\u0930\u093f\u0915 \u0930\u094b\u0917\u094b\u0902 \u0915\u094b \u091c\u093f\u092e\u094d\u092e\u0947\u0926\u093e\u0930 \u0920\u0939\u0930\u093e\u0924\u0947 \u0939\u0948\u0902, \u0932\u0947\u0915\u093f\u0928 \u0939\u093e\u0932 \u0939\u0940 \u092e\u0947\u0902 \u0915\u0908 \u0935\u093f\u0926\u094d\u0935\u093e\u0928\u094b\u0902 \u0915\u093e \u091d\u0941\u0915\u093e\u0935 \u0938\u094d\u0935\u092a\u094d\u0930\u0947\u0930\u093f\u0924 \u092a\u093e\u0917\u0932\u092a\u0928 \u0915\u0940 \u0938\u0902\u092d\u093e\u0935\u0928\u093e \u0915\u0940 \u0913\u0930 \u0930\u0939\u093e \u0939\u0948\u0964 \u092e\u0924\u0932\u092c \u0939\u094b \u0938\u0915\u0924\u093e \u0939\u0948 \u0915\u093f \u0928\u0940\u0924\u094d\u0936\u0947 \u0905\u092a\u0928\u0940 \u092e\u0930\u094d\u091c\u0940 \u0938\u0947 \u092a\u093e\u0917\u0932 \u0939\u094b \u0917\u092f\u093e \u0939\u094b, \u0909\u0938\u0928\u0947 \u092f\u0939 \u091c\u094b\u0916\u093f\u092e \u0909\u0920\u093e\u092f\u093e\u0964 \u0907\u0938 \u092a\u0949\u0921\u0915\u093e\u0938\u094d\u091f \u0938\u0947\u0917\u092e\u0947\u0902\u091f \u092e\u0947\u0902 \u0939\u092e \u0907\u0938 \u0935\u093f\u091a\u093e\u0930 \u0915\u094b \u0914\u0930 \u0905\u0927\u093f\u0915 \u090f\u0915\u094d\u0938\u092a\u094d\u0932\u094b\u0930 \u0915\u0930\u0947\u0902\u0917\u0947 \u0914\u0930 \u0938\u092e\u091d\u0947\u0902\u0917\u0947 \u0915\u093f \u0928\u0940\u0924\u094d\u0936\u0947 \u0928\u0947 \u091c\u093f\u0938 \u0924\u0930\u0939 \u0938\u0947 \u0915\u093f\u092f\u093e \u0909\u0938\u0915\u093e \u0905\u0902\u0924 \u0915\u094d\u092f\u094b\u0902 \u0939\u0941\u0906\u0964"
},
{
"text": "\u0906\u0938\u093e\u0928 \u0914\u0930 \u092e\u0941\u0936\u094d\u0915\u093f\u0932 \u092f\u0947 \u0936\u092c\u094d\u0926 \u0939\u0930 \u0907\u0902\u0938\u093e\u0928 \u0915\u0947 \u0932\u093f\u090f \u092c\u0939\u0941\u0924 \u0939\u0940 \u0930\u093f\u0932\u0947\u091f\u093f\u0935 \u0939\u094b\u0924\u0947 \u0939\u0948\u0964 \u091c\u0948\u0938\u0947 120 \u0915\u093f\u0932\u094b \u0915\u0947 \u0921\u0902\u092c\u0932 \u0915\u0908 \u0932\u094b\u0917\u094b\u0902 \u0915\u0947 \u0932\u093f\u090f \u092c\u0939\u0941\u0924 \u092d\u093e\u0930\u0940 \u0939\u094b\u0924\u093e \u0939\u0948 \u0914\u0930 \u0926\u094b \u091a\u093e\u0930 \u0930\u0947\u092a\u093f\u091f\u0947\u0936\u0928 \u092e\u0947\u0902 \u0939\u0940 \u0935\u0939 \u0916\u0941\u0926 \u0915\u094b \u091a\u094b\u091f \u092a\u0939\u0941\u0902\u091a\u093e \u0938\u0915\u0924\u0947 \u0939\u0948\u0902, \u091c\u092c\u0915\u093f \u0915\u0908 \u0932\u094b\u0917 \u0907\u0924\u0928\u0947 \u0935\u091c\u0928 \u0938\u0947 \u0935\u0949\u0930\u094d\u092e\u0905\u092a \u0915\u0930\u0924\u0947 \u0939\u0948\u0902\u0964 \u0915\u093f\u0938\u0940 \u092d\u0940 \u0915\u093e\u092e \u0915\u0940 \u0921\u093f\u092b\u093f\u0915\u0932\u094d\u091f\u0940 \u0921\u093f\u092a\u0947\u0902\u0921 \u0915\u0930\u0924\u0940 \u0939\u0948\u0964 \u0909\u0938 \u0915\u093e\u092e \u0915\u094b \u0915\u0930\u0928\u0947 \u0935\u093e\u0932\u0947 \u0907\u0902\u0938\u093e\u0928 \u0915\u0940 \u0915\u093e\u092c\u093f\u0932\u093f\u092f\u0924 \u0915\u0947 \u090a\u092a\u0930 \u0905\u0917\u0930 \u090f\u0915 \u0938\u094d\u091f\u0942\u0921\u0947\u0902\u091f \u0928\u0947 \u090f\u0917\u094d\u095b\u0948\u092e \u0915\u0940 \u092a\u0942\u0930\u0940 \u0924\u0948\u092f\u093e\u0930\u0940 \u0915\u0930\u0940 \u0939\u0948 \u0914\u0930 \u092a\u093f\u091b\u0932\u0947 \u0938\u093e\u0932\u094b\u0902 \u0915\u0947 \u0915\u094d\u0935\u0947\u0936\u094d\u091a\u0928 \u092a\u0947\u092a\u0930\u094d\u0938 \u0915\u094b \u092d\u0940 \u0938\u094d\u091f\u0921\u0940 \u0915\u0930 \u0930\u0939\u093e \u0939\u0948 \u0924\u094b \u0909\u0938\u0915\u0947 \u0932\u093f\u090f \u0935\u094b \u090f\u0917\u094d\u095b\u0948\u092e \u0906\u0938\u093e\u0928 \u0939\u094b\u0917\u093e, \u091c\u092c\u0915\u093f \u092c\u093f\u0928\u093e \u0924\u0948\u092f\u093e\u0930\u0940 \u0935\u093e\u0932\u0947 \u0938\u094d\u091f\u0942\u0921\u0947\u0902\u091f \u0915\u094b \u090f\u0917\u094d\u095b\u0948\u092e \u092e\u0941\u0936\u094d\u0915\u093f\u0932 \u0914\u0930 \u0906\u0909\u091f \u0911\u092b \u0938\u093f\u0932\u0947\u092c\u0938 \u0932\u0917\u0947\u0917\u093e\u0964 \u0938\u093f\u092e\u093f\u0932\u0930\u0932\u0940 \u091c\u093f\u0928 \u0932\u094b\u0917\u094b\u0902 \u0915\u094b \u091c\u093f\u0902\u0926\u0917\u0940 \u0915\u0940 \u092a\u0942\u0930\u0940 \u0938\u092e\u091d \u0939\u094b\u0924\u0940 \u0939\u0948 \u0935\u094b \u0936\u0924\u0930\u0902\u091c \u0915\u0947 \u0916\u0947\u0932 \u0915\u0940 \u0924\u0930\u0939 \u0926\u0941\u0936\u094d\u092e\u0928 \u0915\u0940 \u0939\u0930 \u091a\u093e\u0932 \u0915\u094b \u092a\u0939\u0932\u0947 \u0926\u0947\u0916\u0924\u0947 \u0914\u0930 \u0938\u092e\u091d\u0924\u0947 \u0939\u0948\u0902\u0964 \u092a\u0942\u0930\u0947 \u091a\u0947\u0938 \u092c\u094b\u0930\u094d\u0921 \u0915\u093e \u092e\u0941\u0906\u092f\u0928\u093e \u0915\u0930\u0924\u0947 \u0939\u0948\u0902 \u0914\u0930 \u092b\u093f\u0930 \u0905\u0917\u0932\u0940 \u091a\u093e\u0932 \u091a\u0932\u0924\u0947 \u0939\u0948\u0902 \u0907\u0928 \u0911\u0930\u094d\u0921\u0930 \u091f\u0942 \u0935\u093f\u0928 \u0932\u093e\u0907\u092b \u092f\u0942 \u0928\u0940\u0921 \u091f\u0941 \u092a\u0941\u091f \u092f\u094b\u0930\u0938\u0947\u0932\u094d\u092b \u0907\u0928 \u0926 \u092e\u094b\u0938\u094d\u091f \u090f\u0921\u0935\u093e\u0902\u091f\u0947\u091c \u092f\u0938 \u090f\u0915\u094d\u0938\u091f\u0930\u094d\u0928\u0932 \u0910\u0902\u0921 \u0907\u0902\u091f\u0930\u0928\u0932 \u0938\u093f\u091a\u0941\u090f\u0936\u0928\u094d\u0938 \u0906\u092a \u091a\u093e\u0939\u0947 \u0915\u093f\u0938\u0940 \u092d\u0940 \u092c\u0948\u0915\u0917\u094d\u0930\u093e\u0909\u0902\u0921 \u0938\u0947 \u0939\u094b \u0914\u0930 \u0932\u093e\u0907\u092b \u092e\u0947\u0902 \u0915\u093f\u0924\u0928\u0947 \u092d\u0940 \u092a\u0940\u091b\u0947 \u0939\u094b, \u0906\u092a \u0917\u0947\u092e \u0915\u094b \u091c\u0940\u0924 \u0938\u0915\u0924\u0947 \u0939\u094b\u0964 \u0905\u092a\u0928\u0940 \u0932\u093e\u0907\u092b \u092e\u0947\u0902 \u0907\u0928 \u092a\u093e\u0902\u091a \u0932\u093e\u0907\u0938\u0947\u0902\u0938 \u0915\u094b \u0905\u092a\u0928\u093e\u0915\u0930 \u0928\u0902\u092c\u0930 \u0935\u0928 \u0925\u093f\u0902\u0915 \u0907\u091f\u094d\u0938 \u0938\u094d\u092a\u0940\u0915 \u0905\u092c\u093e\u0909\u091f \u092f\u094b\u0930 \u0917\u094b\u0932\u094d\u0938 \u0906\u0938\u093f\u092b \u0926\u0947 \u0939\u0948\u0935 \u0911\u0932\u0930\u0947\u0921\u0940 \u0939\u0948\u092a\u0928\u094d\u0921\u0964 \u0939\u092e\u093e\u0930\u093e \u092e\u0928 \u0939\u092e\u0947\u0936\u093e \u092c\u0938 \u0926\u094b \u0935\u093f\u0930\u094b\u0927\u0940 \u0916\u094d\u092f\u093e\u0932\u094b\u0902 \u0915\u0947 \u092c\u093e\u0930\u0947 \u092e\u0947\u0902 \u0939\u0940 \u0938\u094b\u091a \u0938\u0915\u0924\u093e \u0939\u0948\u0964 \u091c\u0948\u0938\u0947 \u0905\u091a\u094d\u091b\u093e \u0914\u0930 \u092c\u0941\u0930\u093e \u092f\u093e \u092b\u093f\u0930 \u092a\u0949\u095b\u093f\u091f\u093f\u0935 \u0914\u0930 \u0928\u0947\u0917\u0947\u091f\u093f\u0935\u0964 \u0906\u092a \u091c\u0940\u0938 \u092d\u0940 \u0916\u094d\u092f\u093e\u0932 \u092a\u0930 \u0905\u092a\u0928\u0940 \u0905\u091f\u0947\u0928\u094d\u0936\u0928 \u0926\u0947\u0924\u0947 \u0939\u094b \u0914\u0930 \u091c\u0940\u0938 \u092d\u0940 \u0924\u0930\u0939 \u0915\u0940 \u092c\u093e\u0924\u0947\u0902 \u0915\u0930\u0924\u0947 \u0939\u094b\u0964 \u0906\u092a\u0915\u093e \u092e\u0928 \u0906\u092a\u0915\u0947 \u092a\u0942\u0930\u0947 \u0938\u093f\u0938\u094d\u091f\u092e \u0915\u094b \u0909\u0938\u0940 \u092b\u094d\u0932\u0947\u0935\u0930 \u0915\u093e \u092c\u0928\u093e \u0926\u0947\u0924\u093e \u0939\u0948\u0964 \u092e\u0928 \u0938\u093f\u0930\u094d\u092b \u0910\u0921\u093f\u0936\u0928 \u0914\u0930 \u092e\u0932\u094d\u091f\u093f\u092a\u094d\u0932\u093f\u0915\u0947\u0936\u0928 \u0915\u094b \u091c\u093e\u0928\u0924\u093e \u0939\u0948 \u0914\u0930 \u0909\u0938\u0947 \u0938\u092c \u091f\u094d\u0930\u0948\u0915\u094d\u0936\u0928 \u0914\u0930 \u0921\u093f\u0935\u093f\u095b\u0928 \u0915\u0930\u0928\u093e \u0928\u0939\u0940\u0902 \u0906\u0924\u093e\u0964 \u092f\u093e\u0928\u0940 \u0906\u092a \u0915\u093f\u0938\u0940 \u092d\u0940 \u0916\u094d\u092f\u093e\u0932 \u0915\u094b \u0905\u092a\u0928\u0947 \u092e\u0928 \u0938\u0947 \u092b\u094b\u0930\u094d\u0938\u094d\u092b\u0941\u0932\u0940 \u092c\u093e\u0939\u0930 \u0928\u0939\u0940\u0902 \u0928\u093f\u0915\u093e\u0932 \u0938\u0915\u0924\u0947 \u0939\u0948\u0902\u0964 \u0906\u092a \u092c\u0938 \u0928\u090f \u0916\u094d\u092f\u093e\u0932\u094b\u0902 \u0915\u094b \u0910\u0921 \u0915\u0930 \u0938\u0915\u0924\u0947 \u0939\u094b\u0964 \u0926\u0948\u091f \u0938\u094d\u0935\u093e\u0907 \u0916\u0941\u0926 \u0915\u094b \u0930\u093f\u092a\u0940\u091f\u0947\u0921 \u0932\u0940 \u092a\u0949\u095b\u093f\u091f\u093f\u0935, \u092b\u0949\u0930\u094d\u092e\u0947\u0902\u0936\u0928\u094d\u0938 \u0914\u0930 \u0916\u0941\u0926 \u0915\u0947 \u0917\u094b\u0932 \u0938\u0947 \u092b\u0940\u0921 \u0915\u0930\u0928\u093e \u092c\u0939\u0941\u0924 \u095b\u0930\u0942\u0930\u0940 \u0939\u094b\u0924\u093e \u0939\u0948\u0964 \u0916\u0941\u0926 \u0915\u094b \u092a\u0949\u095b\u093f\u091f\u093f\u0935 \u0932\u0940 \u091f\u094d\u0930\u093e\u0902\u0938\u092b\u0949\u0930\u094d\u092e \u0915\u0930\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u091c\u092c \u092d\u0940 \u0906\u092a \u0905\u092a\u0928\u0947 \u0917\u094b\u0932\u094d\u0938 \u0915\u0947 \u092c\u093e\u0930\u0947 \u092e\u0947\u0902 \u0938\u094b\u091a\u094b \u092f\u093e \u092c\u094b\u0932\u094b \u0924\u092c \u0910\u0938\u093e \u092b\u0940\u0932 \u0915\u0930\u094b \u091c\u0948\u0938\u0947 \u0906\u092a\u0928\u0947 \u0909\u0938 \u0917\u094b\u0932 \u0915\u094b \u0911\u0932\u0930\u0947\u0921\u0940 \u0905\u091a\u0940\u0935 \u0915\u0930 \u0932\u093f\u092f\u093e \u0939\u0948\u0964 \u092b\u0949\u0930 \u090f\u0917\u094d\u095b\u0948\u092e\u094d\u092a\u0932 \u092f\u0947 \u092c\u093f\u0932\u0940\u0935 \u0915\u0930\u0928\u0947 \u0938\u0947 \u0915\u093f \u0906\u092a \u0911\u0932\u0930\u0947\u0921\u0940 \u090f\u0915 \u0905\u0930\u092c\u092a\u0924\u093f \u0939\u094b, \u0906\u092a\u0915\u093e \u092c\u0939\u0941\u0924 \u092c\u095c\u093e \u092c\u093f\u095b\u0928\u0947\u0938 \u0939\u0948 \u0906\u092a\u0915\u0947 \u0905\u0902\u0926\u0930 \u092c\u0939\u0941\u0924 \u0938\u093e\u0930\u0947 \u090f\u092e\u094d\u092a\u0932\u0949\u0908\u0938 \u0915\u093e\u092e \u0915\u0930\u0924\u0947 \u0939\u0948\u0902 \u0914\u0930 \u0906\u092a \u0926\u0947\u0936 \u0915\u0940 \u0917\u094d\u0930\u094b\u0925 \u092e\u0947\u0902 \u092c\u0939\u0941\u0924 \u091c\u094d\u092f\u093e\u0926\u093e \u0915\u0949\u0928\u094d\u091f\u094d\u0930\u093f\u092c\u094d\u092f\u0942\u091f \u0915\u0930 \u0930\u0939\u0947 \u0939\u094b\u0964 \u092f\u0942 \u0938\u094d\u091f\u093e\u0930\u094d\u091f \u092c\u093f\u0939\u0947\u0935 \u0907\u0928 \u0932\u093e\u0907\u0915 \u0930\u093f\u091a \u0930\u093f\u0938\u094d\u092a\u0949\u0928\u094d\u0938\u093f\u092c\u0932 \u0910\u0902\u0921 \u092e\u0948\u091a\u094d\u092f\u094b\u0930 \u092a\u0930\u094d\u0938\u0928 \u0906\u092a \u090f\u0915 \u0932\u0940\u0921\u0930 \u0915\u0940 \u0924\u0930\u0939 \u0916\u095c\u0947 \u0939\u094b\u0928\u0947 \u0914\u0930 \u092c\u093e\u0924 \u0915\u0930\u0928\u0947 \u0932\u0917\u0924\u0947 \u0939\u094b\u0964 \u0925\u093f\u0938 \u0907\u0938 \u0928\u0949\u091f \u092b\u0947\u0915 \u0907\u091f \u091f\u093f\u0932 \u092f\u0942 \u092e\u0947\u0915 \u0907\u091f \u0925\u093f\u0938 \u0907\u0938 \u092e\u0947\u0915\u093f\u0902\u0917 \u0907\u091f \u0907\u0902\u091f\u0930\u0928\u0932\u0940 \u092c\u093f\u092b\u094b\u0930 \u092f\u0942 \u092e\u0947\u0915 \u0907\u091f \u090f\u0915\u094d\u0938\u091f\u0930\u094d\u0928\u0932 \u0932\u0940 \u0907\u0902\u091f\u0930\u092a\u094d\u0930\u093f\u091f\u0930 \u092e\u093e\u0907\u0902\u0921\u0938\u0947\u091f \u092c\u093f\u092b\u094b\u0930 \u092f\u0942 \u0910\u0915\u094d\u091a\u0942\u0905\u0932\u0940 \u092c\u093f\u0915\u092e \u0930\u093f\u091c \u0907\u0928\u094d\u091f\u0930\u092a\u094d\u0930\u0947\u091f\u0930 \u0935\u093e\u0907\u091c \u092e\u093e\u0907\u0902\u0921\u0938\u0947\u091f \u092c\u093f\u092b\u094b\u0930 \u092f\u0942 \u090f\u0915\u094d\u091a\u0941\u0905\u0932\u0940 \u092c\u093f\u0915\u092e \u0935\u093e\u0907\u095b \u091c\u094b \u0905\u092c\u0948\u0928\u094d\u0921\u0928 \u0938\u0947 \u092f\u093e \u092b\u093f\u0930 \u0938\u0915\u094d\u0938\u0947\u0938 \u0924\u0941\u092e \u0905\u092a\u0928\u0947 \u092c\u093e\u0939\u0930 \u0926\u0947\u0916\u0928\u093e \u091a\u093e\u0939\u0924\u0947 \u0939\u094b? \u092a\u0939\u0932\u0947 \u0909\u0938\u0947 \u0905\u092a\u0928\u0947 \u0905\u0902\u0926\u0930 \u090f\u0915 \u092a\u0930\u092e\u093e\u0928\u0947\u0902\u091f \u091c\u0917\u0939 \u0926\u094b \u0928\u0902\u092c\u0930 \u091f\u0942 \u091f\u0947\u0915 \u0930\u093f\u0938\u094d\u092a\u0949\u0928\u094d\u0938\u093f\u092c\u093f\u0932\u093f\u091f\u0940 \u0917\u094b\u0932\u094d\u0938 \u0914\u0930 \u090f\u0915\u094d\u0938\u094d\u092a\u094d\u0930\u0947\u0936\u0928\u094d\u0938 \u0924\u094b \u0938\u092c\u0915\u0947 \u0939\u094b\u0924\u0947 \u0939\u0948\u0902 \u0932\u0947\u0915\u093f\u0928 \u0909\u0928 \u0917\u094b\u0932\u094d\u0938 \u092a\u0930 \u0910\u0915\u094d\u0936\u0928 \u0932\u0947\u0928\u093e \u0914\u0930 \u0905\u092a\u0928\u0940 \u0932\u093e\u0907\u092b \u0915\u094b \u0905\u092a\u0928\u0947 \u0917\u094b\u0932\u094d\u0938 \u0915\u0947 \u0905\u0930\u093e\u0909\u0902\u0921 \u0911\u0930\u094d\u0917\u0928\u093e\u0907\u091c \u0915\u0930\u0928\u093e \u092f\u0947 \u092c\u0924\u093e\u0924\u093e \u0939\u0948 \u0915\u093f \u0906\u092a \u0905\u092a\u0928\u0940 \u0932\u093e\u0907\u092b \u0915\u0940 \u0915\u093f\u0924\u0928\u0940 \u0930\u093f\u0938\u094d\u092a\u0949\u0928\u094d\u0938\u093f\u092c\u093f\u0932\u093f\u091f\u0940 \u0932\u0947 \u0930\u0939\u0947 \u0939\u094b\u0964 \u092a\u0930\u094d\u0938\u0928\u0932 \u0930\u093f\u0938\u094d\u092a\u0949\u0928\u094d\u0938\u093f\u092c\u093f\u0932\u093f\u091f\u0940 \u0932\u093e\u0907\u092b \u0915\u094b \u092c\u0939\u0941\u0924 \u0939\u0940 \u0906\u0938\u093e\u0928 \u0914\u0930 \u0938\u093f\u0902\u092a\u0932 \u092c\u0928\u093e \u0926\u0947\u0924\u0940 \u0939\u0948, \u0915\u094d\u092f\u094b\u0902\u0915\u093f \u090f\u0915 \u0930\u093f\u0938\u094d\u092a\u0949\u0928\u094d\u0938\u093f\u092c\u0932 \u0907\u0902\u0938\u093e\u0928 \u0915\u094b \u092a\u0924\u093e \u0939\u094b\u0924\u093e \u0939\u0948\u0964 \u0915\u093f \u0909\u0938\u0915\u0940 \u0932\u093e\u0907\u092b \u0909\u0938\u0915\u0947 \u0915\u0902\u091f\u094d\u0930\u094b\u0932 \u092e\u0947\u0902 \u0939\u0948 \u0914\u0930 \u0935\u094b \u091a\u093e\u0939\u0947 \u091c\u093f\u0924\u0928\u0940 \u092c\u093e\u0930 \u092d\u0940 \u0939\u093e\u0930\u0947 \u092f\u093e \u092b\u093f\u0930 \u0915\u093f\u0924\u0928\u0940 \u092d\u0940 \u092c\u095c\u0940 \u092e\u0941\u0936\u094d\u0915\u093f\u0932 \u092e\u0947\u0902 \u092b\u0902\u0938 \u091c\u093e\u090f, \u0935\u094b \u092b\u093f\u0930 \u092d\u0940 \u0915\u094b\u0908 \u0928\u093e \u0915\u094b\u0908 \u091c\u0941\u0917\u093e\u095c \u0932\u0917\u093e\u0915\u0930 \u091c\u0940\u0924 \u0938\u0915\u0924\u093e \u0939\u0948\u0964 \u0930\u093f\u0938\u094d\u092a\u0949\u0928\u094d\u0938\u093f\u092c\u093f\u0932\u093f\u091f\u0940 \u0915\u094b\u0908 \u092c\u094b\u091d \u0928\u0939\u0940\u0902 \u0939\u0948, \u092c\u0932\u094d\u0915\u093f \u092f\u0939 \u090f\u0915 \u0924\u0930\u0940\u0915\u093e \u0939\u0948 \u091c\u093f\u0938\u0938\u0947 \u0906\u092a \u0926\u0942\u0938\u0930\u094b\u0902 \u0915\u0947 \u090a\u092a\u0930 \u092c\u094b\u091d \u0928\u0939\u0940\u0902 \u092c\u0928\u0924\u0947 \u0939\u0948\u0902 \u0914\u0930 \u0905\u092a\u0928\u0940 \u091c\u093f\u0902\u0926\u0917\u0940 \u0915\u094b \u0915\u0932\u0947 \u0915\u0940 \u0924\u0930\u0939 \u0905\u092a\u0928\u093e \u092e\u0928\u091a\u093e\u0939\u093e \u0906\u0915\u093e\u0930 \u0926\u0947 \u092a\u093e\u0924\u0947 \u0939\u094b\u0964 \u0928\u0902\u092c\u0930 \u0925\u094d\u0930\u0940 \u092b\u093e\u0907\u0928\u094d\u0921 \u092e\u0940\u0928\u093f\u0902\u0917 \u0907\u0928 \u0932\u093e\u0907\u092b \u091c\u093f\u0902\u0926\u0917\u0940 \u092e\u0940\u0928\u093f\u0902\u0917 \u0915\u0947 \u092c\u093f\u0928\u093e \u0905\u0927\u0942\u0930\u0940 \u0914\u0930 \u092c\u0939\u0941\u0924 \u092e\u0941\u0936\u094d\u0915\u093f\u0932 \u0932\u0917\u0924\u0940 \u0939\u0948\u0964 \u092e\u0940\u0928\u093f\u0902\u0917 \u0935\u094b \u091a\u092e\u0915 \u0939\u094b\u0924\u0940 \u0939\u0948 \u091c\u094b \u0906\u092a\u0915\u094b \u091c\u093f\u0902\u0926\u0917\u0940 \u0915\u0947 \u0905\u0902\u0927\u0947\u0930\u0947 \u092e\u0947\u0902 \u0930\u093e\u0938\u094d\u0924\u093e \u0926\u093f\u0916\u093e\u0924\u0940 \u0939\u0948\u0964 \u0906\u092a\u0915\u094b \u092a\u0924\u093e \u0939\u094b\u0924\u093e \u0939\u0948 \u0915\u093f \u0920\u0940\u0915 \u0939\u0948, \u092e\u0948\u0902 \u0907\u0924\u0928\u0940 \u091c\u094d\u092f\u093e\u0926\u093e \u092e\u0941\u0936\u094d\u0915\u093f\u0932\u094b\u0902 \u0915\u094b \u091d\u0947\u0932 \u0930\u0939\u093e \u0939\u0942\u0901\u0964 \u0938\u092b\u0930 \u0915\u0930 \u0930\u0939\u093e \u0939\u0942\u0901 \u092c\u091f \u092f\u0947 \u0938\u092c \u090f\u0915 \u092c\u095c\u0947 \u0915\u0949\u0938\u094d\u091f \u0915\u0947 \u0932\u093f\u090f \u0939\u0948\u0964 \u0907\u0902\u0938\u093e\u0928 \u0915\u094b \u0915\u0937\u094d\u091f \u0909\u0924\u0928\u0947 \u0924\u0902\u0917 \u0928\u0939\u0940\u0902 \u0915\u0930\u0924\u0947 \u091c\u093f\u0924\u0928\u093e \u0909\u0928\u094d\u0939\u0947\u0902 \u0930\u093f\u092f\u0932\u093e\u0907\u091c\u0947\u0936\u0928 \u0924\u0902\u0917 \u0915\u0930\u0924\u0940 \u0939\u0948 \u0915\u093f \u0935\u094b \u092c\u093f\u0928\u093e \u0915\u093f\u0938\u0940 \u0935\u091c\u0939 \u0915\u0947 \u0907\u0928 \u0915\u0937\u094d\u091f\u094b\u0902 \u0915\u094b \u091d\u0947\u0932 \u0930\u0939\u0947 \u0939\u0948\u0902\u0964 \u0907\u0938\u0932\u093f\u090f \u0905\u092a\u0928\u0947 \u091c\u0940\u0928\u0947 \u0915\u0940 \u0935\u091c\u0939 \u0922\u0942\u0902\u0922\u094b\u0964 \u0906\u092a\u0915\u093e \u0932\u093e\u0907\u092b \u092a\u0930\u094d\u092a\u0938 \u0939\u0948 \u0905\u092a\u0928\u0940 \u0939\u093e\u0907\u090f\u0938\u094d\u091f \u092a\u094b\u091f\u0947\u0902\u0936\u093f\u0905\u0932 \u0924\u0915 \u092a\u0939\u0941\u0902\u091a\u0928\u093e, \u0932\u0947\u0915\u093f\u0928 \u0906\u092a \u092f\u0947 \u0915\u093e\u092e \u0915\u0948\u0938\u0947 \u0915\u0930\u094b\u0917\u0947? \u092f\u0939 \u095e\u093f\u0917\u0930 \u0906\u0909\u091f \u0915\u0930\u094b \u0928\u0902\u092c\u0930 \u095e\u094b\u0930 \u091f\u0947\u0915 \u092f\u0941\u0905\u0930 \u092b\u094d\u0930\u0947\u0928\u094d\u0921\u0938 \u090f\u0902\u0921 \u092b\u0948\u092e\u093f\u0932\u0940 \u091f\u0942 \u0926 \u092c\u0948\u091f\u0932 \u0935\u093f\u0926 \u092f\u0942 \u0906\u092a \u092e\u0947\u0902 \u0938\u0947 \u092c\u0939\u0941\u0924 \u0938\u0947 \u0932\u094b\u0917 \u0905\u092a\u0928\u0940 \u092b\u0948\u092e\u093f\u0932\u0940 \u0914\u0930 \u0907\u0935\u0928 \u0905\u092a\u0928\u0947 \u092b\u094d\u0930\u0947\u0902\u0921\u094d\u0938 \u0915\u0947 \u091c\u094d\u092f\u093e\u0926\u093e \u0915\u094d\u0932\u094b\u095b \u0928\u0939\u0940\u0902 \u0939\u0948, \u091c\u093f\u0938\u0915\u0940 \u0935\u091c\u0939 \u0938\u0947 \u0939\u0930 \u0907\u0902\u0938\u093e\u0928 \u0915\u0940 \u0936\u0915\u094d\u0924\u093f \u090f\u0915 \u0939\u0940 \u091c\u0917\u0939 \u092a\u0930 \u0915\u0949\u0928\u094d\u0938\u0928\u094d\u091f\u094d\u0930\u0947\u091f \u0939\u094b\u0928\u0947 \u0915\u0947 \u092c\u091c\u093e\u092f \u0906\u092a \u0932\u094b\u0917\u094b\u0902 \u0915\u0940 \u0936\u0915\u094d\u0924\u093f \u0907\u0927\u0930 \u0909\u0927\u0930 \u092c\u093f\u0916\u0930\u0940 \u0939\u0941\u0908 \u0939\u0948 \u0914\u0930 \u0906\u092a \u0915\u093f\u0938\u0940 \u092c\u095c\u0947 \u0917\u094b\u0932 \u092a\u0930 \u0915\u093e\u092e \u0928\u0939\u0940\u0902 \u0915\u0930 \u0938\u0915\u0924\u0947 \u0939\u0948\u0902\u0964 \u091f\u0940\u092e \u090f\u092b\u0930\u094d\u091f \u092e\u0947\u0902 \u0907\u0924\u0928\u0940 \u0924\u093e\u0915\u0924 \u0939\u094b\u0924\u0940 \u0939\u0948 \u0915\u093f \u0935\u0939 \u090f\u0915 \u092a\u0939\u093e\u095c \u0915\u094b \u092d\u0940 \u0939\u093f\u0932\u093e \u0926\u0947\u0964 \u091c\u093f\u0924\u0928\u093e \u092c\u095c\u093e \u0917\u094b\u0932 \u0939\u094b\u0917\u093e \u0909\u0924\u0928\u093e \u0939\u0940 \u0906\u092a\u0915\u094b \u0926\u0942\u0938\u0930\u094b\u0902 \u0915\u093e \u0938\u093e\u0925 \u091a\u093e\u0939\u093f\u090f \u0939\u094b\u0917\u093e, \u091c\u092c\u0915\u093f \u091b\u094b\u091f\u0947 \u0917\u094b\u095b \u0924\u094b \u0906\u092a \u0905\u0915\u0947\u0932\u0947 \u0939\u0940 \u0905\u091a\u0940\u0935 \u0915\u0930 \u0932\u094b\u0917\u0947\u0964 \u0932\u0947\u0915\u093f\u0928 \u091b\u094b\u091f\u0947 \u0917\u094b\u0932\u094d\u0938 \u0915\u0947 \u0938\u093e\u0925 \u092a\u094d\u0930\u0949\u092c\u094d\u0932\u092e \u092f\u0947 \u0939\u094b\u0924\u0940 \u0939\u0948 \u0915\u093f \u0935\u094b \u0906\u092a\u0915\u094b \u0939\u092e\u0947\u0936\u093e \u0939\u0940 \u092e\u0940\u0921\u093f\u092f\u093e \u0939\u094b\u0915\u0930 \u092c\u0928\u093e\u090f \u0930\u0916\u0924\u0947 \u0939\u0948\u0902\u0964 \u0906\u092a\u0915\u094b \u092d\u0940 \u0932\u093e\u0907\u092b \u0915\u093e \u092e\u095b\u093e \u0938\u093f\u0930\u094d\u092b \u0924\u092d\u0940 \u0906\u090f\u0917\u093e \u091c\u092c \u0906\u092a \u090f\u0915 \u0905\u0928\u0930\u093f\u092f\u0932\u093f\u0938\u094d\u091f\u093f\u0915 \u0914\u0930 \u092a\u093e\u0917\u0932\u094b\u0902 \u0935\u093e\u0932\u0940 \u091a\u0940\u095b \u0915\u093e \u092a\u0940\u091b\u093e \u0915\u0930\u094b\u0917\u0947\u0964 \u091c\u092c \u0924\u0915 \u0906\u092a \u0916\u0941\u0926 \u0915\u094b \u0905\u092a\u0928\u0940 \u0932\u093f\u092e\u093f\u091f\u094d\u0938 \u0915\u0947 \u092a\u093e\u0938 \u0938\u094d\u091f\u094d\u0930\u0947\u091a \u0928\u0939\u0940\u0902 \u0915\u0930\u094b\u0917\u0947, \u0924\u092c \u0924\u0915 \u0906\u092a\u0915\u094b \u092f\u0939 \u092b\u0940\u0932 \u0939\u0940 \u0928\u0939\u0940\u0902 \u0939\u094b\u0917\u093e\u0964 \u0915\u094d\u092f\u093e \u0906\u092a \u091c\u093f\u0928\u094d\u0926\u093e \u0939\u094b \u0914\u0930 \u0906\u092a \u0905\u092a\u0928\u0940 \u0938\u094b\u091a \u0938\u0947 \u091c\u094d\u092f\u093e\u0926\u093e \u092a\u0949\u0935\u0930\u092b\u0941\u0932 \u0939\u094b? \u0907\u0938\u0932\u093f\u090f \u0905\u092a\u0928\u0947 \u092b\u094d\u0930\u0947\u0902\u0921\u094d\u0938 \u0914\u0930 \u092b\u0948\u092e\u093f\u0932\u0940 \u0915\u094b\u0938\u092e\u0947 \u092a\u0947\u091c \u092a\u0930 \u0932\u093e\u0913 \u0909\u0928\u0938\u0947 \u0917\u094b\u0932\u094d\u0938, \u092a\u0930\u094d\u092a\u0938 \u0914\u0930 \u092e\u0940\u0928\u093f\u0902\u0917 \u0915\u0940 \u092c\u093e\u0924\u0947\u0902 \u0915\u0930\u094b\u0964 \u092a\u0939\u0932\u0947 \u0916\u0941\u0926 \u0915\u0940 \u0914\u0930 \u092b\u093f\u0930 \u0909\u0928\u0915\u0940 \u091f\u094d\u0930\u093e\u0902\u0938\u092b\u0949\u0930\u094d\u092e\u0947\u0936\u0928 \u092a\u0930 \u0927\u094d\u092f\u093e\u0928 \u0926\u094b \u0914\u0930 \u0938\u093e\u0925 \u092e\u0947\u0902 \u0917\u094d\u0930\u094b \u0915\u0930\u0928\u0947 \u0915\u093e \u092a\u094d\u0932\u093e\u0928 \u092c\u0928\u093e \u0939\u094b\u0964 \u0928\u0902\u092c\u0930 \u092b\u093e\u0907\u0935 \u092c\u093f\u0915\u092e \u092c\u093f\u0917\u0930 \u0926\u0947\u0928 \u092f\u0942 \u0911\u092b\u093f\u0930\u094d\u0938 \u0924\u0941\u092e\u094d\u0939\u0947 \u0932\u093e\u0907\u092b \u0915\u093e\u092b\u0940 \u092e\u0948\u0928\u0947\u091c\u0947\u092c\u0932 \u0914\u0930 \u0906\u0938\u093e\u0928 \u0932\u0917\u0928\u0947 \u0932\u0917\u0947 \u0917\u0940 \u091c\u092c \u0924\u0941\u092e \u092f\u0947 \u0915\u0930\u094b\u0917\u0947 \u0915\u093f \u0924\u0941\u092e \u0905\u092a\u0928\u0947 \u0921\u0930\u094b \u0938\u0947 \u091c\u094d\u092f\u093e\u0926\u093e \u092c\u095c\u0947 \u0939\u094b, \u0939\u0930 \u0921\u0930 \u092e\u094c\u0924 \u0938\u0947 \u0939\u0940 \u091c\u0941\u095c\u093e \u0939\u094b\u0924\u093e \u0939\u0948 \u0914\u0930 \u0921\u0930 \u0915\u094b \u0913\u0935\u0930 \u0915\u092e \u0915\u0930\u0928\u0947 \u0915\u093e \u092e\u0924\u0932\u092c \u0939\u094b\u0924\u093e \u0939\u0948 \u092e\u094c\u0924 \u0915\u094b \u0913\u0935\u0930 \u0915\u092e \u0915\u0930\u0928\u093e\u0964 \u0905\u092c \u0939\u092e \u0905\u092a\u0928\u0947 \u0936\u0930\u0940\u0930 \u0915\u094b \u0924\u094b \u092e\u0930\u0928\u0947 \u0938\u0947 \u0930\u094b\u0915 \u0928\u0939\u0940\u0902 \u0938\u0915\u0924\u0947 \u0932\u0947\u0915\u093f\u0928 \u0939\u092e \u092f\u0947 \u091c\u0930\u0942\u0930 \u0938\u092e\u091d \u0938\u0915\u0924\u0947 \u0939\u0948\u0902 \u0915\u093f \u091c\u0940\u0938 \u0907\u0928\u094d\u091f\u0947\u0932\u093f\u091c\u0947\u0928\u094d\u0938 \u0928\u0947 \u0939\u092e\u0947\u0902 \u092c\u0928\u093e\u092f\u093e \u0939\u0948\u0964 \u0935\u094b \u0915\u092d\u0940 \u092e\u0930 \u092f\u093e \u092b\u093f\u0930 \u092a\u0948\u0926\u093e \u0939\u094b \u0939\u0940 \u0928\u0939\u0940\u0902 \u0938\u0915\u0924\u0940\u0964 \u0935\u094b \u0905\u0928\u0928\u094d\u0924 \u0914\u0930 \u0936\u093e\u0936\u094d\u0935\u0924 \u0939\u0948\u0964 \u0905\u0917\u0930 \u0906\u092a \u0907\u0938 \u092c\u093e\u0930\u0947 \u092e\u0947\u0902 \u0905\u092a\u0928\u0940 \u0908\u0917\u094b \u0928\u093e\u092e \u0915\u0947 \u0924\u093f\u0928\u0915\u0947 \u0915\u093e \u0938\u0939\u093e\u0930\u093e \u0932\u0947\u0928\u0947 \u0915\u0940 \u0915\u094b\u0936\u093f\u0936 \u0915\u0930\u094b\u0917\u0947 \u0924\u094b \u0906\u092a\u0915\u093e \u0921\u0942\u092c\u0928\u093e \u0924\u092f \u0939\u0948\u0964 \u0907\u0938\u0932\u093f\u090f \u0915\u092d\u0940 \u092d\u0940 \u0916\u0941\u0926 \u0915\u094b \u0905\u092a\u0928\u0947 \u0936\u0930\u0940\u0930 \u092f\u093e \u092b\u093f\u0930 \u092e\u0928\u0938\u0947 \u0906\u0907\u0921\u0947\u0902\u091f\u093f\u092b\u093e\u0907 \u092e\u0924 \u0915\u0930\u094b\u0964 \u0916\u0941\u0926 \u0915\u0940 \u0906\u0907\u0921\u0947\u0902\u091f\u093f\u091f\u0940 \u092e\u0947\u0902 \u091b\u0941\u092a\u0947 \u091d\u0942\u0920 \u0915\u094b \u0926\u0947\u0916\u094b\u0964 \u091c\u093f\u0924\u0928\u093e \u0906\u092a \u0905\u092a\u0928\u0940 \u0908\u0917\u094b \u0915\u094b \u0938\u0940\u0930\u093f\u092f\u0938\u0932\u0940 \u0932\u094b\u0917\u0947, \u0909\u0924\u0928\u093e \u0939\u0940 \u0906\u092a \u0921\u0930, \u0932\u093e\u0932\u091a, \u092e\u094b\u0939, \u092a\u0940\u095c\u093e \u0914\u0930 \u0917\u0941\u0938\u094d\u0938\u0947 \u091c\u0948\u0938\u0940 \u091a\u0940\u091c\u094b\u0902 \u092e\u0947\u0902 \u0930\u0939\u094b\u0917\u0947 \u0913\u0935\u0930 \u0915\u092e\u093f\u0902\u0917 \u092b\u093f\u0932\u094d\u0938 \u0907\u0938\u0915\u093e \u0938\u094d\u092a\u093f\u0930\u093f\u091a\u0941\u0905\u0932 \u091f\u093e\u0938\u094d\u0915 \u0928\u0949\u091f \u0938\u093e\u0907\u0915\u094b\u0932\u0949\u091c\u093f\u0915\u0932 \u0906\u092a\u0915\u093e \u0926\u093f\u092e\u093e\u0917 \u0914\u0930 \u0936\u0930\u0940\u0930 \u092a\u094d\u0930\u094b\u0917\u094d\u0930\u093e\u092e\u094d\u0921 \u0939\u0948, \u0921\u0930\u094b \u0938\u0947 \u0926\u0942\u0930 \u092d\u093e\u0917\u0928\u0947 \u0915\u0947 \u0932\u093f\u090f \u0939\u0948 \u0915\u094d\u092f\u094b\u0902\u0915\u093f \u092e\u094c\u0924 \u0906\u092a\u0915\u0947 \u0936\u0930\u0940\u0930 \u0914\u0930 \u092e\u0928 \u0938\u0947 \u092c\u095d\u0940 \u0939\u0948\u0964 \u0905\u092a\u0928\u0947 \u0905\u0902\u0926\u0930 \u0909\u0938 \u0924\u093e\u0915\u0924 \u0915\u094b \u0922\u0942\u0902\u0922\u094b \u091c\u094b \u092e\u094c\u0924 \u0938\u0947 \u092d\u0940 \u092c\u095c\u0940 \u0939\u0948 \u0914\u0930 \u0924\u092c \u0924\u0941\u092e \u092c\u094b\u0932\u094b\u0917\u0947 \u0915\u093f \u091c\u093f\u0902\u0926\u0917\u0940 \u0909\u0924\u0928\u0940 \u092e\u0941\u0936\u094d\u0915\u093f\u0932 \u0914\u0930 \u0926\u0941\u0916\u0926\u093e\u092f\u0940 \u0928\u0939\u0940\u0902 \u0939\u0948 \u091c\u093f\u0924\u0928\u093e \u0939\u092e \u0909\u0938\u0947 \u092c\u0928\u093e \u0926\u0947\u0924\u0947 \u0939\u0948\u0902\u0964 \u0938\u094d\u091f\u0949\u092a \u092e\u0947\u0915\u093f\u0902\u0917 \u092f\u094b\u0930 \u0932\u093e\u0907\u092b \u0905\u0928\u0928\u0947\u0938\u0947\u0938\u0930\u0940 \u0932\u0940 \u0939\u093e\u0930\u094d\u0921 \u0924\u0941\u092e\u094d\u0939\u093e\u0930\u0940 \u091c\u093f\u0902\u0926\u0917\u0940 \u0915\u092d\u0940 \u0928\u093e \u0905\u0902\u0924 \u0939\u094b\u0928\u0947 \u0935\u093e\u0932\u0940 \u0932\u0940\u0932\u093e \u0915\u093e \u0939\u093f\u0938\u094d\u0938\u093e \u0939\u0948\u0964 \u0907\u0938\u0915\u093e \u0930\u0938 \u0932\u094b \u0914\u0930 \u091c\u093f\u0924\u0928\u093e \u0939\u094b \u0938\u0915\u0947 \u0909\u0924\u0928\u093e \u0907\u0938 \u0926\u0941\u0928\u093f\u092f\u093e \u092e\u0947\u0902 \u092c\u0926\u0932\u093e\u0935 \u0932\u093e\u0913\u0964 \u092b\u0941\u091f \u092a\u0930 \u0914\u0930 \u0916\u0941\u0926 \u0915\u0940 \u092a\u094b\u091f\u0947\u0902\u0936\u093f\u0905\u0932 \u092a\u0930 \u0915\u093e\u092e \u0915\u0930\u0915\u0947 \u0932\u093e\u0907\u092b \u0907\u095b \u0930\u093f\u092f\u0932\u0940 \u0939\u093e\u0930\u094d\u0921 \u092b\u0949\u0930 \u092f\u0942 \u092c\u091f \u0907\u091f \u0907\u0938 \u0935\u0947\u0907\u091f\u094d \u092b\u0949\u0930 \u0926\u094b\u095b \u0939\u0942\u0901 \u0914\u0930 \u0932\u093f\u0935\u093f\u0902\u0917 \u0905\u092a \u091f\u0941 \u0926\u0947\u0930 \u092a\u094b\u091f\u0947\u0902\u0936\u093f\u0905\u0932\u0964",
"target": "\u0906\u0938\u093e\u0928 \u0939\u094b \u092f\u093e \u092e\u0941\u0936\u094d\u0915\u093f\u0932, \u092f\u0947 \u0936\u092c\u094d\u0926 \u091c\u094d\u092f\u093e\u0926\u093e\u0924\u0930 \u0932\u094b\u0917\u094b\u0902 \u0915\u0947 \u0932\u093f\u090f \u0938\u093e\u092a\u0947\u0915\u094d\u0937 \u0939\u094b\u0924\u0947 \u0939\u0948\u0902\u0964 \u091c\u094b \u0906\u092a\u0915\u0947 \u0932\u093f\u090f \u0906\u0938\u093e\u0928 \u0939\u0948 \u0935\u0939 \u0915\u093f\u0938\u0940 \u0914\u0930 \u0915\u0947 \u0932\u093f\u090f \u092e\u0941\u0936\u094d\u0915\u093f\u0932 \u0939\u094b \u0938\u0915\u0924\u093e \u0939\u0948 \u0914\u0930 \u0907\u0938\u0915\u0947 \u0935\u093f\u092a\u0930\u0940\u0924\u0964 \u0915\u093f\u0938\u0940 \u0915\u093e\u0930\u094d\u092f \u0915\u0940 \u0915\u0920\u093f\u0928\u093e\u0908 \u0909\u0938\u0947 \u0915\u0930\u0928\u0947 \u0935\u093e\u0932\u0947 \u0935\u094d\u092f\u0915\u094d\u0924\u093f \u0915\u0940 \u0915\u094d\u0937\u092e\u0924\u093e \u092a\u0930 \u0928\u093f\u0930\u094d\u092d\u0930 \u0915\u0930\u0924\u0940 \u0939\u0948\u0964 \u090f\u0915 \u0916\u093e\u0938 \u092e\u093e\u0928\u0938\u093f\u0915\u0924\u093e \u0930\u0916\u0928\u0947 \u0935\u093e\u0932\u094b\u0902 \u0915\u0947 \u0932\u093f\u090f \u091c\u0940\u0935\u0928 \u0906\u0938\u093e\u0928 \u0939\u094b\u0924\u093e \u0939\u0948 \u0914\u0930 \u0907\u0938 \u092a\u0949\u0921\u0915\u093e\u0938\u094d\u091f \u0938\u0947\u0917\u092e\u0947\u0902\u091f \u092e\u0947\u0902 \u0939\u092e \u0910\u0938\u0940 \u0939\u0940 \u092e\u093e\u0928\u0938\u093f\u0915\u0924\u093e \u0915\u0947 \u092c\u093e\u0930\u0947 \u092e\u0947\u0902 \u092c\u093e\u0924 \u0915\u0930\u0947\u0902\u0917\u0947\u0964"
}
]
```
### Dataset Fields
The dataset has the following fields (also called "features"):
```json
{
"text": "Value(dtype='string', id=None)",
"target": "Value(dtype='string', id=None)"
}
```
### Dataset Splits
This dataset is split into a train and validation split. The split sizes are as follow:
| Split name | Num samples |
| ------------ | ------------------- |
| train | 451 |
| valid | 113 |
|
viditsorg/autotrain-data-mbart-finetune-hindi
|
[
"task_categories:summarization",
"region:us"
] |
2023-10-24T11:54:12+00:00
|
{"task_categories": ["summarization"]}
|
2023-10-24T11:56:07+00:00
|
[] |
[] |
TAGS
#task_categories-summarization #region-us
|
AutoTrain Dataset for project: mbart-finetune-hindi
===================================================
Dataset Description
-------------------
This dataset has been automatically processed by AutoTrain for project mbart-finetune-hindi.
### Languages
The BCP-47 code for the dataset's language is unk.
Dataset Structure
-----------------
### Data Instances
A sample from this dataset looks as follows:
### Dataset Fields
The dataset has the following fields (also called "features"):
### Dataset Splits
This dataset is split into a train and validation split. The split sizes are as follow:
|
[
"### Languages\n\n\nThe BCP-47 code for the dataset's language is unk.\n\n\nDataset Structure\n-----------------",
"### Data Instances\n\n\nA sample from this dataset looks as follows:",
"### Dataset Fields\n\n\nThe dataset has the following fields (also called \"features\"):",
"### Dataset Splits\n\n\nThis dataset is split into a train and validation split. The split sizes are as follow:"
] |
[
"TAGS\n#task_categories-summarization #region-us \n",
"### Languages\n\n\nThe BCP-47 code for the dataset's language is unk.\n\n\nDataset Structure\n-----------------",
"### Data Instances\n\n\nA sample from this dataset looks as follows:",
"### Dataset Fields\n\n\nThe dataset has the following fields (also called \"features\"):",
"### Dataset Splits\n\n\nThis dataset is split into a train and validation split. The split sizes are as follow:"
] |
[
16,
27,
17,
23,
27
] |
[
"passage: TAGS\n#task_categories-summarization #region-us \n### Languages\n\n\nThe BCP-47 code for the dataset's language is unk.\n\n\nDataset Structure\n-----------------### Data Instances\n\n\nA sample from this dataset looks as follows:### Dataset Fields\n\n\nThe dataset has the following fields (also called \"features\"):### Dataset Splits\n\n\nThis dataset is split into a train and validation split. The split sizes are as follow:"
] |
b39e09e45becd0e92c706556caad1f2da3f71944
|
Translating...
|
irlab-udc/sharegpt_galician
|
[
"license:apache-2.0",
"region:us"
] |
2023-10-24T12:17:19+00:00
|
{"license": "apache-2.0"}
|
2023-10-24T12:17:39+00:00
|
[] |
[] |
TAGS
#license-apache-2.0 #region-us
|
Translating...
|
[] |
[
"TAGS\n#license-apache-2.0 #region-us \n"
] |
[
14
] |
[
"passage: TAGS\n#license-apache-2.0 #region-us \n"
] |
ded661ae4f75547f35ddbe2857dd62870bca420d
|
# LTR
LTR -- Lingvanex Test References for MT Evaluation from English into a total of 30 target languages for a big variety of cases.
## TEST CASES
| Parameter | Description |
|-----------|-------------|
| Length | Sentences from 1 to 100 words. |
| Domain | Medicine (12%), Automobile (11%), Finance (8%) |
| Tokenizer | Jupiter is 1.000.000 km far. Ask Mr. Johnson for training |
| Tags | I want to eat <tag> and swim |
| Capitalisation (Case) | HELLO my Dear frIEND |
| Different languages in one text (Up to 3 languages) | I see "Купалинка" performance near the theater. |
| Styling | Hello Dude! |
| Errors (Grammar, OCR) | I neet (need) to buy a kat (cat)|
| Abbreviations | The model was named 15.BVcX-10 |
| Named Entities | Let’s go to New York city |
| Idioms | A piece of cake. Once in a blue moon’ |
| Formulas (Math, Physics, Chemistry) | Cr2(SO4)3 + CO2 + H2O + K2SO4 + KNO3 |
| Romanian numbers | It was in MCMXVII year |
| Unicode Special Characters | №%&*/# |
# Updates
Data package has been released on 11/18/2023. Enjoy!
# License
The LTR data set is released under the [CC BY-SA 4.0 license](https://huggingface.co/datasets/lingvanex/lingvanex_test_references/blob/main/LICENSE.md).
# How to Cite
```
@inproceedings{
title = "{LTR} Lingvanex Test References for {MT} Evaluation",
author = "Aliaksei Rudak",
url = "https://lingvanex.com",
Year = "2023"
}
```
|
lingvanex/lingvanex_test_references
|
[
"task_categories:translation",
"multilinguality:translation",
"size_categories:1K<n<10K",
"language:af",
"language:ar",
"language:be",
"language:bp",
"language:bs",
"language:cs",
"language:da",
"language:de",
"language:el",
"language:es",
"language:et",
"language:fa",
"language:fi",
"language:fr",
"language:he",
"language:hu",
"language:id",
"language:it",
"language:lt",
"language:lv",
"language:no",
"language:pl",
"language:pt",
"language:ro",
"language:ru",
"language:sq",
"language:sr",
"language:sv",
"language:tr",
"language:uk",
"license:cc-by-nc-sa-4.0",
"region:us"
] |
2023-10-24T12:21:32+00:00
|
{"language": ["af", "ar", "be", "bp", "bs", "cs", "da", "de", "el", "es", "et", "fa", "fi", "fr", "he", "hu", "id", "it", "lt", "lv", false, "pl", "pt", "ro", "ru", "sq", "sr", "sv", "tr", "uk"], "license": "cc-by-nc-sa-4.0", "multilinguality": ["translation"], "size_categories": ["1K<n<10K"], "task_categories": ["translation"], "task_ids": [], "configs": [{"config_name": "en-cs", "data_files": [{"split": "test", "path": "data/en-cs.csv"}]}, {"config_name": "en-et", "data_files": [{"split": "test", "path": "data/en-et.csv"}]}, {"config_name": "en-hu", "data_files": [{"split": "test", "path": "data/en-hu.csv"}]}, {"config_name": "en-no", "data_files": [{"split": "test", "path": "data/en-no.csv"}]}, {"config_name": "en-ru", "data_files": [{"split": "test", "path": "data/en-ru.csv"}]}, {"config_name": "en-uk", "data_files": [{"split": "test", "path": "data/en-uk.csv"}]}, {"config_name": "en-af", "data_files": [{"split": "test", "path": "data/en-af.csv"}]}, {"config_name": "en-da", "data_files": [{"split": "test", "path": "data/en-da.csv"}]}, {"config_name": "en-fa", "data_files": [{"split": "test", "path": "data/en-fa.csv"}]}, {"config_name": "en-id", "data_files": [{"split": "test", "path": "data/en-id.csv"}]}, {"config_name": "en-pl", "data_files": [{"split": "test", "path": "data/en-pl.csv"}]}, {"config_name": "en-sq", "data_files": [{"split": "test", "path": "data/en-sq.csv"}]}, {"config_name": "en-ar", "data_files": [{"split": "test", "path": "data/en-ar.csv"}]}, {"config_name": "en-de", "data_files": [{"split": "test", "path": "data/en-de.csv"}]}, {"config_name": "en-fi", "data_files": [{"split": "test", "path": "data/en-fi.csv"}]}, {"config_name": "en-it", "data_files": [{"split": "test", "path": "data/en-it.csv"}]}, {"config_name": "en-pt_br", "data_files": [{"split": "test", "path": "data/en-pt_br.csv"}]}, {"config_name": "en-sr", "data_files": [{"split": "test", "path": "data/en-sr.csv"}]}, {"config_name": "en-be", "data_files": [{"split": "test", "path": "data/en-be.csv"}]}, {"config_name": "en-el", "data_files": [{"split": "test", "path": "data/en-el.csv"}]}, {"config_name": "en-fr", "data_files": [{"split": "test", "path": "data/en-fr.csv"}]}, {"config_name": "en-lt", "data_files": [{"split": "test", "path": "data/en-lt.csv"}]}, {"config_name": "en-pt", "data_files": [{"split": "test", "path": "data/en-pt.csv"}]}, {"config_name": "en-sv", "data_files": [{"split": "test", "path": "data/en-sv.csv"}]}, {"config_name": "en-bs", "data_files": [{"split": "test", "path": "data/en-bs.csv"}]}, {"config_name": "en-es", "data_files": [{"split": "test", "path": "data/en-es.csv"}]}, {"config_name": "en-he", "data_files": [{"split": "test", "path": "data/en-he.csv"}]}, {"config_name": "en-lv", "data_files": [{"split": "test", "path": "data/en-lv.csv"}]}, {"config_name": "en-ro", "data_files": [{"split": "test", "path": "data/en-ro.csv"}]}, {"config_name": "en-tr", "data_files": [{"split": "test", "path": "data/en-tr.csv"}]}, {"config_name": "en-uk", "data_files": [{"split": "test", "path": "data/en-uk.csv"}]}]}
|
2023-10-24T12:48:54+00:00
|
[] |
[
"af",
"ar",
"be",
"bp",
"bs",
"cs",
"da",
"de",
"el",
"es",
"et",
"fa",
"fi",
"fr",
"he",
"hu",
"id",
"it",
"lt",
"lv",
"no",
"pl",
"pt",
"ro",
"ru",
"sq",
"sr",
"sv",
"tr",
"uk"
] |
TAGS
#task_categories-translation #multilinguality-translation #size_categories-1K<n<10K #language-Afrikaans #language-Arabic #language-Belarusian #language-bp #language-Bosnian #language-Czech #language-Danish #language-German #language-Modern Greek (1453-) #language-Spanish #language-Estonian #language-Persian #language-Finnish #language-French #language-Hebrew #language-Hungarian #language-Indonesian #language-Italian #language-Lithuanian #language-Latvian #language-Norwegian #language-Polish #language-Portuguese #language-Romanian #language-Russian #language-Albanian #language-Serbian #language-Swedish #language-Turkish #language-Ukrainian #license-cc-by-nc-sa-4.0 #region-us
|
LTR
===
LTR -- Lingvanex Test References for MT Evaluation from English into a total of 30 target languages for a big variety of cases.
TEST CASES
----------
Updates
=======
Data package has been released on 11/18/2023. Enjoy!
License
=======
The LTR data set is released under the CC BY-SA 4.0 license.
How to Cite
===========
|
[] |
[
"TAGS\n#task_categories-translation #multilinguality-translation #size_categories-1K<n<10K #language-Afrikaans #language-Arabic #language-Belarusian #language-bp #language-Bosnian #language-Czech #language-Danish #language-German #language-Modern Greek (1453-) #language-Spanish #language-Estonian #language-Persian #language-Finnish #language-French #language-Hebrew #language-Hungarian #language-Indonesian #language-Italian #language-Lithuanian #language-Latvian #language-Norwegian #language-Polish #language-Portuguese #language-Romanian #language-Russian #language-Albanian #language-Serbian #language-Swedish #language-Turkish #language-Ukrainian #license-cc-by-nc-sa-4.0 #region-us \n"
] |
[
216
] |
[
"passage: TAGS\n#task_categories-translation #multilinguality-translation #size_categories-1K<n<10K #language-Afrikaans #language-Arabic #language-Belarusian #language-bp #language-Bosnian #language-Czech #language-Danish #language-German #language-Modern Greek (1453-) #language-Spanish #language-Estonian #language-Persian #language-Finnish #language-French #language-Hebrew #language-Hungarian #language-Indonesian #language-Italian #language-Lithuanian #language-Latvian #language-Norwegian #language-Polish #language-Portuguese #language-Romanian #language-Russian #language-Albanian #language-Serbian #language-Swedish #language-Turkish #language-Ukrainian #license-cc-by-nc-sa-4.0 #region-us \n"
] |
eeb5a229dd58dfe75d720d33990ecbc2bc5c7c24
|
# Dataset Card for "SciCoT"
Combination of sciq, medmcqa, and pubmed_qa (human annotated part), with a maximum of 3k examples taken from each.
|
euclaise/SciCoT
|
[
"license:cc-by-nc-3.0",
"region:us"
] |
2023-10-24T12:25:49+00:00
|
{"license": "cc-by-nc-3.0", "dataset_info": {"features": [{"name": "rationale", "dtype": "string"}, {"name": "target", "dtype": "string"}, {"name": "source", "dtype": "string"}, {"name": "prompt", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 4559510, "num_examples": 7000}], "download_size": 2872385, "dataset_size": 4559510}}
|
2023-11-21T19:31:01+00:00
|
[] |
[] |
TAGS
#license-cc-by-nc-3.0 #region-us
|
# Dataset Card for "SciCoT"
Combination of sciq, medmcqa, and pubmed_qa (human annotated part), with a maximum of 3k examples taken from each.
|
[
"# Dataset Card for \"SciCoT\"\n\nCombination of sciq, medmcqa, and pubmed_qa (human annotated part), with a maximum of 3k examples taken from each."
] |
[
"TAGS\n#license-cc-by-nc-3.0 #region-us \n",
"# Dataset Card for \"SciCoT\"\n\nCombination of sciq, medmcqa, and pubmed_qa (human annotated part), with a maximum of 3k examples taken from each."
] |
[
17,
47
] |
[
"passage: TAGS\n#license-cc-by-nc-3.0 #region-us \n# Dataset Card for \"SciCoT\"\n\nCombination of sciq, medmcqa, and pubmed_qa (human annotated part), with a maximum of 3k examples taken from each."
] |
f17f342cb93265f6cff2e997d4742483f9ddfc0d
|
## Dataset Details
### Dataset Description
This dataset is the downscaled version of the [Food121](https://huggingface.co/datasets/ItsNotRohit/Food121) dataset. All images are downscaled to a maximum of 224*224.
This dataset is the combination of the [Food101](https://huggingface.co/datasets/food101), [Indian Food Classification](https://www.kaggle.com/datasets/l33tc0d3r/indian-food-classification) and [The-massive-Indian-Food-Dataset](https://www.kaggle.com/datasets/anshulmehtakaggl/themassiveindianfooddataset) datasets.
This Dataset aims to be a viable dataset for Image Classification of Foods with an added Indian context. This dataset has 121 classes with each class having 800 images in the train split and 200 images in the test split.
### Dataset Sources
- **Food101:** https://huggingface.co/datasets/food101
- **Indian Food Classification:** https://www.kaggle.com/datasets/l33tc0d3r/indian-food-classification
- **The-massive-Indian-Food-Dataset:** https://www.kaggle.com/datasets/anshulmehtakaggl/themassiveindianfooddataset
|
ItsNotRohit/Food121-224
|
[
"task_categories:image-classification",
"size_categories:10K<n<100K",
"language:en",
"license:mit",
"food101",
"image classification",
"region:us"
] |
2023-10-24T12:26:42+00:00
|
{"language": ["en"], "license": "mit", "size_categories": ["10K<n<100K"], "task_categories": ["image-classification"], "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "test", "path": "data/test-*"}]}], "dataset_info": {"features": [{"name": "image", "dtype": "image"}, {"name": "label", "dtype": {"class_label": {"names": {"0": "apple_pie", "1": "baby_back_ribs", "2": "baklava", "3": "beef_carpaccio", "4": "beef_tartare", "5": "beet_salad", "6": "beignets", "7": "bibimbap", "8": "biryani", "9": "bread_pudding", "10": "breakfast_burrito", "11": "bruschetta", "12": "caesar_salad", "13": "cannoli", "14": "caprese_salad", "15": "carrot_cake", "16": "ceviche", "17": "chai", "18": "chapati", "19": "cheese_plate", "20": "cheesecake", "21": "chicken_curry", "22": "chicken_quesadilla", "23": "chicken_wings", "24": "chocolate_cake", "25": "chocolate_mousse", "26": "chole_bhature", "27": "churros", "28": "clam_chowder", "29": "club_sandwich", "30": "crab_cakes", "31": "creme_brulee", "32": "croque_madame", "33": "cup_cakes", "34": "dabeli", "35": "dal", "36": "deviled_eggs", "37": "dhokla", "38": "donuts", "39": "dosa", "40": "dumplings", "41": "edamame", "42": "eggs_benedict", "43": "escargots", "44": "falafel", "45": "filet_mignon", "46": "fish_and_chips", "47": "foie_gras", "48": "french_fries", "49": "french_onion_soup", "50": "french_toast", "51": "fried_calamari", "52": "fried_rice", "53": "frozen_yogurt", "54": "garlic_bread", "55": "gnocchi", "56": "greek_salad", "57": "grilled_cheese_sandwich", "58": "grilled_salmon", "59": "guacamole", "60": "gyoza", "61": "hamburger", "62": "hot_and_sour_soup", "63": "hot_dog", "64": "huevos_rancheros", "65": "hummus", "66": "ice_cream", "67": "idli", "68": "jalebi", "69": "kathi_rolls", "70": "kofta", "71": "kulfi", "72": "lasagna", "73": "lobster_bisque", "74": "lobster_roll_sandwich", "75": "macaroni_and_cheese", "76": "macarons", "77": "miso_soup", "78": "momos", "79": "mussels", "80": "naan", "81": "nachos", "82": "omelette", "83": "onion_rings", "84": "oysters", "85": "pad_thai", "86": "paella", "87": "pakoda", "88": "pancakes", "89": "pani_puri", "90": "panna_cotta", "91": "panner_butter_masala", "92": "pav_bhaji", "93": "peking_duck", "94": "pho", "95": "pizza", "96": "pork_chop", "97": "poutine", "98": "prime_rib", "99": "pulled_pork_sandwich", "100": "ramen", "101": "ravioli", "102": "red_velvet_cake", "103": "risotto", "104": "samosa", "105": "sashimi", "106": "scallops", "107": "seaweed_salad", "108": "shrimp_and_grits", "109": "spaghetti_bolognese", "110": "spaghetti_carbonara", "111": "spring_rolls", "112": "steak", "113": "strawberry_shortcake", "114": "sushi", "115": "tacos", "116": "takoyaki", "117": "tiramisu", "118": "tuna_tartare", "119": "vadapav", "120": "waffles"}}}}], "splits": [{"name": "train", "num_bytes": 2004526002, "num_examples": 96800}, {"name": "test", "num_bytes": 513682668.4, "num_examples": 24200}], "download_size": 3295817653, "dataset_size": 2518208670.4}, "tags": ["food101", "image classification"]}
|
2023-10-28T06:04:35+00:00
|
[] |
[
"en"
] |
TAGS
#task_categories-image-classification #size_categories-10K<n<100K #language-English #license-mit #food101 #image classification #region-us
|
## Dataset Details
### Dataset Description
This dataset is the downscaled version of the Food121 dataset. All images are downscaled to a maximum of 224*224.
This dataset is the combination of the Food101, Indian Food Classification and The-massive-Indian-Food-Dataset datasets.
This Dataset aims to be a viable dataset for Image Classification of Foods with an added Indian context. This dataset has 121 classes with each class having 800 images in the train split and 200 images in the test split.
### Dataset Sources
- Food101: URL
- Indian Food Classification: URL
- The-massive-Indian-Food-Dataset: URL
|
[
"## Dataset Details",
"### Dataset Description\n\nThis dataset is the downscaled version of the Food121 dataset. All images are downscaled to a maximum of 224*224.\n\n\nThis dataset is the combination of the Food101, Indian Food Classification and The-massive-Indian-Food-Dataset datasets.\nThis Dataset aims to be a viable dataset for Image Classification of Foods with an added Indian context. This dataset has 121 classes with each class having 800 images in the train split and 200 images in the test split.",
"### Dataset Sources\n\n- Food101: URL\n- Indian Food Classification: URL\n- The-massive-Indian-Food-Dataset: URL"
] |
[
"TAGS\n#task_categories-image-classification #size_categories-10K<n<100K #language-English #license-mit #food101 #image classification #region-us \n",
"## Dataset Details",
"### Dataset Description\n\nThis dataset is the downscaled version of the Food121 dataset. All images are downscaled to a maximum of 224*224.\n\n\nThis dataset is the combination of the Food101, Indian Food Classification and The-massive-Indian-Food-Dataset datasets.\nThis Dataset aims to be a viable dataset for Image Classification of Foods with an added Indian context. This dataset has 121 classes with each class having 800 images in the train split and 200 images in the test split.",
"### Dataset Sources\n\n- Food101: URL\n- Indian Food Classification: URL\n- The-massive-Indian-Food-Dataset: URL"
] |
[
45,
4,
115,
33
] |
[
"passage: TAGS\n#task_categories-image-classification #size_categories-10K<n<100K #language-English #license-mit #food101 #image classification #region-us \n## Dataset Details### Dataset Description\n\nThis dataset is the downscaled version of the Food121 dataset. All images are downscaled to a maximum of 224*224.\n\n\nThis dataset is the combination of the Food101, Indian Food Classification and The-massive-Indian-Food-Dataset datasets.\nThis Dataset aims to be a viable dataset for Image Classification of Foods with an added Indian context. This dataset has 121 classes with each class having 800 images in the train split and 200 images in the test split.### Dataset Sources\n\n- Food101: URL\n- Indian Food Classification: URL\n- The-massive-Indian-Food-Dataset: URL"
] |
1fa1eca99c82ccb23ca0ede5458cfd7784d7a7f1
|
# Dataset Card for "Synthetic_Runyankole_MMS"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
mekaneeky/Synthetic_Runyankole_MMS
|
[
"region:us"
] |
2023-10-24T12:28:26+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "dev", "path": "data/dev-*"}, {"split": "test", "path": "data/test-*"}]}], "dataset_info": {"features": [{"name": "eng", "dtype": "string"}, {"name": "lug", "dtype": "string"}, {"name": "ach", "dtype": "string"}, {"name": "teo", "dtype": "string"}, {"name": "lgg", "dtype": "string"}, {"name": "nyn", "dtype": "string"}, {"name": "runyankole_synthetic_audio", "sequence": {"sequence": "float32"}}], "splits": [{"name": "train", "num_bytes": 12669819104, "num_examples": 23947}, {"name": "dev", "num_bytes": 268490157, "num_examples": 500}, {"name": "test", "num_bytes": 262253648, "num_examples": 500}], "download_size": 13209162621, "dataset_size": 13200562909}}
|
2023-10-24T12:41:37+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "Synthetic_Runyankole_MMS"
More Information needed
|
[
"# Dataset Card for \"Synthetic_Runyankole_MMS\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"Synthetic_Runyankole_MMS\"\n\nMore Information needed"
] |
[
6,
22
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"Synthetic_Runyankole_MMS\"\n\nMore Information needed"
] |
11eb2e697f12e88d90842095d847685f15cb925c
|
# Dataset Card for "encodec_24khz-opt-125m-pretrained-ft-librispeech_asr"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
cmu-mlsp/encodec_24khz-opt-125m-pretrained-ft-librispeech_asr
|
[
"region:us"
] |
2023-10-24T12:29:07+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "validation", "path": "data/validation-*"}, {"split": "test", "path": "data/test-*"}]}], "dataset_info": {"features": [{"name": "file", "dtype": "string"}, {"name": "audio", "dtype": {"audio": {"sampling_rate": 24000}}}, {"name": "text", "dtype": "string"}, {"name": "speaker_id", "dtype": "int64"}, {"name": "chapter_id", "dtype": "int64"}, {"name": "id", "dtype": "string"}, {"name": "audio_codes", "sequence": {"sequence": "int64"}}], "splits": [{"name": "train", "num_bytes": 17829358082.086, "num_examples": 28539}, {"name": "validation", "num_bytes": 955281891.125, "num_examples": 2703}, {"name": "test", "num_bytes": 958024726.5, "num_examples": 2620}], "download_size": 18905275151, "dataset_size": 19742664699.711}}
|
2023-10-24T12:39:34+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "encodec_24khz-opt-125m-pretrained-ft-librispeech_asr"
More Information needed
|
[
"# Dataset Card for \"encodec_24khz-opt-125m-pretrained-ft-librispeech_asr\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"encodec_24khz-opt-125m-pretrained-ft-librispeech_asr\"\n\nMore Information needed"
] |
[
6,
35
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"encodec_24khz-opt-125m-pretrained-ft-librispeech_asr\"\n\nMore Information needed"
] |
12d0d457032b4d725dcb596cda20c853a9af6943
|
# Dataset Card for "sv_corpora_parliament_processed"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
enakilci/sv_corpora_parliament_processed
|
[
"region:us"
] |
2023-10-24T12:40:22+00:00
|
{"dataset_info": {"features": [{"name": "text", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 292351437, "num_examples": 1892723}], "download_size": 0, "dataset_size": 292351437}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-31T10:45:38+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "sv_corpora_parliament_processed"
More Information needed
|
[
"# Dataset Card for \"sv_corpora_parliament_processed\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"sv_corpora_parliament_processed\"\n\nMore Information needed"
] |
[
6,
20
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"sv_corpora_parliament_processed\"\n\nMore Information needed"
] |
e41c41cb1750ec694bd226c583a8215c1950af10
|
# NDC Sector Classification
This dataset is built from the tagged NDC ([Climate Watch](https://www.climatewatchdata.org/data-explorer/historical-emissions?historical-emissions-data-sources=climate-watch&historical-emissions-gases=all-ghg&historical-emissions-regions=All%20Selected&historical-emissions-sectors=total-including-lucf%2Ctotal-including-lucf&page=1)) paragraphs made by [GIZ Data Service Center](https://www.giz.de/expertise/html/63018.html) and available on Hugging Face ([GIZ/policy_qa_v0](https://huggingface.co/datasets/GIZ/policy_qa_v0)).
The NDC urls have been taken from [IGES NDC Database](https://www.iges.or.jp/en/pub/iges-indc-ndc-database/en).
Each NDC have been classified in a specific sector if it contains at least a paragraph classified as the specific sector. Each NDC can be associated to multiple sector.
The dataset contains 250 document classified in 18 sectors. The followin plot shows the number of documents tagged as each sector.
## NDC Data
The csv containing the tagged NDC is structured as follows:
- `Country`: The country to which the NDC refers.
- `Document`: The type of document (INDC, First NDC, Second NDC).
- `Language`: The original language of the NDC.
- `Sector`: A json whose keys represent the sectors mentioned in the NDC and whose values represent the number of paragraphs that mention the specific secotor.
- `URL`: The pdf url.
## Sector Data
The json containing the sectors' description follows the scheme below:
```json
{
"topic_list_id":"UUID",
"topics":[
{
"topic_id":"integer",
"topic_name":"string",
"definitions":[
{
"lang":"string",
"description":"string"
}
]
}
]
}
```
**Note:** The descriptions have been taken from Wikipedia (en). The Spanish version is a translation of the english one.
|
andreaponti/NDC-sectors
|
[
"task_categories:text-classification",
"size_categories:n<1K",
"language:en",
"language:es",
"climate",
"region:us"
] |
2023-10-24T12:43:19+00:00
|
{"language": ["en", "es"], "size_categories": ["n<1K"], "task_categories": ["text-classification"], "pretty_name": "NDC Sector Classification", "tags": ["climate"], "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "NDC_sectors.csv"}]}, {"config_name": "sector_description", "data_files": "sectors.json"}]}
|
2023-10-24T13:21:11+00:00
|
[] |
[
"en",
"es"
] |
TAGS
#task_categories-text-classification #size_categories-n<1K #language-English #language-Spanish #climate #region-us
|
# NDC Sector Classification
This dataset is built from the tagged NDC (Climate Watch) paragraphs made by GIZ Data Service Center and available on Hugging Face (GIZ/policy_qa_v0).
The NDC urls have been taken from IGES NDC Database.
Each NDC have been classified in a specific sector if it contains at least a paragraph classified as the specific sector. Each NDC can be associated to multiple sector.
The dataset contains 250 document classified in 18 sectors. The followin plot shows the number of documents tagged as each sector.
!image/png
## NDC Data
The csv containing the tagged NDC is structured as follows:
- 'Country': The country to which the NDC refers.
- 'Document': The type of document (INDC, First NDC, Second NDC).
- 'Language': The original language of the NDC.
- 'Sector': A json whose keys represent the sectors mentioned in the NDC and whose values represent the number of paragraphs that mention the specific secotor.
- 'URL': The pdf url.
## Sector Data
The json containing the sectors' description follows the scheme below:
Note: The descriptions have been taken from Wikipedia (en). The Spanish version is a translation of the english one.
|
[
"# NDC Sector Classification\nThis dataset is built from the tagged NDC (Climate Watch) paragraphs made by GIZ Data Service Center and available on Hugging Face (GIZ/policy_qa_v0).\nThe NDC urls have been taken from IGES NDC Database.\n\nEach NDC have been classified in a specific sector if it contains at least a paragraph classified as the specific sector. Each NDC can be associated to multiple sector.\n\nThe dataset contains 250 document classified in 18 sectors. The followin plot shows the number of documents tagged as each sector.\n!image/png",
"## NDC Data\nThe csv containing the tagged NDC is structured as follows:\n- 'Country': The country to which the NDC refers.\n- 'Document': The type of document (INDC, First NDC, Second NDC).\n- 'Language': The original language of the NDC.\n- 'Sector': A json whose keys represent the sectors mentioned in the NDC and whose values represent the number of paragraphs that mention the specific secotor.\n- 'URL': The pdf url.",
"## Sector Data\nThe json containing the sectors' description follows the scheme below:\n\nNote: The descriptions have been taken from Wikipedia (en). The Spanish version is a translation of the english one."
] |
[
"TAGS\n#task_categories-text-classification #size_categories-n<1K #language-English #language-Spanish #climate #region-us \n",
"# NDC Sector Classification\nThis dataset is built from the tagged NDC (Climate Watch) paragraphs made by GIZ Data Service Center and available on Hugging Face (GIZ/policy_qa_v0).\nThe NDC urls have been taken from IGES NDC Database.\n\nEach NDC have been classified in a specific sector if it contains at least a paragraph classified as the specific sector. Each NDC can be associated to multiple sector.\n\nThe dataset contains 250 document classified in 18 sectors. The followin plot shows the number of documents tagged as each sector.\n!image/png",
"## NDC Data\nThe csv containing the tagged NDC is structured as follows:\n- 'Country': The country to which the NDC refers.\n- 'Document': The type of document (INDC, First NDC, Second NDC).\n- 'Language': The original language of the NDC.\n- 'Sector': A json whose keys represent the sectors mentioned in the NDC and whose values represent the number of paragraphs that mention the specific secotor.\n- 'URL': The pdf url.",
"## Sector Data\nThe json containing the sectors' description follows the scheme below:\n\nNote: The descriptions have been taken from Wikipedia (en). The Spanish version is a translation of the english one."
] |
[
40,
132,
120,
43
] |
[
"passage: TAGS\n#task_categories-text-classification #size_categories-n<1K #language-English #language-Spanish #climate #region-us \n# NDC Sector Classification\nThis dataset is built from the tagged NDC (Climate Watch) paragraphs made by GIZ Data Service Center and available on Hugging Face (GIZ/policy_qa_v0).\nThe NDC urls have been taken from IGES NDC Database.\n\nEach NDC have been classified in a specific sector if it contains at least a paragraph classified as the specific sector. Each NDC can be associated to multiple sector.\n\nThe dataset contains 250 document classified in 18 sectors. The followin plot shows the number of documents tagged as each sector.\n!image/png## NDC Data\nThe csv containing the tagged NDC is structured as follows:\n- 'Country': The country to which the NDC refers.\n- 'Document': The type of document (INDC, First NDC, Second NDC).\n- 'Language': The original language of the NDC.\n- 'Sector': A json whose keys represent the sectors mentioned in the NDC and whose values represent the number of paragraphs that mention the specific secotor.\n- 'URL': The pdf url.## Sector Data\nThe json containing the sectors' description follows the scheme below:\n\nNote: The descriptions have been taken from Wikipedia (en). The Spanish version is a translation of the english one."
] |
2a61ff1d149dc98c0ddd1f81744d433f251dc128
|
# Dataset Card for "processed_demo"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
PeterLawrence/processed_demo
|
[
"region:us"
] |
2023-10-24T12:46:34+00:00
|
{"dataset_info": {"features": [{"name": "messages", "list": [{"name": "content", "dtype": "string"}, {"name": "role", "dtype": "string"}]}], "splits": [{"name": "train", "num_bytes": 26986, "num_examples": 34}], "download_size": 5785, "dataset_size": 26986}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-11-21T16:48:09+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "processed_demo"
More Information needed
|
[
"# Dataset Card for \"processed_demo\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"processed_demo\"\n\nMore Information needed"
] |
[
6,
14
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"processed_demo\"\n\nMore Information needed"
] |
3ccf620ec528f5d6d5a177e6fd75e44fc00be64e
|
# A Tale of Pronouns: Attributions on WinoMT
This dataset contains the pre-computed feature attribution scores
relative to the paper [A Tale of Pronouns: Interpretability Informs Gender Bias Mitigation for Fairer Instruction-Tuned Machine Translation](https://arxiv.org/abs/2310.12127).
## Dataset Details
We release the integrated gradient token-level attributions computed for each WinoMT translated example into Spanish and German
with Flan-T5-XXL and mtT0-XXL.
We computed the scores using [inseq](https://inseq.readthedocs.io/en/latest/).
The files here contain the FeatureAttributionOutput saved with the library's `save()` utility.
Therefore, you can load them using the `load()`. Please refer to inseq's doc to know more details.
Note that each file in the repository is compressed (`gz`) and contains a variable number of samples
(you can check how many samples by 1) loading the file into a variable (e.g., `out`) and print `len(out.sequence_attributions)`.
Each set of attribution must be loaded with a specific `inseq` version. See details below:
- Flan-T5-XXL (En-Es): v0.5.0
- Flan-T5-XXL (En-De): WIP
- mT0-XXL: v0.4.0
- mT0-XXL: v0.4.0
If you are having issues in loading the weights, feel free to reach out.
### Dataset Description
<!-- Provide a longer summary of what this dataset is. -->
- **Curated by:** [Giuseppe Attanasio](https://gattanasio.cc)
- **Language(s) (NLP):** Spanish, German
- **License:** Apache 2
### Dataset Sources [optional]
<!-- Provide the basic links for the dataset. -->
- **Repository:** [GitHub](https://github.com/MilaNLProc/interpretability-mt-gender-bias)
- **Paper:** [arxiv](https://arxiv.org/abs/2310.12127) (to appear in EMNLP 2023)
### Recommendations
TBD
**BibTeX:**
```bibtex
@article{attanasio2023tale,
title={A Tale of Pronouns: Interpretability Informs Gender Bias Mitigation for Fairer Instruction-Tuned Machine Translation},
author={Attanasio, Giuseppe and Plaza-del-Arco, Flor Miriam and Nozza, Debora and Lauscher, Anne},
journal={arXiv preprint arXiv:2310.12127},
year={2023}
}
```
## Dataset Card Contact
[Giuseppe Attanasio](mailto:[email protected])
|
MilaNLProc/a-tale-of-pronouns
|
[
"license:apache-2.0",
"arxiv:2310.12127",
"region:us"
] |
2023-10-24T13:11:01+00:00
|
{"license": "apache-2.0", "pretty_name": "A Tale of Pronouns"}
|
2023-11-18T10:22:12+00:00
|
[
"2310.12127"
] |
[] |
TAGS
#license-apache-2.0 #arxiv-2310.12127 #region-us
|
# A Tale of Pronouns: Attributions on WinoMT
This dataset contains the pre-computed feature attribution scores
relative to the paper A Tale of Pronouns: Interpretability Informs Gender Bias Mitigation for Fairer Instruction-Tuned Machine Translation.
## Dataset Details
We release the integrated gradient token-level attributions computed for each WinoMT translated example into Spanish and German
with Flan-T5-XXL and mtT0-XXL.
We computed the scores using inseq.
The files here contain the FeatureAttributionOutput saved with the library's 'save()' utility.
Therefore, you can load them using the 'load()'. Please refer to inseq's doc to know more details.
Note that each file in the repository is compressed ('gz') and contains a variable number of samples
(you can check how many samples by 1) loading the file into a variable (e.g., 'out') and print 'len(out.sequence_attributions)'.
Each set of attribution must be loaded with a specific 'inseq' version. See details below:
- Flan-T5-XXL (En-Es): v0.5.0
- Flan-T5-XXL (En-De): WIP
- mT0-XXL: v0.4.0
- mT0-XXL: v0.4.0
If you are having issues in loading the weights, feel free to reach out.
### Dataset Description
- Curated by: Giuseppe Attanasio
- Language(s) (NLP): Spanish, German
- License: Apache 2
### Dataset Sources [optional]
- Repository: GitHub
- Paper: arxiv (to appear in EMNLP 2023)
### Recommendations
TBD
BibTeX:
## Dataset Card Contact
Giuseppe Attanasio
|
[
"# A Tale of Pronouns: Attributions on WinoMT\n\nThis dataset contains the pre-computed feature attribution scores \nrelative to the paper A Tale of Pronouns: Interpretability Informs Gender Bias Mitigation for Fairer Instruction-Tuned Machine Translation.",
"## Dataset Details\n\nWe release the integrated gradient token-level attributions computed for each WinoMT translated example into Spanish and German\nwith Flan-T5-XXL and mtT0-XXL. \n\nWe computed the scores using inseq.\nThe files here contain the FeatureAttributionOutput saved with the library's 'save()' utility.\nTherefore, you can load them using the 'load()'. Please refer to inseq's doc to know more details.\n\nNote that each file in the repository is compressed ('gz') and contains a variable number of samples \n(you can check how many samples by 1) loading the file into a variable (e.g., 'out') and print 'len(out.sequence_attributions)'.\n\nEach set of attribution must be loaded with a specific 'inseq' version. See details below:\n- Flan-T5-XXL (En-Es): v0.5.0\n- Flan-T5-XXL (En-De): WIP\n- mT0-XXL: v0.4.0\n- mT0-XXL: v0.4.0\n\nIf you are having issues in loading the weights, feel free to reach out.",
"### Dataset Description\n\n\n\n- Curated by: Giuseppe Attanasio\n- Language(s) (NLP): Spanish, German\n- License: Apache 2",
"### Dataset Sources [optional]\n\n\n\n- Repository: GitHub\n- Paper: arxiv (to appear in EMNLP 2023)",
"### Recommendations\n\nTBD\n\nBibTeX:",
"## Dataset Card Contact\n\nGiuseppe Attanasio"
] |
[
"TAGS\n#license-apache-2.0 #arxiv-2310.12127 #region-us \n",
"# A Tale of Pronouns: Attributions on WinoMT\n\nThis dataset contains the pre-computed feature attribution scores \nrelative to the paper A Tale of Pronouns: Interpretability Informs Gender Bias Mitigation for Fairer Instruction-Tuned Machine Translation.",
"## Dataset Details\n\nWe release the integrated gradient token-level attributions computed for each WinoMT translated example into Spanish and German\nwith Flan-T5-XXL and mtT0-XXL. \n\nWe computed the scores using inseq.\nThe files here contain the FeatureAttributionOutput saved with the library's 'save()' utility.\nTherefore, you can load them using the 'load()'. Please refer to inseq's doc to know more details.\n\nNote that each file in the repository is compressed ('gz') and contains a variable number of samples \n(you can check how many samples by 1) loading the file into a variable (e.g., 'out') and print 'len(out.sequence_attributions)'.\n\nEach set of attribution must be loaded with a specific 'inseq' version. See details below:\n- Flan-T5-XXL (En-Es): v0.5.0\n- Flan-T5-XXL (En-De): WIP\n- mT0-XXL: v0.4.0\n- mT0-XXL: v0.4.0\n\nIf you are having issues in loading the weights, feel free to reach out.",
"### Dataset Description\n\n\n\n- Curated by: Giuseppe Attanasio\n- Language(s) (NLP): Spanish, German\n- License: Apache 2",
"### Dataset Sources [optional]\n\n\n\n- Repository: GitHub\n- Paper: arxiv (to appear in EMNLP 2023)",
"### Recommendations\n\nTBD\n\nBibTeX:",
"## Dataset Card Contact\n\nGiuseppe Attanasio"
] |
[
23,
63,
285,
32,
31,
13,
9
] |
[
"passage: TAGS\n#license-apache-2.0 #arxiv-2310.12127 #region-us \n# A Tale of Pronouns: Attributions on WinoMT\n\nThis dataset contains the pre-computed feature attribution scores \nrelative to the paper A Tale of Pronouns: Interpretability Informs Gender Bias Mitigation for Fairer Instruction-Tuned Machine Translation.## Dataset Details\n\nWe release the integrated gradient token-level attributions computed for each WinoMT translated example into Spanish and German\nwith Flan-T5-XXL and mtT0-XXL. \n\nWe computed the scores using inseq.\nThe files here contain the FeatureAttributionOutput saved with the library's 'save()' utility.\nTherefore, you can load them using the 'load()'. Please refer to inseq's doc to know more details.\n\nNote that each file in the repository is compressed ('gz') and contains a variable number of samples \n(you can check how many samples by 1) loading the file into a variable (e.g., 'out') and print 'len(out.sequence_attributions)'.\n\nEach set of attribution must be loaded with a specific 'inseq' version. See details below:\n- Flan-T5-XXL (En-Es): v0.5.0\n- Flan-T5-XXL (En-De): WIP\n- mT0-XXL: v0.4.0\n- mT0-XXL: v0.4.0\n\nIf you are having issues in loading the weights, feel free to reach out.### Dataset Description\n\n\n\n- Curated by: Giuseppe Attanasio\n- Language(s) (NLP): Spanish, German\n- License: Apache 2### Dataset Sources [optional]\n\n\n\n- Repository: GitHub\n- Paper: arxiv (to appear in EMNLP 2023)### Recommendations\n\nTBD\n\nBibTeX:## Dataset Card Contact\n\nGiuseppe Attanasio"
] |
2970493cc65246f2ee51214c5dd649cd43f028ea
|
# Dataset Card for "31425212"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
result-kand2-sdxl-wuerst-karlo/31425212
|
[
"region:us"
] |
2023-10-24T13:14:18+00:00
|
{"dataset_info": {"features": [{"name": "result", "dtype": "string"}, {"name": "id", "dtype": "int64"}], "splits": [{"name": "train", "num_bytes": 203, "num_examples": 10}], "download_size": 1410, "dataset_size": 203}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T13:14:19+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "31425212"
More Information needed
|
[
"# Dataset Card for \"31425212\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"31425212\"\n\nMore Information needed"
] |
[
6,
14
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"31425212\"\n\nMore Information needed"
] |
2d54cfdab7724ac745832c9c9293d7a5b6cab5f9
|
# Dataset Card for "yun-car-dataset"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
zhangyi617/yun-car-dataset
|
[
"region:us"
] |
2023-10-24T13:19:37+00:00
|
{"dataset_info": {"features": [{"name": "image", "dtype": "image"}, {"name": "text", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 11217081.0, "num_examples": 33}], "download_size": 11218227, "dataset_size": 11217081.0}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T13:19:39+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "yun-car-dataset"
More Information needed
|
[
"# Dataset Card for \"yun-car-dataset\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"yun-car-dataset\"\n\nMore Information needed"
] |
[
6,
16
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"yun-car-dataset\"\n\nMore Information needed"
] |
5c9db0f0e8dc3d605c264a4fd0b2bf74de904358
|
## Overview
This data was used to train model:
https://huggingface.co/PDBEurope/BiomedNLP-PubMedBERT-ProteinStructure-NER-v1.2
There are 19 different entity types in this dataset:
"chemical", "complex_assembly", "evidence", "experimental_method", "gene", "mutant",
"oligomeric_state", "protein", "protein_state", "protein_type", "ptm", "residue_name",
"residue_name_number","residue_number", "residue_range", "site", "species", "structure_element",
"taxonomy_domain"
The data prepared as IOB formated input has been used during training, development
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (https://www.teamtat.org/) and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
| document ID | number of annotations in BioC XML | number of annotations in IOB/JSON/CSV | number of sentences |
| --- | --- | --- | --- |
| PMC4850273 | 1121 | 1121 | 204 |
| PMC4784909 | 865 | 865 | 204 |
| PMC4850288 | 716 | 708 | 146 |
| PMC4887326 | 933 | 933 | 152 |
| PMC4833862 | 1044 | 1044 | 192 |
| PMC4832331 | 739 | 718 | 134 |
| PMC4852598 | 1229 | 1218 | 250 |
| PMC4786784 | 1549 | 1549 | 232 |
| PMC4848090 | 987 | 985 | 191 |
| PMC4792962 | 1268 | 1268 | 256 |
| total | 10451 | 10409 | 1961 |
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: https://bioc.sourceforge.net/
## Raw BioC XML files
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw_BioC_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_raw.xml.
## Annotations in IOB format
The IOB formated files can be found in the directory: "annotation_IOB"
The four files are as follows:
* all.tsv --> all sentences and annotations used to create model
"PDBEurope/BiomedNLP-PubMedBERT-ProteinStructure-NER-v1.2"; 1961 sentences
* train.tsv --> training subset of the data; 1372 sentences
* dev.tsv --> development subset of the data; 294 sentences
* test.tsv --> testing subset of the data; 295 sentences
The total number of annotations is: 10409
## Annotations in BioC JSON
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
* "begin" --> character start position of annotation
* "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
```json
[{"sourceid":"4784909",
"sourcedb":"",
"project":"",
"target":"",
"text":"",
"denotations":[{"span":{"begin":24,
"end":34},
"obj":"chemical,CHEBI:,[email protected],2023-03-21T15:19:42Z",
"id":"4500"},
{"span":{"begin":50,
"end":59},
"obj":"taxonomy_domain,DUMMY:,[email protected],2023-03-21T15:15:03Z",
"id":"1281"}]
}
]
```
## Annotations in BioC XML
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
* "offset" --> start character position as defined by offset value
* "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
```xml
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE collection SYSTEM "BioC.dtd">
<collection>
<source>PMC</source>
<date>20140719</date>
<key>pmc.key</key>
<document>
<id>4784909</id>
<passage>
<offset>0</offset>
<text>The Structural Basis of Coenzyme A Recycling in a Bacterial Organelle</text>
<annotation id="4500">
<infon key="type">chemical</infon>
<infon key="identifier">CHEBI:</infon>
<infon key="annotator">[email protected]</infon>
<infon key="updated_at">2023-03-21T15:19:42Z</infon>
<location offset="24" length="10"/>
<text>Coenzyme A</text>
</annotation>
</passage>
</document>
</collection>
```
## Annotations in CSV
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno_start" --> character start position of the annotation
* "anno_end" --> character end position of the annotation
* "anno_text" --> text covered by the annotation
* "entity_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
## Annotations in JSON
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation_JSON" under the name "annotations.json".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
* "sid" --> unique sentence ID
* "sent" --> sentence text as string
* "section" --> publication section the sentence is in
* "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
```json
{"PMC4850273": {"annotations":
[{"sid": 0,
"sent": "Molecular Dissection of Xyloglucan Recognition in a Prominent Human Gut Symbiont",
"section": "TITLE",
"ner": [
[24,34,"Xyloglucan","chemical"],
[62,67,"Human","species"],]
},]
}}
```
|
PDBEurope/protein_structure_NER_model_v1.2
|
[
"language:en",
"license:mit",
"biology",
"protein structure",
"token classification",
"region:us"
] |
2023-10-24T13:43:14+00:00
|
{"language": ["en"], "license": "mit", "tags": ["biology", "protein structure", "token classification"], "configs": [{"config_name": "protein_structure_NER_model_v1.2", "data_files": [{"split": "train", "path": "annotation_IOB/train.tsv"}, {"split": "dev", "path": "annotation_IOB/dev.tsv"}, {"split": "test", "path": "annotation_IOB/test.tsv"}]}]}
|
2023-11-01T09:51:43+00:00
|
[] |
[
"en"
] |
TAGS
#language-English #license-mit #biology #protein structure #token classification #region-us
|
Overview
--------
This data was used to train model:
URL
There are 19 different entity types in this dataset:
"chemical", "complex\_assembly", "evidence", "experimental\_method", "gene", "mutant",
"oligomeric\_state", "protein", "protein\_state", "protein\_type", "ptm", "residue\_name",
"residue\_name\_number","residue\_number", "residue\_range", "site", "species", "structure\_element",
"taxonomy\_domain"
The data prepared as IOB formated input has been used during training, development
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (URL and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: URL
Raw BioC XML files
------------------
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw\_BioC\_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_raw.xml.
Annotations in IOB format
-------------------------
The IOB formated files can be found in the directory: "annotation\_IOB"
The four files are as follows:
* URL --> all sentences and annotations used to create model
"PDBEurope/BiomedNLP-PubMedBERT-ProteinStructure-NER-v1.2"; 1961 sentences
* URL --> training subset of the data; 1372 sentences
* URL --> development subset of the data; 294 sentences
* URL --> testing subset of the data; 295 sentences
The total number of annotations is: 10409
Annotations in BioC JSON
------------------------
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
+ "begin" --> character start position of annotation
+ "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
Annotations in BioC XML
-----------------------
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID\_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated\_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
+ "offset" --> start character position as defined by offset value
+ "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
Annotations in CSV
------------------
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation\_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno\_start" --> character start position of the annotation
* "anno\_end" --> character end position of the annotation
* "anno\_text" --> text covered by the annotation
* "entity\_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
Annotations in JSON
-------------------
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation\_JSON" under the name "URL".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
+ "sid" --> unique sentence ID
+ "sent" --> sentence text as string
+ "section" --> publication section the sentence is in
+ "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
|
[] |
[
"TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
[
27
] |
[
"passage: TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
f94586d73f5a8fc81ed649ac8986d35f850020e6
|
## Overview
This data was used to train model:
https://huggingface.co/PDBEurope/BiomedNLP-PubMedBERT-ProteinStructure-NER-v1.4
There are 19 different entity types in this dataset:
"chemical", "complex_assembly", "evidence", "experimental_method", "gene", "mutant",
"oligomeric_state", "protein", "protein_state", "protein_type", "ptm", "residue_name",
"residue_name_number","residue_number", "residue_range", "site", "species", "structure_element",
"taxonomy_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (https://www.teamtat.org/) and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
| document ID | number of annotations in BioC XML | number of annotations in IOB/JSON/CSV | number of sentences |
| --- | --- | --- | --- |
| PMC4850273 | 1121 | 1121 | 204 |
| PMC4784909 | 865 | 865 | 204 |
| PMC4850288 | 716 | 708 | 146 |
| PMC4887326 | 933 | 933 | 152 |
| PMC4833862 | 1044 | 1044 | 192 |
| PMC4832331 | 739 | 718 | 134 |
| PMC4852598 | 1229 | 1218 | 250 |
| PMC4786784 | 1549 | 1549 | 232 |
| PMC4848090 | 987 | 985 | 191 |
| PMC4792962 | 1268 | 1268 | 256 |
| PMC4841544 | 1434 | 1433 | 273 |
| PMC4772114 | 825 | 825 | 166 |
| PMC4872110 | 1276 | 1276 | 253 |
| PMC4848761 | 887 | 883 | 252 |
| PMC4919469 | 1628 | 1616 | 336 |
| PMC4880283 | 771 | 771 | 166 |
| PMC4937829 | 625 | 625 | 181 |
| PMC4968113 | 1238 | 1238 | 292 |
| PMC4854314 | 481 | 471 | 139 |
| PMC4871749 | 383 | 383 | 76 |
| total | 19999 | 19930 | 4095 |
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: https://bioc.sourceforge.net/
## Raw BioC XML files
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw_BioC_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_raw.xml.
## Annotations in IOB format
The IOB formated files can be found in the directory: "annotation_IOB"
The four files are as follows:
* all.tsv --> all sentences and annotations used to create model
"PDBEurope/BiomedNLP-PubMedBERT-ProteinStructure-NER-v1.4"; 4095 sentences
* train.tsv --> training subset of the data; 2866 sentences
* dev.tsv --> development subset of the data; 614 sentences
* test.tsv --> testing subset of the data; 615 sentences
The total number of annotations is: 19930
## Annotations in BioC JSON
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
* "begin" --> character start position of annotation
* "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
```json
[{"sourceid":"4784909",
"sourcedb":"",
"project":"",
"target":"",
"text":"",
"denotations":[{"span":{"begin":24,
"end":34},
"obj":"chemical,CHEBI:,[email protected],2023-03-21T15:19:42Z",
"id":"4500"},
{"span":{"begin":50,
"end":59},
"obj":"taxonomy_domain,DUMMY:,[email protected],2023-03-21T15:15:03Z",
"id":"1281"}]
}
]
```
## Annotations in BioC XML
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
* "offset" --> start character position as defined by offset value
* "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
```xml
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE collection SYSTEM "BioC.dtd">
<collection>
<source>PMC</source>
<date>20140719</date>
<key>pmc.key</key>
<document>
<id>4784909</id>
<passage>
<offset>0</offset>
<text>The Structural Basis of Coenzyme A Recycling in a Bacterial Organelle</text>
<annotation id="4500">
<infon key="type">chemical</infon>
<infon key="identifier">CHEBI:</infon>
<infon key="annotator">[email protected]</infon>
<infon key="updated_at">2023-03-21T15:19:42Z</infon>
<location offset="24" length="10"/>
<text>Coenzyme A</text>
</annotation>
</passage>
</document>
</collection>
```
## Annotations in CSV
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno_start" --> character start position of the annotation
* "anno_end" --> character end position of the annotation
* "anno_text" --> text covered by the annotation
* "entity_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
## Annotations in JSON
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation_JSON" under the name "annotations.json".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
* "sid" --> unique sentence ID
* "sent" --> sentence text as string
* "section" --> publication section the sentence is in
* "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
```json
{"PMC4850273": {"annotations":
[{"sid": 0,
"sent": "Molecular Dissection of Xyloglucan Recognition in a Prominent Human Gut Symbiont",
"section": "TITLE",
"ner": [
[24,34,"Xyloglucan","chemical"],
[62,67,"Human","species"],]
},]
}}
```
|
PDBEurope/protein_structure_NER_model_v1.4
|
[
"language:en",
"license:mit",
"biology",
"protein structure",
"token classification",
"region:us"
] |
2023-10-24T13:54:44+00:00
|
{"language": ["en"], "license": "mit", "tags": ["biology", "protein structure", "token classification"], "configs": [{"config_name": "protein_structure_NER_model_v1.4", "data_files": [{"split": "train", "path": "annotation_IOB/train.tsv"}, {"split": "dev", "path": "annotation_IOB/dev.tsv"}, {"split": "test", "path": "annotation_IOB/test.tsv"}]}]}
|
2023-11-01T10:00:43+00:00
|
[] |
[
"en"
] |
TAGS
#language-English #license-mit #biology #protein structure #token classification #region-us
|
Overview
--------
This data was used to train model:
URL
There are 19 different entity types in this dataset:
"chemical", "complex\_assembly", "evidence", "experimental\_method", "gene", "mutant",
"oligomeric\_state", "protein", "protein\_state", "protein\_type", "ptm", "residue\_name",
"residue\_name\_number","residue\_number", "residue\_range", "site", "species", "structure\_element",
"taxonomy\_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (URL and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: URL
Raw BioC XML files
------------------
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw\_BioC\_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_raw.xml.
Annotations in IOB format
-------------------------
The IOB formated files can be found in the directory: "annotation\_IOB"
The four files are as follows:
* URL --> all sentences and annotations used to create model
"PDBEurope/BiomedNLP-PubMedBERT-ProteinStructure-NER-v1.4"; 4095 sentences
* URL --> training subset of the data; 2866 sentences
* URL --> development subset of the data; 614 sentences
* URL --> testing subset of the data; 615 sentences
The total number of annotations is: 19930
Annotations in BioC JSON
------------------------
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
+ "begin" --> character start position of annotation
+ "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
Annotations in BioC XML
-----------------------
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID\_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated\_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
+ "offset" --> start character position as defined by offset value
+ "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
Annotations in CSV
------------------
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation\_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno\_start" --> character start position of the annotation
* "anno\_end" --> character end position of the annotation
* "anno\_text" --> text covered by the annotation
* "entity\_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
Annotations in JSON
-------------------
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation\_JSON" under the name "URL".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
+ "sid" --> unique sentence ID
+ "sent" --> sentence text as string
+ "section" --> publication section the sentence is in
+ "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
|
[] |
[
"TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
[
27
] |
[
"passage: TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
3ca0c4a9aa9ba1d665bba410d8300f1ae3e82126
|
## Overview
This data was used to train model:
https://huggingface.co/PDBEurope/BiomedNLP-PubMedBERT-ProteinStructure-NER-v2.1
There are 20 different entity types in this dataset:
"bond_interaction", "chemical", "complex_assembly", "evidence", "experimental_method", "gene",
"mutant", "oligomeric_state", "protein", "protein_state", "protein_type", "ptm", "residue_name",
"residue_name_number","residue_number", "residue_range", "site", "species", "structure_element",
"taxonomy_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (https://www.teamtat.org/) and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
| document ID | number of annotations in BioC XML | number of annotations in IOB/JSON/CSV | number of sentences |
| --- | --- | --- | --- |
| PMC4850273 | 1129 | 1129 | 205 |
| PMC4784909 | 868 | 868 | 204 |
| PMC4850288 | 718 | 710 | 146 |
| PMC4887326 | 942 | 942 | 152 |
| PMC4833862 | 1044 | 1044 | 192 |
| PMC4832331 | 739 | 718 | 134 |
| PMC4852598 | 1239 | 1228 | 250 |
| PMC4786784 | 1573 | 1573 | 232 |
| PMC4848090 | 1002 | 1000 | 192 |
| PMC4792962 | 1297 | 1297 | 256 |
| PMC4841544 | 1460 | 1459 | 274 |
| PMC4772114 | 824 | 824 | 165 |
| PMC4872110 | 1283 | 1283 | 250 |
| PMC4848761 | 888 | 884 | 252 |
| PMC4919469 | 1636 | 1624 | 336 |
| PMC4880283 | 783 | 783 | 166 |
| PMC4968113 | 1245 | 1245 | 292 |
| PMC4937829 | 633 | 633 | 181 |
| PMC4854314 | 498 | 488 | 139 |
| PMC4871749 | 411 | 411 | 79 |
| PMC4869123 | 922 | 922 | 195 |
| PMC4888278 | 580 | 580 | 102 |
| PMC4795551 | 1475 | 1475 | 297 |
| PMC4831588 | 1087 | 1070 | 224 |
| PMC4918766 | 1027 | 1027 | 210 |
| PMC4802042 | 1445 | 1445 | 268 |
| PMC4896748 | 2652 | 2638 | 480 |
| PMC4781976 | 115 | 113 | 24 |
| PMC4802085 | 983 | 983 | 193 |
| PMC4887163 | 856 | 856 | 196|
| total | 31354 | 31252 | 6286 |
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: https://bioc.sourceforge.net/
## Raw BioC XML files
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw_BioC_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_raw.xml.
## Annotations in IOB format
The IOB formated files can be found in the directory: "annotation_IOB"
The four files are as follows:
* all.tsv --> all sentences and annotations used to create model
"PDBEurope/BiomedNLP-PubMedBERT-ProteinStructure-NER-v2.1"; 6286 sentences
* train.tsv --> training subset of the data; 4400 sentences
* dev.tsv --> development subset of the data; 943 sentences
* test.tsv --> testing subset of the data; 943 sentences
The total number of annotations is: 31252
## Annotations in BioC JSON
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
* "begin" --> character start position of annotation
* "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
```json
[{"sourceid":"4784909",
"sourcedb":"",
"project":"",
"target":"",
"text":"",
"denotations":[{"span":{"begin":24,
"end":34},
"obj":"chemical,CHEBI:,[email protected],2023-03-21T15:19:42Z",
"id":"4500"},
{"span":{"begin":50,
"end":59},
"obj":"taxonomy_domain,DUMMY:,[email protected],2023-03-21T15:15:03Z",
"id":"1281"}]
}
]
```
## Annotations in BioC XML
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
* "offset" --> start character position as defined by offset value
* "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
```xml
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE collection SYSTEM "BioC.dtd">
<collection>
<source>PMC</source>
<date>20140719</date>
<key>pmc.key</key>
<document>
<id>4784909</id>
<passage>
<offset>0</offset>
<text>The Structural Basis of Coenzyme A Recycling in a Bacterial Organelle</text>
<annotation id="4500">
<infon key="type">chemical</infon>
<infon key="identifier">CHEBI:</infon>
<infon key="annotator">[email protected]</infon>
<infon key="updated_at">2023-03-21T15:19:42Z</infon>
<location offset="24" length="10"/>
<text>Coenzyme A</text>
</annotation>
</passage>
</document>
</collection>
```
## Annotations in CSV
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno_start" --> character start position of the annotation
* "anno_end" --> character end position of the annotation
* "anno_text" --> text covered by the annotation
* "entity_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
## Annotations in JSON
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation_JSON" under the name "annotations.json".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
* "sid" --> unique sentence ID
* "sent" --> sentence text as string
* "section" --> publication section the sentence is in
* "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
```json
{"PMC4850273": {"annotations":
[{"sid": 0,
"sent": "Molecular Dissection of Xyloglucan Recognition in a Prominent Human Gut Symbiont",
"section": "TITLE",
"ner": [
[24,34,"Xyloglucan","chemical"],
[62,67,"Human","species"],]
},]
}}
```
|
PDBEurope/protein_structure_NER_model_v2.1
|
[
"language:en",
"license:mit",
"biology",
"protein structure",
"token classification",
"region:us"
] |
2023-10-24T13:57:29+00:00
|
{"language": ["en"], "license": "mit", "tags": ["biology", "protein structure", "token classification"], "configs": [{"config_name": "protein_structure_NER_model_v2.1", "data_files": [{"split": "train", "path": "annotation_IOB/train.tsv"}, {"split": "dev", "path": "annotation_IOB/dev.tsv"}, {"split": "test", "path": "annotation_IOB/test.tsv"}]}]}
|
2023-11-01T10:19:37+00:00
|
[] |
[
"en"
] |
TAGS
#language-English #license-mit #biology #protein structure #token classification #region-us
|
Overview
--------
This data was used to train model:
URL
There are 20 different entity types in this dataset:
"bond\_interaction", "chemical", "complex\_assembly", "evidence", "experimental\_method", "gene",
"mutant", "oligomeric\_state", "protein", "protein\_state", "protein\_type", "ptm", "residue\_name",
"residue\_name\_number","residue\_number", "residue\_range", "site", "species", "structure\_element",
"taxonomy\_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (URL and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: URL
Raw BioC XML files
------------------
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw\_BioC\_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_raw.xml.
Annotations in IOB format
-------------------------
The IOB formated files can be found in the directory: "annotation\_IOB"
The four files are as follows:
* URL --> all sentences and annotations used to create model
"PDBEurope/BiomedNLP-PubMedBERT-ProteinStructure-NER-v2.1"; 6286 sentences
* URL --> training subset of the data; 4400 sentences
* URL --> development subset of the data; 943 sentences
* URL --> testing subset of the data; 943 sentences
The total number of annotations is: 31252
Annotations in BioC JSON
------------------------
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
+ "begin" --> character start position of annotation
+ "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
Annotations in BioC XML
-----------------------
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID\_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated\_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
+ "offset" --> start character position as defined by offset value
+ "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
Annotations in CSV
------------------
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation\_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno\_start" --> character start position of the annotation
* "anno\_end" --> character end position of the annotation
* "anno\_text" --> text covered by the annotation
* "entity\_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
Annotations in JSON
-------------------
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation\_JSON" under the name "URL".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
+ "sid" --> unique sentence ID
+ "sent" --> sentence text as string
+ "section" --> publication section the sentence is in
+ "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
|
[] |
[
"TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
[
27
] |
[
"passage: TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
6d940b23013695b98b28de82d67a104577739d84
|
## Overview
This data was used to train model:
https://huggingface.co/PDBEurope/BiomedNLP-PubMedBERT-ProteinStructure-NER-v3.1
There are 20 different entity types in this dataset:
"bond_interaction", "chemical", "complex_assembly", "evidence", "experimental_method", "gene",
"mutant", "oligomeric_state", "protein", "protein_state", "protein_type", "ptm", "residue_name",
"residue_name_number","residue_number", "residue_range", "site", "species", "structure_element",
"taxonomy_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (https://www.teamtat.org/) and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
| document ID | number of annotations in BioC XML | number of annotations in IOB/JSON/CSV | number of sentences |
| --- | --- | --- | --- |
| PMC4850273 | 1129 | 1129 | 205 |
| PMC4784909 | 868 | 868 | 204 |
| PMC4850288 | 717 | 709 | 146 |
| PMC4887326 | 942 | 942 | 152 |
| PMC4833862 | 1044 | 1044 | 192 |
| PMC4832331 | 739 | 718 | 134 |
| PMC4852598 | 1239 | 1228 | 250 |
| PMC4786784 | 1573 | 1573 | 232 |
| PMC4848090 | 1000 | 998 | 192 |
| PMC4792962 | 1297 | 1297 | 256 |
| PMC4841544 | 1460 | 1459 | 274 |
| PMC4772114 | 824 | 824 | 165 |
| PMC4872110 | 1283 | 1283 | 250 |
| PMC4848761 | 888 | 884 | 252 |
| PMC4919469 | 1636 | 1624 | 336 |
| PMC4880283 | 783 | 783 | 166 |
| PMC4968113 | 1245 | 1245 | 292 |
| PMC4937829 | 633 | 633 | 181 |
| PMC4854314 | 498 | 488 | 139 |
| PMC4871749 | 411 | 411 | 79 |
| PMC4869123 | 922 | 922 | 195 |
| PMC4888278 | 580 | 580 | 102 |
| PMC4795551 | 1475 | 1475 | 297 |
| PMC4831588 | 1087 | 1070 | 224 |
| PMC4918766 | 1027 | 1027 | 210 |
| PMC4802042 | 1441 | 1441 | 264 |
| PMC4896748 | 2652 | 2638 | 480 |
| PMC4781976 | 115 | 113 | 24 |
| PMC4802085 | 983 | 983 | 193 |
| PMC4887163 | 856 | 856 | 196|
| PMC4918759 | 803 | 803 | 175 |
| PMC4855620 | 563 | 563 | 122 |
| PMC4822050 | 1521 | 1521 | 249 |
| PMC4822561 | 367 | 366 | 84 |
| PMC4885502 | 577 | 577 | 97 |
| PMC4746701 | 1130 | 1130 | 245 |
| PMC4820378 | 733 | 733 | 170 |
| PMC4773095 | 1323 | 1323 | 252 |
| PMC4857006 | 1358 | 1358 | 249 |
| PMC4774019 | 532 | 530 | 117 |
| total | 40254 | 40149 | 8042 |
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: https://bioc.sourceforge.net/
## Raw BioC XML files
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw_BioC_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_raw.xml.
## Annotations in IOB format
The IOB formated files can be found in the directory: "annotation_IOB"
The four files are as follows:
* all.tsv --> all sentences and annotations used to create model
"PDBEurope/BiomedNLP-PubMedBERT-ProteinStructure-NER-v3.1"; 8042 sentences
* train.tsv --> training subset of the data; 5629 sentences
* dev.tsv --> development subset of the data; 1206 sentences
* test.tsv --> testing subset of the data; 1207 sentences
The total number of annotations is: 40149
## Annotations in BioC JSON
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
* "begin" --> character start position of annotation
* "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
```json
[{"sourceid":"4784909",
"sourcedb":"",
"project":"",
"target":"",
"text":"",
"denotations":[{"span":{"begin":24,
"end":34},
"obj":"chemical,CHEBI:,[email protected],2023-03-21T15:19:42Z",
"id":"4500"},
{"span":{"begin":50,
"end":59},
"obj":"taxonomy_domain,DUMMY:,[email protected],2023-03-21T15:15:03Z",
"id":"1281"}]
}
]
```
## Annotations in BioC XML
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
* "offset" --> start character position as defined by offset value
* "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
```xml
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE collection SYSTEM "BioC.dtd">
<collection>
<source>PMC</source>
<date>20140719</date>
<key>pmc.key</key>
<document>
<id>4784909</id>
<passage>
<offset>0</offset>
<text>The Structural Basis of Coenzyme A Recycling in a Bacterial Organelle</text>
<annotation id="4500">
<infon key="type">chemical</infon>
<infon key="identifier">CHEBI:</infon>
<infon key="annotator">[email protected]</infon>
<infon key="updated_at">2023-03-21T15:19:42Z</infon>
<location offset="24" length="10"/>
<text>Coenzyme A</text>
</annotation>
</passage>
</document>
</collection>
```
## Annotations in CSV
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno_start" --> character start position of the annotation
* "anno_end" --> character end position of the annotation
* "anno_text" --> text covered by the annotation
* "entity_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
## Annotations in JSON
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation_JSON" under the name "annotations.json".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
* "sid" --> unique sentence ID
* "sent" --> sentence text as string
* "section" --> publication section the sentence is in
* "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
```json
{"PMC4850273": {"annotations":
[{"sid": 0,
"sent": "Molecular Dissection of Xyloglucan Recognition in a Prominent Human Gut Symbiont",
"section": "TITLE",
"ner": [
[24,34,"Xyloglucan","chemical"],
[62,67,"Human","species"],]
},]
}}
```
|
PDBEurope/protein_structure_NER_model_v3.1
|
[
"language:en",
"license:mit",
"biology",
"protein structure",
"token classification",
"region:us"
] |
2023-10-24T14:00:15+00:00
|
{"language": ["en"], "license": "mit", "tags": ["biology", "protein structure", "token classification"], "configs": [{"config_name": "protein_structure_NER_model_v3.1", "data_files": [{"split": "train", "path": "annotation_IOB/train.tsv"}, {"split": "dev", "path": "annotation_IOB/dev.tsv"}, {"split": "test", "path": "annotation_IOB/test.tsv"}]}]}
|
2023-11-01T10:27:48+00:00
|
[] |
[
"en"
] |
TAGS
#language-English #license-mit #biology #protein structure #token classification #region-us
|
Overview
--------
This data was used to train model:
URL
There are 20 different entity types in this dataset:
"bond\_interaction", "chemical", "complex\_assembly", "evidence", "experimental\_method", "gene",
"mutant", "oligomeric\_state", "protein", "protein\_state", "protein\_type", "ptm", "residue\_name",
"residue\_name\_number","residue\_number", "residue\_range", "site", "species", "structure\_element",
"taxonomy\_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (URL and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: URL
Raw BioC XML files
------------------
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw\_BioC\_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_raw.xml.
Annotations in IOB format
-------------------------
The IOB formated files can be found in the directory: "annotation\_IOB"
The four files are as follows:
* URL --> all sentences and annotations used to create model
"PDBEurope/BiomedNLP-PubMedBERT-ProteinStructure-NER-v3.1"; 8042 sentences
* URL --> training subset of the data; 5629 sentences
* URL --> development subset of the data; 1206 sentences
* URL --> testing subset of the data; 1207 sentences
The total number of annotations is: 40149
Annotations in BioC JSON
------------------------
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
+ "begin" --> character start position of annotation
+ "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
Annotations in BioC XML
-----------------------
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID\_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated\_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
+ "offset" --> start character position as defined by offset value
+ "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
Annotations in CSV
------------------
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation\_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno\_start" --> character start position of the annotation
* "anno\_end" --> character end position of the annotation
* "anno\_text" --> text covered by the annotation
* "entity\_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
Annotations in JSON
-------------------
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation\_JSON" under the name "URL".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
+ "sid" --> unique sentence ID
+ "sent" --> sentence text as string
+ "section" --> publication section the sentence is in
+ "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
|
[] |
[
"TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
[
27
] |
[
"passage: TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
1d43168eae8d6c7904319cf951ed8de2752da047
|
## Overview
This data was used to train model:
https://huggingface.co/mevol/BiomedNLP-PubMedBERT-ProteinStructure-NER-v1.2
There are 19 different entity types in this dataset:
"chemical", "complex_assembly", "evidence", "experimental_method", "gene", "mutant",
"oligomeric_state", "protein", "protein_state", "protein_type", "ptm", "residue_name",
"residue_name_number","residue_number", "residue_range", "site", "species", "structure_element",
"taxonomy_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (https://www.teamtat.org/) and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
| document ID | number of annotations in BioC XML | number of annotations in IOB/JSON/CSV | number of sentences |
| --- | --- | --- | --- |
| PMC4850273 | 1121 | 1121 | 204 |
| PMC4784909 | 865 | 865 | 204 |
| PMC4850288 | 716 | 708 | 146 |
| PMC4887326 | 933 | 933 | 152 |
| PMC4833862 | 1044 | 1044 | 192 |
| PMC4832331 | 739 | 718 | 134 |
| PMC4852598 | 1229 | 1218 | 250 |
| PMC4786784 | 1549 | 1549 | 232 |
| PMC4848090 | 987 | 985 | 191 |
| PMC4792962 | 1268 | 1268 | 256 |
| total | 10451 | 10409 | 1961 |
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat (https://www.teamtat.org/). More about the BioC
format can be found here: https://bioc.sourceforge.net/
## Raw BioC XML files
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw_BioC_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_raw.xml.
## Annotations in IOB format
The IOB formated files can be found in the directory: "annotation_IOB"
The four files are as follows:
* all.tsv --> all sentences and annotations used to create model
"mevol/BiomedNLP-PubMedBERT-ProteinStructure-NER-v1.2"; 1961 sentences
* train.tsv --> training subset of the data; 1372 sentences
* dev.tsv --> development subset of the data; 294 sentences
* test.tsv --> testing subset of the data; 295 sentences
The total number of annotations is: 10409
## Annotations in BioC JSON
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
* "begin" --> character start position of annotation
* "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
```json
[{"sourceid":"4784909",
"sourcedb":"",
"project":"",
"target":"",
"text":"",
"denotations":[{"span":{"begin":24,
"end":34},
"obj":"chemical,CHEBI:,[email protected],2023-03-21T15:19:42Z",
"id":"4500"},
{"span":{"begin":50,
"end":59},
"obj":"taxonomy_domain,DUMMY:,[email protected],2023-03-21T15:15:03Z",
"id":"1281"}]
}
]
```
## Annotations in BioC XML
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
* "offset" --> start character position as defined by offset value
* "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
```xml
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE collection SYSTEM "BioC.dtd">
<collection>
<source>PMC</source>
<date>20140719</date>
<key>pmc.key</key>
<document>
<id>4784909</id>
<passage>
<offset>0</offset>
<text>The Structural Basis of Coenzyme A Recycling in a Bacterial Organelle</text>
<annotation id="4500">
<infon key="type">chemical</infon>
<infon key="identifier">CHEBI:</infon>
<infon key="annotator">[email protected]</infon>
<infon key="updated_at">2023-03-21T15:19:42Z</infon>
<location offset="24" length="10"/>
<text>Coenzyme A</text>
</annotation>
</passage>
</document>
</collection>
```
## Annotations in CSV
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno_start" --> character start position of the annotation
* "anno_end" --> character end position of the annotation
* "anno_text" --> text covered by the annotation
* "entity_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
## Annotations in JSON
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation_JSON" under the name "annotations.json".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
* "sid" --> unique sentence ID
* "sent" --> sentence text as string
* "section" --> publication section the sentence is in
* "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
```json
{"PMC4850273": {"annotations":
[{"sid": 0,
"sent": "Molecular Dissection of Xyloglucan Recognition in a Prominent Human Gut Symbiont",
"section": "TITLE",
"ner": [
[24,34,"Xyloglucan","chemical"],
[62,67,"Human","species"],]
},]
}}
```
|
mevol/protein_structure_NER_model_v1.2
|
[
"language:en",
"license:mit",
"biology",
"protein structure",
"token classification",
"region:us"
] |
2023-10-24T14:03:08+00:00
|
{"language": ["en"], "license": "mit", "tags": ["biology", "protein structure", "token classification"], "configs": [{"config_name": "protein_structure_NER_model_v1.2", "data_files": [{"split": "train", "path": "annotation_IOB/train.tsv"}, {"split": "dev", "path": "annotation_IOB/dev.tsv"}, {"split": "test", "path": "annotation_IOB/test.tsv"}]}]}
|
2023-11-01T09:50:22+00:00
|
[] |
[
"en"
] |
TAGS
#language-English #license-mit #biology #protein structure #token classification #region-us
|
Overview
--------
This data was used to train model:
URL
There are 19 different entity types in this dataset:
"chemical", "complex\_assembly", "evidence", "experimental\_method", "gene", "mutant",
"oligomeric\_state", "protein", "protein\_state", "protein\_type", "ptm", "residue\_name",
"residue\_name\_number","residue\_number", "residue\_range", "site", "species", "structure\_element",
"taxonomy\_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (URL and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat (URL More about the BioC
format can be found here: URL
Raw BioC XML files
------------------
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw\_BioC\_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_raw.xml.
Annotations in IOB format
-------------------------
The IOB formated files can be found in the directory: "annotation\_IOB"
The four files are as follows:
* URL --> all sentences and annotations used to create model
"mevol/BiomedNLP-PubMedBERT-ProteinStructure-NER-v1.2"; 1961 sentences
* URL --> training subset of the data; 1372 sentences
* URL --> development subset of the data; 294 sentences
* URL --> testing subset of the data; 295 sentences
The total number of annotations is: 10409
Annotations in BioC JSON
------------------------
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
+ "begin" --> character start position of annotation
+ "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
Annotations in BioC XML
-----------------------
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID\_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated\_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
+ "offset" --> start character position as defined by offset value
+ "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
Annotations in CSV
------------------
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation\_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno\_start" --> character start position of the annotation
* "anno\_end" --> character end position of the annotation
* "anno\_text" --> text covered by the annotation
* "entity\_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
Annotations in JSON
-------------------
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation\_JSON" under the name "URL".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
+ "sid" --> unique sentence ID
+ "sent" --> sentence text as string
+ "section" --> publication section the sentence is in
+ "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
|
[] |
[
"TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
[
27
] |
[
"passage: TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
27f52fbd973d35850f5a955e84cda146b13e5d5c
|
## Overview
This data was used to train model:
https://huggingface.co/mevol/BiomedNLP-PubMedBERT-ProteinStructure-NER-v1.4
There are 19 different entity types in this dataset:
"chemical", "complex_assembly", "evidence", "experimental_method", "gene", "mutant",
"oligomeric_state", "protein", "protein_state", "protein_type", "ptm", "residue_name",
"residue_name_number","residue_number", "residue_range", "site", "species", "structure_element",
"taxonomy_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (https://www.teamtat.org/) and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
| document ID | number of annotations in BioC XML | number of annotations in IOB/JSON/CSV | number of sentences |
| --- | --- | --- | --- |
| PMC4850273 | 1121 | 1121 | 204 |
| PMC4784909 | 865 | 865 | 204 |
| PMC4850288 | 716 | 708 | 146 |
| PMC4887326 | 933 | 933 | 152 |
| PMC4833862 | 1044 | 1044 | 192 |
| PMC4832331 | 739 | 718 | 134 |
| PMC4852598 | 1229 | 1218 | 250 |
| PMC4786784 | 1549 | 1549 | 232 |
| PMC4848090 | 987 | 985 | 191 |
| PMC4792962 | 1268 | 1268 | 256 |
| PMC4841544 | 1434 | 1433 | 273 |
| PMC4772114 | 825 | 825 | 166 |
| PMC4872110 | 1276 | 1276 | 253 |
| PMC4848761 | 887 | 883 | 252 |
| PMC4919469 | 1628 | 1616 | 336 |
| PMC4880283 | 771 | 771 | 166 |
| PMC4937829 | 625 | 625 | 181 |
| PMC4968113 | 1238 | 1238 | 292 |
| PMC4854314 | 481 | 471 | 139 |
| PMC4871749 | 383 | 383 | 76 |
| total | 19999 | 19930 | 4095 |
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: https://bioc.sourceforge.net/
## Raw BioC XML files
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw_BioC_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_raw.xml.
## Annotations in IOB format
The IOB formated files can be found in the directory: "annotation_IOB"
The four files are as follows:
* all.tsv --> all sentences and annotations used to create model
"mevol/BiomedNLP-PubMedBERT-ProteinStructure-NER-v1.4"; 4095 sentences
* train.tsv --> training subset of the data; 2866 sentences
* dev.tsv --> development subset of the data; 614 sentences
* test.tsv --> testing subset of the data; 615 sentences
The total number of annotations is: 19930
## Annotations in BioC JSON
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
* "begin" --> character start position of annotation
* "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
```json
[{"sourceid":"4784909",
"sourcedb":"",
"project":"",
"target":"",
"text":"",
"denotations":[{"span":{"begin":24,
"end":34},
"obj":"chemical,CHEBI:,[email protected],2023-03-21T15:19:42Z",
"id":"4500"},
{"span":{"begin":50,
"end":59},
"obj":"taxonomy_domain,DUMMY:,[email protected],2023-03-21T15:15:03Z",
"id":"1281"}]
}
]
```
## Annotations in BioC XML
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
* "offset" --> start character position as defined by offset value
* "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
```xml
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE collection SYSTEM "BioC.dtd">
<collection>
<source>PMC</source>
<date>20140719</date>
<key>pmc.key</key>
<document>
<id>4784909</id>
<passage>
<offset>0</offset>
<text>The Structural Basis of Coenzyme A Recycling in a Bacterial Organelle</text>
<annotation id="4500">
<infon key="type">chemical</infon>
<infon key="identifier">CHEBI:</infon>
<infon key="annotator">[email protected]</infon>
<infon key="updated_at">2023-03-21T15:19:42Z</infon>
<location offset="24" length="10"/>
<text>Coenzyme A</text>
</annotation>
</passage>
</document>
</collection>
```
## Annotations in CSV
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno_start" --> character start position of the annotation
* "anno_end" --> character end position of the annotation
* "anno_text" --> text covered by the annotation
* "entity_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
## Annotations in JSON
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation_JSON" under the name "annotations.json".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
* "sid" --> unique sentence ID
* "sent" --> sentence text as string
* "section" --> publication section the sentence is in
* "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
```json
{"PMC4850273": {"annotations":
[{"sid": 0,
"sent": "Molecular Dissection of Xyloglucan Recognition in a Prominent Human Gut Symbiont",
"section": "TITLE",
"ner": [
[24,34,"Xyloglucan","chemical"],
[62,67,"Human","species"],]
},]
}}
```
|
mevol/protein_structure_NER_model_v1.4
|
[
"language:en",
"license:mit",
"biology",
"protein structure",
"token classification",
"region:us"
] |
2023-10-24T14:04:04+00:00
|
{"language": ["en"], "license": "mit", "tags": ["biology", "protein structure", "token classification"], "configs": [{"config_name": "protein_structure_NER_model_v1.4", "data_files": [{"split": "train", "path": "annotation_IOB/train.tsv"}, {"split": "dev", "path": "annotation_IOB/dev.tsv"}, {"split": "test", "path": "annotation_IOB/test.tsv"}]}]}
|
2023-11-01T10:00:18+00:00
|
[] |
[
"en"
] |
TAGS
#language-English #license-mit #biology #protein structure #token classification #region-us
|
Overview
--------
This data was used to train model:
URL
There are 19 different entity types in this dataset:
"chemical", "complex\_assembly", "evidence", "experimental\_method", "gene", "mutant",
"oligomeric\_state", "protein", "protein\_state", "protein\_type", "ptm", "residue\_name",
"residue\_name\_number","residue\_number", "residue\_range", "site", "species", "structure\_element",
"taxonomy\_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (URL and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: URL
Raw BioC XML files
------------------
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw\_BioC\_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_raw.xml.
Annotations in IOB format
-------------------------
The IOB formated files can be found in the directory: "annotation\_IOB"
The four files are as follows:
* URL --> all sentences and annotations used to create model
"mevol/BiomedNLP-PubMedBERT-ProteinStructure-NER-v1.4"; 4095 sentences
* URL --> training subset of the data; 2866 sentences
* URL --> development subset of the data; 614 sentences
* URL --> testing subset of the data; 615 sentences
The total number of annotations is: 19930
Annotations in BioC JSON
------------------------
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
+ "begin" --> character start position of annotation
+ "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
Annotations in BioC XML
-----------------------
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID\_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated\_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
+ "offset" --> start character position as defined by offset value
+ "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
Annotations in CSV
------------------
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation\_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno\_start" --> character start position of the annotation
* "anno\_end" --> character end position of the annotation
* "anno\_text" --> text covered by the annotation
* "entity\_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
Annotations in JSON
-------------------
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation\_JSON" under the name "URL".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
+ "sid" --> unique sentence ID
+ "sent" --> sentence text as string
+ "section" --> publication section the sentence is in
+ "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
|
[] |
[
"TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
[
27
] |
[
"passage: TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
a2d2293a2225b9efb57cf8fd03b40b2b02395845
|
## Overview
This data was used to train model:
https://huggingface.co/mevol/BiomedNLP-PubMedBERT-ProteinStructure-NER-v2.1
There are 20 different entity types in this dataset:
"bond_interaction", "chemical", "complex_assembly", "evidence", "experimental_method", "gene",
"mutant", "oligomeric_state", "protein", "protein_state", "protein_type", "ptm", "residue_name",
"residue_name_number","residue_number", "residue_range", "site", "species", "structure_element",
"taxonomy_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (https://www.teamtat.org/) and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
| document ID | number of annotations in BioC XML | number of annotations in IOB/JSON/CSV | number of sentences |
| --- | --- | --- | --- |
| PMC4850273 | 1129 | 1129 | 205 |
| PMC4784909 | 868 | 868 | 204 |
| PMC4850288 | 718 | 710 | 146 |
| PMC4887326 | 942 | 942 | 152 |
| PMC4833862 | 1044 | 1044 | 192 |
| PMC4832331 | 739 | 718 | 134 |
| PMC4852598 | 1239 | 1228 | 250 |
| PMC4786784 | 1573 | 1573 | 232 |
| PMC4848090 | 1002 | 1000 | 192 |
| PMC4792962 | 1297 | 1297 | 256 |
| PMC4841544 | 1460 | 1459 | 274 |
| PMC4772114 | 824 | 824 | 165 |
| PMC4872110 | 1283 | 1283 | 250 |
| PMC4848761 | 888 | 884 | 252 |
| PMC4919469 | 1636 | 1624 | 336 |
| PMC4880283 | 783 | 783 | 166 |
| PMC4968113 | 1245 | 1245 | 292 |
| PMC4937829 | 633 | 633 | 181 |
| PMC4854314 | 498 | 488 | 139 |
| PMC4871749 | 411 | 411 | 79 |
| PMC4869123 | 922 | 922 | 195 |
| PMC4888278 | 580 | 580 | 102 |
| PMC4795551 | 1475 | 1475 | 297 |
| PMC4831588 | 1087 | 1070 | 224 |
| PMC4918766 | 1027 | 1027 | 210 |
| PMC4802042 | 1445 | 1445 | 268 |
| PMC4896748 | 2652 | 2638 | 480 |
| PMC4781976 | 115 | 113 | 24 |
| PMC4802085 | 983 | 983 | 193 |
| PMC4887163 | 856 | 856 | 196|
| total | 31354 | 31252 | 6286 |
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: https://bioc.sourceforge.net/
## Raw BioC XML files
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw_BioC_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_raw.xml.
## Annotations in IOB format
The IOB formated files can be found in the directory: "annotation_IOB"
The four files are as follows:
* all.tsv --> all sentences and annotations used to create model
"mevol/BiomedNLP-PubMedBERT-ProteinStructure-NER-v2.1"; 6286 sentences
* train.tsv --> training subset of the data; 4400 sentences
* dev.tsv --> development subset of the data; 943 sentences
* test.tsv --> testing subset of the data; 943 sentences
The total number of annotations is: 31252
## Annotations in BioC JSON
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
* "begin" --> character start position of annotation
* "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
```json
[{"sourceid":"4784909",
"sourcedb":"",
"project":"",
"target":"",
"text":"",
"denotations":[{"span":{"begin":24,
"end":34},
"obj":"chemical,CHEBI:,[email protected],2023-03-21T15:19:42Z",
"id":"4500"},
{"span":{"begin":50,
"end":59},
"obj":"taxonomy_domain,DUMMY:,[email protected],2023-03-21T15:15:03Z",
"id":"1281"}]
}
]
```
## Annotations in BioC XML
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
* "offset" --> start character position as defined by offset value
* "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
```xml
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE collection SYSTEM "BioC.dtd">
<collection>
<source>PMC</source>
<date>20140719</date>
<key>pmc.key</key>
<document>
<id>4784909</id>
<passage>
<offset>0</offset>
<text>The Structural Basis of Coenzyme A Recycling in a Bacterial Organelle</text>
<annotation id="4500">
<infon key="type">chemical</infon>
<infon key="identifier">CHEBI:</infon>
<infon key="annotator">[email protected]</infon>
<infon key="updated_at">2023-03-21T15:19:42Z</infon>
<location offset="24" length="10"/>
<text>Coenzyme A</text>
</annotation>
</passage>
</document>
</collection>
```
## Annotations in CSV
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno_start" --> character start position of the annotation
* "anno_end" --> character end position of the annotation
* "anno_text" --> text covered by the annotation
* "entity_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
## Annotations in JSON
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation_JSON" under the name "annotations.json".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
* "sid" --> unique sentence ID
* "sent" --> sentence text as string
* "section" --> publication section the sentence is in
* "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
```json
{"PMC4850273": {"annotations":
[{"sid": 0,
"sent": "Molecular Dissection of Xyloglucan Recognition in a Prominent Human Gut Symbiont",
"section": "TITLE",
"ner": [
[24,34,"Xyloglucan","chemical"],
[62,67,"Human","species"],]
},]
}}
```
|
mevol/protein_structure_NER_model_v2.1
|
[
"language:en",
"license:mit",
"biology",
"protein structure",
"token classification",
"region:us"
] |
2023-10-24T14:05:05+00:00
|
{"language": ["en"], "license": "mit", "tags": ["biology", "protein structure", "token classification"], "configs": [{"config_name": "protein_structure_NER_model_v2.1", "data_files": [{"split": "train", "path": "annotation_IOB/train.tsv"}, {"split": "dev", "path": "annotation_IOB/dev.tsv"}, {"split": "test", "path": "annotation_IOB/test.tsv"}]}]}
|
2023-11-01T10:19:00+00:00
|
[] |
[
"en"
] |
TAGS
#language-English #license-mit #biology #protein structure #token classification #region-us
|
Overview
--------
This data was used to train model:
URL
There are 20 different entity types in this dataset:
"bond\_interaction", "chemical", "complex\_assembly", "evidence", "experimental\_method", "gene",
"mutant", "oligomeric\_state", "protein", "protein\_state", "protein\_type", "ptm", "residue\_name",
"residue\_name\_number","residue\_number", "residue\_range", "site", "species", "structure\_element",
"taxonomy\_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (URL and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: URL
Raw BioC XML files
------------------
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw\_BioC\_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_raw.xml.
Annotations in IOB format
-------------------------
The IOB formated files can be found in the directory: "annotation\_IOB"
The four files are as follows:
* URL --> all sentences and annotations used to create model
"mevol/BiomedNLP-PubMedBERT-ProteinStructure-NER-v2.1"; 6286 sentences
* URL --> training subset of the data; 4400 sentences
* URL --> development subset of the data; 943 sentences
* URL --> testing subset of the data; 943 sentences
The total number of annotations is: 31252
Annotations in BioC JSON
------------------------
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
+ "begin" --> character start position of annotation
+ "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
Annotations in BioC XML
-----------------------
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID\_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated\_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
+ "offset" --> start character position as defined by offset value
+ "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
Annotations in CSV
------------------
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation\_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno\_start" --> character start position of the annotation
* "anno\_end" --> character end position of the annotation
* "anno\_text" --> text covered by the annotation
* "entity\_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
Annotations in JSON
-------------------
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation\_JSON" under the name "URL".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
+ "sid" --> unique sentence ID
+ "sent" --> sentence text as string
+ "section" --> publication section the sentence is in
+ "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
|
[] |
[
"TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
[
27
] |
[
"passage: TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
af353a7d9b9cf6e721859a4f1389f2ef001bffb8
|
## Overview
This data was used to train model:
https://huggingface.co/mevol/BiomedNLP-PubMedBERT-ProteinStructure-NER-v3.1
There are 20 different entity types in this dataset:
"bond_interaction", "chemical", "complex_assembly", "evidence", "experimental_method", "gene",
"mutant", "oligomeric_state", "protein", "protein_state", "protein_type", "ptm", "residue_name",
"residue_name_number","residue_number", "residue_range", "site", "species", "structure_element",
"taxonomy_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (https://www.teamtat.org/) and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
| document ID | number of annotations in BioC XML | number of annotations in IOB/JSON/CSV | number of sentences |
| --- | --- | --- | --- |
| PMC4850273 | 1129 | 1129 | 205 |
| PMC4784909 | 868 | 868 | 204 |
| PMC4850288 | 717 | 709 | 146 |
| PMC4887326 | 942 | 942 | 152 |
| PMC4833862 | 1044 | 1044 | 192 |
| PMC4832331 | 739 | 718 | 134 |
| PMC4852598 | 1239 | 1228 | 250 |
| PMC4786784 | 1573 | 1573 | 232 |
| PMC4848090 | 1000 | 998 | 192 |
| PMC4792962 | 1297 | 1297 | 256 |
| PMC4841544 | 1460 | 1459 | 274 |
| PMC4772114 | 824 | 824 | 165 |
| PMC4872110 | 1283 | 1283 | 250 |
| PMC4848761 | 888 | 884 | 252 |
| PMC4919469 | 1636 | 1624 | 336 |
| PMC4880283 | 783 | 783 | 166 |
| PMC4968113 | 1245 | 1245 | 292 |
| PMC4937829 | 633 | 633 | 181 |
| PMC4854314 | 498 | 488 | 139 |
| PMC4871749 | 411 | 411 | 79 |
| PMC4869123 | 922 | 922 | 195 |
| PMC4888278 | 580 | 580 | 102 |
| PMC4795551 | 1475 | 1475 | 297 |
| PMC4831588 | 1087 | 1070 | 224 |
| PMC4918766 | 1027 | 1027 | 210 |
| PMC4802042 | 1441 | 1441 | 264 |
| PMC4896748 | 2652 | 2638 | 480 |
| PMC4781976 | 115 | 113 | 24 |
| PMC4802085 | 983 | 983 | 193 |
| PMC4887163 | 856 | 856 | 196|
| PMC4918759 | 803 | 803 | 175 |
| PMC4855620 | 563 | 563 | 122 |
| PMC4822050 | 1521 | 1521 | 249 |
| PMC4822561 | 367 | 366 | 84 |
| PMC4885502 | 577 | 577 | 97 |
| PMC4746701 | 1130 | 1130 | 245 |
| PMC4820378 | 733 | 733 | 170 |
| PMC4773095 | 1323 | 1323 | 252 |
| PMC4857006 | 1358 | 1358 | 249 |
| PMC4774019 | 532 | 530 | 117 |
| total | 40254 | 40149 | 8042 |
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: https://bioc.sourceforge.net/
## Raw BioC XML files
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw_BioC_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_raw.xml.
## Annotations in IOB format
The IOB formated files can be found in the directory: "annotation_IOB"
The four files are as follows:
* all.tsv --> all sentences and annotations used to create model
"mevol/BiomedNLP-PubMedBERT-ProteinStructure-NER-v3.1"; 8042 sentences
* train.tsv --> training subset of the data; 5629 sentences
* dev.tsv --> development subset of the data; 1206 sentences
* test.tsv --> testing subset of the data; 1207 sentences
The total number of annotations is: 40149
## Annotations in BioC JSON
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
* "begin" --> character start position of annotation
* "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
```json
[{"sourceid":"4784909",
"sourcedb":"",
"project":"",
"target":"",
"text":"",
"denotations":[{"span":{"begin":24,
"end":34},
"obj":"chemical,CHEBI:,[email protected],2023-03-21T15:19:42Z",
"id":"4500"},
{"span":{"begin":50,
"end":59},
"obj":"taxonomy_domain,DUMMY:,[email protected],2023-03-21T15:15:03Z",
"id":"1281"}]
}
]
```
## Annotations in BioC XML
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated_BioC_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
* "offset" --> start character position as defined by offset value
* "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
```xml
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE collection SYSTEM "BioC.dtd">
<collection>
<source>PMC</source>
<date>20140719</date>
<key>pmc.key</key>
<document>
<id>4784909</id>
<passage>
<offset>0</offset>
<text>The Structural Basis of Coenzyme A Recycling in a Bacterial Organelle</text>
<annotation id="4500">
<infon key="type">chemical</infon>
<infon key="identifier">CHEBI:</infon>
<infon key="annotator">[email protected]</infon>
<infon key="updated_at">2023-03-21T15:19:42Z</infon>
<location offset="24" length="10"/>
<text>Coenzyme A</text>
</annotation>
</passage>
</document>
</collection>
```
## Annotations in CSV
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno_start" --> character start position of the annotation
* "anno_end" --> character end position of the annotation
* "anno_text" --> text covered by the annotation
* "entity_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
## Annotations in JSON
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation_JSON" under the name "annotations.json".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
* "sid" --> unique sentence ID
* "sent" --> sentence text as string
* "section" --> publication section the sentence is in
* "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
```json
{"PMC4850273": {"annotations":
[{"sid": 0,
"sent": "Molecular Dissection of Xyloglucan Recognition in a Prominent Human Gut Symbiont",
"section": "TITLE",
"ner": [
[24,34,"Xyloglucan","chemical"],
[62,67,"Human","species"],]
},]
}}
```
|
mevol/protein_structure_NER_model_v3.1
|
[
"language:en",
"license:mit",
"biology",
"protein structure",
"token classification",
"region:us"
] |
2023-10-24T14:06:19+00:00
|
{"language": ["en"], "license": "mit", "tags": ["biology", "protein structure", "token classification"], "configs": [{"config_name": "protein_structure_NER_model_v3.1", "data_files": [{"split": "train", "path": "annotation_IOB/train.tsv"}, {"split": "dev", "path": "annotation_IOB/dev.tsv"}, {"split": "test", "path": "annotation_IOB/test.tsv"}]}]}
|
2023-11-01T10:27:27+00:00
|
[] |
[
"en"
] |
TAGS
#language-English #license-mit #biology #protein structure #token classification #region-us
|
Overview
--------
This data was used to train model:
URL
There are 20 different entity types in this dataset:
"bond\_interaction", "chemical", "complex\_assembly", "evidence", "experimental\_method", "gene",
"mutant", "oligomeric\_state", "protein", "protein\_state", "protein\_type", "ptm", "residue\_name",
"residue\_name\_number","residue\_number", "residue\_range", "site", "species", "structure\_element",
"taxonomy\_domain"
The data prepared as IOB formated input has been used during training, develiopment
and testing. Additional data formats such as JSON and XML as well as CSV files are
also available and are described below.
Annotation was carried out with the free annotation tool TeamTat (URL and
documents were downloaded as BioC XML before converting them to IOB, annotation only JSON and CSV format.
The number of annotations and sentences in each file is given below:
Documents and annotations are easiest viewed by using the BioC XML files and opening
them in free annotation tool TeamTat. More about the BioC
format can be found here: URL
Raw BioC XML files
------------------
These are the raw, un-annotated XML files for the publications in the dataset in BioC format.
The files are found in the directory: "raw\_BioC\_XML".
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_raw.xml.
Annotations in IOB format
-------------------------
The IOB formated files can be found in the directory: "annotation\_IOB"
The four files are as follows:
* URL --> all sentences and annotations used to create model
"mevol/BiomedNLP-PubMedBERT-ProteinStructure-NER-v3.1"; 8042 sentences
* URL --> training subset of the data; 5629 sentences
* URL --> development subset of the data; 1206 sentences
* URL --> testing subset of the data; 1207 sentences
The total number of annotations is: 40149
Annotations in BioC JSON
------------------------
The BioC formated JSON files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_JSON"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID"\_ann.json
Each document JSON contains the following relevant keys:
* "sourceid" --> giving the numerical part of the unique PubMedCentral ID
* "text" --> containing the complete raw text of the publication as a string
* "denotations" --> containing a list of all the annotations for the text
Each annotation is a dictionary with the following keys:
* "span" --> gives the start and end of the annotatiom span defined by sub keys:
+ "begin" --> character start position of annotation
+ "end" --> character end position of annotation
* "obj" --> a string containing a number of terms that can be separated by ","; the order
of the terms gives the following: entity type, reference to ontology, annotator,
time stamp
* "id" --> unique annotation ID
Here an example:
Annotations in BioC XML
-----------------------
The BioC formated XML files of the publications have been downloaded from the annotation
tool TeamTat. The files are found in the directory: "annotated\_BioC\_XML"
There is one file for each document and they follow standard naming
"unique PubMedCentral ID\_ann.xml
The key XML tags to be able to visualise the annotations in TeamTat as well as extracting
them to create the training data are "passage" and "offset". The "passage" tag encloses a
text passage or paragraph to which the annotations are linked. "Offset" gives the passage/
paragraph offset and allows to determine the character starting and ending postions of the
annotations. The tag "text" encloses the raw text of the passage.
Each annotation in the XML file is tagged as below:
* "annotation id=" --> giving the unique ID of the annotation
* "infon key="type"" --> giving the entity type of the annotation
* "infon key="identifier"" --> giving a reference to an ontology for the annotation
* "infon key="annotator"" --> giving the annotator
* "infon key="updated\_at"" --> providing a time stamp for annotation creation/update
* "location" --> start and end character positions for the annotated text span
+ "offset" --> start character position as defined by offset value
+ "length" --> length of the annotation span; sum of "offset" and "length" creates
the end character position
Here is a basic example of what the BioC XML looks like. Additional tags for document
management are not given. Please refer to the documenttation to find out more.
Annotations in CSV
------------------
The annotations and the relevant sentences they have been found in have also been made
available as tab-separated CSV files, one for each publication in the dataset. The files can
be found in directory "annotation\_CSV". Each file is named as "unique PubMedCentral ID".csv.
The column labels in the CSV files are as follows:
* "anno\_start" --> character start position of the annotation
* "anno\_end" --> character end position of the annotation
* "anno\_text" --> text covered by the annotation
* "entity\_type" --> entity type of the annotation
* "sentence" --> sentence text in which the annotation was found
* "section" --> publication section in which the annotation was found
Annotations in JSON
-------------------
A combined JSON file was created only containing the relevant sentences and associated
annotations for each publication in the dataset. The file can be found in directory
"annotation\_JSON" under the name "URL".
The following keys are used:
* "PMC4850273" --> unique PubMedCentral of the publication
* "annotations" --> list of dictionaries for the relevant, annotated sentences of the
document; each dictionary has the following sub keys
+ "sid" --> unique sentence ID
+ "sent" --> sentence text as string
+ "section" --> publication section the sentence is in
+ "ner" --> nested list of annotations; each sublist contains the following items:
start character position, end character position, annotation text,
entity type
Here is an example of a sentence and its annotations:
|
[] |
[
"TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
[
27
] |
[
"passage: TAGS\n#language-English #license-mit #biology #protein structure #token classification #region-us \n"
] |
f6e0f43af24049ed2ff07892efa464891fb9d83a
|
# Dataset Card for "e-ner-roberta-base"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
lltala/e-ner-roberta-base
|
[
"region:us"
] |
2023-10-24T14:17:00+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "validation", "path": "data/validation-*"}]}], "dataset_info": {"features": [{"name": "doc_id", "dtype": "string"}, {"name": "id", "dtype": "string"}, {"name": "ner_tags", "sequence": {"class_label": {"names": {"0": "O", "1": "B-PER", "2": "I-PER", "3": "B-ORG", "4": "I-ORG", "5": "B-LOC", "6": "I-LOC"}}}}, {"name": "tokens", "sequence": "string"}], "splits": [{"name": "train", "num_bytes": 6380478, "num_examples": 840}, {"name": "validation", "num_bytes": 676038, "num_examples": 90}], "download_size": 776863, "dataset_size": 7056516}}
|
2023-10-24T14:17:03+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "e-ner-roberta-base"
More Information needed
|
[
"# Dataset Card for \"e-ner-roberta-base\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"e-ner-roberta-base\"\n\nMore Information needed"
] |
[
6,
18
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"e-ner-roberta-base\"\n\nMore Information needed"
] |
c6f57fd67643e45df43f8f93f1fced31216dce28
|
# Dataset Card for "ocr_correction"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
pablouribe/ocr_correction
|
[
"region:us"
] |
2023-10-24T14:35:18+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "test", "path": "data/test-*"}]}], "dataset_info": {"features": [{"name": "text", "dtype": "string"}, {"name": "ocr_text", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 141292368.8, "num_examples": 12800}, {"name": "test", "num_bytes": 35323092.2, "num_examples": 3200}], "download_size": 106441640, "dataset_size": 176615461.0}}
|
2023-10-24T14:35:35+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "ocr_correction"
More Information needed
|
[
"# Dataset Card for \"ocr_correction\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"ocr_correction\"\n\nMore Information needed"
] |
[
6,
15
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"ocr_correction\"\n\nMore Information needed"
] |
cad3867335327626039a4e75fad0abdd8976fdc3
|
<img src="https://cdn-uploads.huggingface.co/production/uploads/5df9c78eda6d0311fd3d541f/CmusIT5OlSXvFrbTJ7l-C.png" alt="Taiwan LLM Logo" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
## Performance
Data source: https://github.com/mtkresearch/MR-Models/tree/main/TC-Eval
# Citation for Taiwan LLM
```
@misc{lin2023taiwan,
title={Taiwan LLM: Bridging the Linguistic Divide with a Culturally Aligned Language Model},
author={Yen-Ting Lin and Yun-Nung Chen},
year={2023},
eprint={2311.17487},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
|
yentinglin/TC-Eval
|
[
"task_categories:question-answering",
"task_categories:text-classification",
"size_categories:1K<n<10K",
"language:zh",
"arxiv:2311.17487",
"region:us"
] |
2023-10-24T14:58:14+00:00
|
{"language": ["zh"], "size_categories": ["1K<n<10K"], "task_categories": ["question-answering", "text-classification"], "pretty_name": "TMLU", "configs": [{"config_name": "FGC", "data_files": [{"split": "test", "path": "fgc.jsonl"}]}, {"config_name": "DRCD", "data_files": [{"split": "test", "path": "drcd.jsonl"}]}, {"config_name": "TMMLU", "data_files": [{"split": "test", "path": "tmmlu.jsonl"}]}, {"config_name": "XSUM_TC", "data_files": [{"split": "test", "path": "xsum.jsonl"}]}, {"config_name": "TTQA", "data_files": [{"split": "test", "path": "TTQA_mc.jsonl"}]}, {"config_name": "TCIC", "data_files": [{"split": "test", "path": "TCIC_1.2.0.json"}]}, {"config_name": "PenguinsInTable_TC", "data_files": [{"split": "test", "path": "PenguinsInTable_TC.jsonl"}]}, {"config_name": "IMDB_TC", "data_files": [{"split": "test", "path": "imdb.jsonl"}]}]}
|
2023-12-01T02:38:36+00:00
|
[
"2311.17487"
] |
[
"zh"
] |
TAGS
#task_categories-question-answering #task_categories-text-classification #size_categories-1K<n<10K #language-Chinese #arxiv-2311.17487 #region-us
|
<img src="URL alt="Taiwan LLM Logo" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
## Performance
!image/png
Data source: URL
for Taiwan LLM
|
[
"## Performance\n\n\n!image/png\n\n\nData source: URL\n\nfor Taiwan LLM"
] |
[
"TAGS\n#task_categories-question-answering #task_categories-text-classification #size_categories-1K<n<10K #language-Chinese #arxiv-2311.17487 #region-us \n",
"## Performance\n\n\n!image/png\n\n\nData source: URL\n\nfor Taiwan LLM"
] |
[
55,
14
] |
[
"passage: TAGS\n#task_categories-question-answering #task_categories-text-classification #size_categories-1K<n<10K #language-Chinese #arxiv-2311.17487 #region-us \n## Performance\n\n\n!image/png\n\n\nData source: URL\n\nfor Taiwan LLM"
] |
5e4a598e7cfe4336318f71e2664c2452c815353e
|
# Dataset Card for "helm-samsum-dolly-lima"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
quyanh/helm-samsum-dolly-lima
|
[
"region:us"
] |
2023-10-24T15:03:51+00:00
|
{"dataset_info": {"features": [{"name": "prompt", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 22189219.45245881, "num_examples": 21963}], "download_size": 17712616, "dataset_size": 22189219.45245881}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-25T08:27:23+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "helm-samsum-dolly-lima"
More Information needed
|
[
"# Dataset Card for \"helm-samsum-dolly-lima\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"helm-samsum-dolly-lima\"\n\nMore Information needed"
] |
[
6,
20
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"helm-samsum-dolly-lima\"\n\nMore Information needed"
] |
4cd977073b94c7bf4b5804e666cc2e5663a0dca3
|
# Dataset Card for "Romanian Bible Paraphrase Corpus"
## Table of Contents
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)
## Dataset Description
<!--
- **Paper:** Fake News Opensources
-->
- **Homepage:** [https://github.com/AndyTheFactory/ro-paraphrase-bible](https://github.com/AndyTheFactory/ro-paraphrase-bible)
- **Repository:** [https://github.com/AndyTheFactory/ro-paraphrase-bible](https://github.com/AndyTheFactory/ro-paraphrase-bible)
- **Point of Contact:** [Andrei Paraschiv](https://github.com/AndyTheFactory)
-
### Dataset Summary
A paraphprase corpus created from 10 different Romanian language Bible versions. Since the Bible has all paragraphs uniquely numbered an alignment between two
versions is straighforward.
We compiled a complete combination of paragraph pairs from the following Bible versions:
- Română Noul Testament Interconfesional 2009
- Biblia în versuri 2014
- Biblia Traducerea Fidela 2015
- Biblia în Versiune Actualizată 2018
- Ediția Dumitru Cornilescu revizuită 2022
- Noua Traducere Românească
- Noul Testament SBR 2023
- Versiunea Biblia Romano-Catolică 2020
- Biblia sau Sfânta Scriptură cu Trimiteri 1924, Dumitru Cornilescu
- Traducere Literală Cornilescu 1931
In order to provide a similarity score between two paragraphs, we use FuzzyWuzzy library to compute the partial_token_sort_ratio (PTSoR)
between the lemmatized versions of the two texts. The final similarity score between matching paragraphs was computed as follows:
> score = 50 + ptsor // 2
where ptsor = partial_token_sort_ratio(lemmatize(text1), lemmatize(text2))
We also provide a set of non-matching texts, generated from a random pairing of paragraphs.
The score for the non-matching paragraphs was computed using this formula:
> score = int(ptsor / 2 +10)
The final dataset contains 904,815 similar records and 218,977 non matching records, totaling 1,123,927
### Languages
Romanian
## Dataset Structure
### Data Instances
An example record looks as follows.
```
{
'text1': 'Poporul a început să strige: „Este glasul lui Dumnezeu, nu al unui om!”',
'text2': 'Norodul a strigat: „Glas de Dumnezeu, nu de om!”',
'score': 82
}
```
### Data Fields
- `text1`, `text2`: compared Paragraphs
- `score`: 0 - 100 matching score (integer) - records with score > 50 are matches
### Data Statistics
**Score Distribution**:
Statistics for matching record scores:
```
count 904950.000000
mean 83.998228
std 7.083741
min 50.000000
25% 80.000000
50% 84.000000
75% 89.000000
max 100.000000
```
Statistics for non-matching record scores:
```
count 218977.000000
mean 29.400873
std 3.339921
min 10.000000
25% 27.000000
50% 29.000000
75% 32.000000
max 40.000000
```
## Dataset Creation
### Curation Rationale
### Source Data
Romanian Language Bible translations
#### Initial Data Collection and Normalization
#### Who are the source language producers?
### Annotations
#### Annotation process
#### Who are the annotators?
### Personal and Sensitive Information
## Considerations for Using the Data
### Social Impact of Dataset
### Discussion of Biases
### Other Known Limitations
The dataset is biased towards some archaic language. Additionally, it involves religious figures and phrase constructions that are rarly used in the common language.
## Additional Information
### Dataset Curators
### Licensing Information
This data is available and distributed under Apache-2.0 license
### Citation Information
```
tbd
```
### Contributions
|
andyP/ro-paraphrase-bible
|
[
"task_categories:sentence-similarity",
"task_ids:text-scoring",
"task_ids:semantic-similarity-scoring",
"task_ids:semantic-similarity-classification",
"annotations_creators:machine-generated",
"language_creators:found",
"multilinguality:monolingual",
"size_categories:100K<n<1M",
"language:ro",
"license:apache-2.0",
"sentence similarity",
"paraphrase",
"romanian",
"nlp",
"doi:10.57967/hf/1712",
"region:us"
] |
2023-10-24T15:09:37+00:00
|
{"annotations_creators": ["machine-generated"], "language_creators": ["found"], "language": ["ro"], "license": "apache-2.0", "multilinguality": ["monolingual"], "size_categories": ["100K<n<1M"], "task_categories": ["sentence-similarity"], "task_ids": ["text-scoring", "semantic-similarity-scoring", "semantic-similarity-classification"], "pretty_name": "Romanian Bible Paraphrase Corpus", "tags": ["sentence similarity", "paraphrase", "romanian", "nlp"], "dataset_info": {"features": [{"name": "text1", "dtype": "string"}, {"name": "text2", "dtype": "string"}, {"name": "score", "dtype": "int8"}]}}
|
2024-01-29T21:59:39+00:00
|
[] |
[
"ro"
] |
TAGS
#task_categories-sentence-similarity #task_ids-text-scoring #task_ids-semantic-similarity-scoring #task_ids-semantic-similarity-classification #annotations_creators-machine-generated #language_creators-found #multilinguality-monolingual #size_categories-100K<n<1M #language-Romanian #license-apache-2.0 #sentence similarity #paraphrase #romanian #nlp #doi-10.57967/hf/1712 #region-us
|
# Dataset Card for "Romanian Bible Paraphrase Corpus"
## Table of Contents
- Dataset Description
- Dataset Summary
- Supported Tasks and Leaderboards
- Languages
- Dataset Structure
- Data Instances
- Data Fields
- Data Splits
- Dataset Creation
- Curation Rationale
- Source Data
- Annotations
- Personal and Sensitive Information
- Considerations for Using the Data
- Social Impact of Dataset
- Discussion of Biases
- Other Known Limitations
- Additional Information
- Dataset Curators
- Licensing Information
- Citation Information
- Contributions
## Dataset Description
- Homepage: URL
- Repository: URL
- Point of Contact: Andrei Paraschiv
-
### Dataset Summary
A paraphprase corpus created from 10 different Romanian language Bible versions. Since the Bible has all paragraphs uniquely numbered an alignment between two
versions is straighforward.
We compiled a complete combination of paragraph pairs from the following Bible versions:
- Română Noul Testament Interconfesional 2009
- Biblia în versuri 2014
- Biblia Traducerea Fidela 2015
- Biblia în Versiune Actualizată 2018
- Ediția Dumitru Cornilescu revizuită 2022
- Noua Traducere Românească
- Noul Testament SBR 2023
- Versiunea Biblia Romano-Catolică 2020
- Biblia sau Sfânta Scriptură cu Trimiteri 1924, Dumitru Cornilescu
- Traducere Literală Cornilescu 1931
In order to provide a similarity score between two paragraphs, we use FuzzyWuzzy library to compute the partial_token_sort_ratio (PTSoR)
between the lemmatized versions of the two texts. The final similarity score between matching paragraphs was computed as follows:
> score = 50 + ptsor // 2
where ptsor = partial_token_sort_ratio(lemmatize(text1), lemmatize(text2))
We also provide a set of non-matching texts, generated from a random pairing of paragraphs.
The score for the non-matching paragraphs was computed using this formula:
> score = int(ptsor / 2 +10)
The final dataset contains 904,815 similar records and 218,977 non matching records, totaling 1,123,927
### Languages
Romanian
## Dataset Structure
### Data Instances
An example record looks as follows.
### Data Fields
- 'text1', 'text2': compared Paragraphs
- 'score': 0 - 100 matching score (integer) - records with score > 50 are matches
### Data Statistics
Score Distribution:
!Score distribution
Statistics for matching record scores:
Statistics for non-matching record scores:
## Dataset Creation
### Curation Rationale
### Source Data
Romanian Language Bible translations
#### Initial Data Collection and Normalization
#### Who are the source language producers?
### Annotations
#### Annotation process
#### Who are the annotators?
### Personal and Sensitive Information
## Considerations for Using the Data
### Social Impact of Dataset
### Discussion of Biases
### Other Known Limitations
The dataset is biased towards some archaic language. Additionally, it involves religious figures and phrase constructions that are rarly used in the common language.
## Additional Information
### Dataset Curators
### Licensing Information
This data is available and distributed under Apache-2.0 license
### Contributions
|
[
"# Dataset Card for \"Romanian Bible Paraphrase Corpus\"",
"## Table of Contents\n- Dataset Description\n - Dataset Summary\n - Supported Tasks and Leaderboards\n - Languages\n- Dataset Structure\n - Data Instances\n - Data Fields\n - Data Splits\n- Dataset Creation\n - Curation Rationale\n - Source Data\n - Annotations\n - Personal and Sensitive Information\n- Considerations for Using the Data\n - Social Impact of Dataset\n - Discussion of Biases\n - Other Known Limitations\n- Additional Information\n - Dataset Curators\n - Licensing Information\n - Citation Information\n - Contributions",
"## Dataset Description\n\n- Homepage: URL\n- Repository: URL\n- Point of Contact: Andrei Paraschiv\n-",
"### Dataset Summary\n\nA paraphprase corpus created from 10 different Romanian language Bible versions. Since the Bible has all paragraphs uniquely numbered an alignment between two \nversions is straighforward. \nWe compiled a complete combination of paragraph pairs from the following Bible versions:\n\n- Română Noul Testament Interconfesional 2009\n- Biblia în versuri 2014\n- Biblia Traducerea Fidela 2015\n- Biblia în Versiune Actualizată 2018\n- Ediția Dumitru Cornilescu revizuită 2022\n- Noua Traducere Românească\n- Noul Testament SBR 2023\n- Versiunea Biblia Romano-Catolică 2020\n- Biblia sau Sfânta Scriptură cu Trimiteri 1924, Dumitru Cornilescu\n- Traducere Literală Cornilescu 1931\n\nIn order to provide a similarity score between two paragraphs, we use FuzzyWuzzy library to compute the partial_token_sort_ratio (PTSoR)\nbetween the lemmatized versions of the two texts. The final similarity score between matching paragraphs was computed as follows:\n\n > score = 50 + ptsor // 2\n\nwhere ptsor = partial_token_sort_ratio(lemmatize(text1), lemmatize(text2))\n\n\nWe also provide a set of non-matching texts, generated from a random pairing of paragraphs. \nThe score for the non-matching paragraphs was computed using this formula:\n\n > score = int(ptsor / 2 +10)\n\nThe final dataset contains 904,815 similar records and 218,977 non matching records, totaling 1,123,927",
"### Languages\n\nRomanian",
"## Dataset Structure",
"### Data Instances\n\n\nAn example record looks as follows.",
"### Data Fields\n\n- 'text1', 'text2': compared Paragraphs\n- 'score': 0 - 100 matching score (integer) - records with score > 50 are matches",
"### Data Statistics\nScore Distribution:\n\n!Score distribution\n\n\n\nStatistics for matching record scores:\n\n\nStatistics for non-matching record scores:",
"## Dataset Creation",
"### Curation Rationale",
"### Source Data\n\nRomanian Language Bible translations",
"#### Initial Data Collection and Normalization",
"#### Who are the source language producers?",
"### Annotations",
"#### Annotation process",
"#### Who are the annotators?",
"### Personal and Sensitive Information",
"## Considerations for Using the Data",
"### Social Impact of Dataset",
"### Discussion of Biases",
"### Other Known Limitations\n\nThe dataset is biased towards some archaic language. Additionally, it involves religious figures and phrase constructions that are rarly used in the common language.",
"## Additional Information",
"### Dataset Curators",
"### Licensing Information\n\nThis data is available and distributed under Apache-2.0 license",
"### Contributions"
] |
[
"TAGS\n#task_categories-sentence-similarity #task_ids-text-scoring #task_ids-semantic-similarity-scoring #task_ids-semantic-similarity-classification #annotations_creators-machine-generated #language_creators-found #multilinguality-monolingual #size_categories-100K<n<1M #language-Romanian #license-apache-2.0 #sentence similarity #paraphrase #romanian #nlp #doi-10.57967/hf/1712 #region-us \n",
"# Dataset Card for \"Romanian Bible Paraphrase Corpus\"",
"## Table of Contents\n- Dataset Description\n - Dataset Summary\n - Supported Tasks and Leaderboards\n - Languages\n- Dataset Structure\n - Data Instances\n - Data Fields\n - Data Splits\n- Dataset Creation\n - Curation Rationale\n - Source Data\n - Annotations\n - Personal and Sensitive Information\n- Considerations for Using the Data\n - Social Impact of Dataset\n - Discussion of Biases\n - Other Known Limitations\n- Additional Information\n - Dataset Curators\n - Licensing Information\n - Citation Information\n - Contributions",
"## Dataset Description\n\n- Homepage: URL\n- Repository: URL\n- Point of Contact: Andrei Paraschiv\n-",
"### Dataset Summary\n\nA paraphprase corpus created from 10 different Romanian language Bible versions. Since the Bible has all paragraphs uniquely numbered an alignment between two \nversions is straighforward. \nWe compiled a complete combination of paragraph pairs from the following Bible versions:\n\n- Română Noul Testament Interconfesional 2009\n- Biblia în versuri 2014\n- Biblia Traducerea Fidela 2015\n- Biblia în Versiune Actualizată 2018\n- Ediția Dumitru Cornilescu revizuită 2022\n- Noua Traducere Românească\n- Noul Testament SBR 2023\n- Versiunea Biblia Romano-Catolică 2020\n- Biblia sau Sfânta Scriptură cu Trimiteri 1924, Dumitru Cornilescu\n- Traducere Literală Cornilescu 1931\n\nIn order to provide a similarity score between two paragraphs, we use FuzzyWuzzy library to compute the partial_token_sort_ratio (PTSoR)\nbetween the lemmatized versions of the two texts. The final similarity score between matching paragraphs was computed as follows:\n\n > score = 50 + ptsor // 2\n\nwhere ptsor = partial_token_sort_ratio(lemmatize(text1), lemmatize(text2))\n\n\nWe also provide a set of non-matching texts, generated from a random pairing of paragraphs. \nThe score for the non-matching paragraphs was computed using this formula:\n\n > score = int(ptsor / 2 +10)\n\nThe final dataset contains 904,815 similar records and 218,977 non matching records, totaling 1,123,927",
"### Languages\n\nRomanian",
"## Dataset Structure",
"### Data Instances\n\n\nAn example record looks as follows.",
"### Data Fields\n\n- 'text1', 'text2': compared Paragraphs\n- 'score': 0 - 100 matching score (integer) - records with score > 50 are matches",
"### Data Statistics\nScore Distribution:\n\n!Score distribution\n\n\n\nStatistics for matching record scores:\n\n\nStatistics for non-matching record scores:",
"## Dataset Creation",
"### Curation Rationale",
"### Source Data\n\nRomanian Language Bible translations",
"#### Initial Data Collection and Normalization",
"#### Who are the source language producers?",
"### Annotations",
"#### Annotation process",
"#### Who are the annotators?",
"### Personal and Sensitive Information",
"## Considerations for Using the Data",
"### Social Impact of Dataset",
"### Discussion of Biases",
"### Other Known Limitations\n\nThe dataset is biased towards some archaic language. Additionally, it involves religious figures and phrase constructions that are rarly used in the common language.",
"## Additional Information",
"### Dataset Curators",
"### Licensing Information\n\nThis data is available and distributed under Apache-2.0 license",
"### Contributions"
] |
[
140,
14,
120,
24,
349,
6,
6,
14,
45,
33,
5,
7,
10,
10,
10,
5,
5,
9,
8,
8,
7,
8,
45,
5,
6,
19,
5
] |
[
"passage: TAGS\n#task_categories-sentence-similarity #task_ids-text-scoring #task_ids-semantic-similarity-scoring #task_ids-semantic-similarity-classification #annotations_creators-machine-generated #language_creators-found #multilinguality-monolingual #size_categories-100K<n<1M #language-Romanian #license-apache-2.0 #sentence similarity #paraphrase #romanian #nlp #doi-10.57967/hf/1712 #region-us \n# Dataset Card for \"Romanian Bible Paraphrase Corpus\"## Table of Contents\n- Dataset Description\n - Dataset Summary\n - Supported Tasks and Leaderboards\n - Languages\n- Dataset Structure\n - Data Instances\n - Data Fields\n - Data Splits\n- Dataset Creation\n - Curation Rationale\n - Source Data\n - Annotations\n - Personal and Sensitive Information\n- Considerations for Using the Data\n - Social Impact of Dataset\n - Discussion of Biases\n - Other Known Limitations\n- Additional Information\n - Dataset Curators\n - Licensing Information\n - Citation Information\n - Contributions## Dataset Description\n\n- Homepage: URL\n- Repository: URL\n- Point of Contact: Andrei Paraschiv\n-"
] |
135683c1090f9b5a036d573d6416a793fbb6c21b
|
# Dataset Card for "Synthetic_Runyankole_VITS_22.5k"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
mekaneeky/Synthetic_Runyankole_VITS_22.5k
|
[
"region:us"
] |
2023-10-24T15:19:12+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "dev", "path": "data/dev-*"}, {"split": "test", "path": "data/test-*"}]}], "dataset_info": {"features": [{"name": "eng", "dtype": "string"}, {"name": "lug", "dtype": "string"}, {"name": "ach", "dtype": "string"}, {"name": "teo", "dtype": "string"}, {"name": "lgg", "dtype": "string"}, {"name": "nyn", "dtype": "string"}, {"name": "ID", "dtype": "string"}, {"name": "runyankole_synthetic_audio", "sequence": {"sequence": "float32"}}], "splits": [{"name": "train", "num_bytes": 14611930432, "num_examples": 23947}, {"name": "dev", "num_bytes": 304694860, "num_examples": 500}, {"name": "test", "num_bytes": 324234504, "num_examples": 500}], "download_size": 15255437028, "dataset_size": 15240859796}}
|
2023-10-24T15:34:22+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "Synthetic_Runyankole_VITS_22.5k"
More Information needed
|
[
"# Dataset Card for \"Synthetic_Runyankole_VITS_22.5k\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"Synthetic_Runyankole_VITS_22.5k\"\n\nMore Information needed"
] |
[
6,
26
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"Synthetic_Runyankole_VITS_22.5k\"\n\nMore Information needed"
] |
4119abd116793aa36e0705036d629ef832337113
|
# Dataset Card for "nietzsche"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
krasaee/nietzsche
|
[
"region:us"
] |
2023-10-24T15:54:51+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}], "dataset_info": {"features": [{"name": "text", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 9929433, "num_examples": 60480}], "download_size": 6288420, "dataset_size": 9929433}}
|
2023-10-26T06:47:27+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "nietzsche"
More Information needed
|
[
"# Dataset Card for \"nietzsche\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"nietzsche\"\n\nMore Information needed"
] |
[
6,
13
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"nietzsche\"\n\nMore Information needed"
] |
8ec6b2c337f95c758d9190d51f56ee911d39347d
|
# Dataset Card for Extended CodeSearchNet corpus
|
andstor/code_search_net_extended
|
[
"annotations_creators:no-annotation",
"language_creators:machine-generated",
"multilinguality:multilingual",
"size_categories:100K<n<1M",
"size_categories:10K<n<100K",
"size_categories:1M<n<10M",
"source_datasets:original",
"language:code",
"license:other",
"region:us"
] |
2023-10-24T16:20:13+00:00
|
{"annotations_creators": ["no-annotation"], "language_creators": ["machine-generated"], "language": ["code"], "license": ["other"], "multilinguality": ["multilingual"], "size_categories": ["100K<n<1M", "10K<n<100K", "1M<n<10M"], "source_datasets": ["original"]}
|
2023-10-24T16:26:32+00:00
|
[] |
[
"code"
] |
TAGS
#annotations_creators-no-annotation #language_creators-machine-generated #multilinguality-multilingual #size_categories-100K<n<1M #size_categories-10K<n<100K #size_categories-1M<n<10M #source_datasets-original #language-code #license-other #region-us
|
# Dataset Card for Extended CodeSearchNet corpus
|
[
"# Dataset Card for Extended CodeSearchNet corpus"
] |
[
"TAGS\n#annotations_creators-no-annotation #language_creators-machine-generated #multilinguality-multilingual #size_categories-100K<n<1M #size_categories-10K<n<100K #size_categories-1M<n<10M #source_datasets-original #language-code #license-other #region-us \n",
"# Dataset Card for Extended CodeSearchNet corpus"
] |
[
91,
12
] |
[
"passage: TAGS\n#annotations_creators-no-annotation #language_creators-machine-generated #multilinguality-multilingual #size_categories-100K<n<1M #size_categories-10K<n<100K #size_categories-1M<n<10M #source_datasets-original #language-code #license-other #region-us \n# Dataset Card for Extended CodeSearchNet corpus"
] |
72d6a549c6f362562450552e85595312e0d490ac
|
# Bangumi Image Base of K-on!
This is the image base of bangumi K-ON!, we detected 51 characters, 8731 images in total. The full dataset is [here](all.zip).
**Please note that these image bases are not guaranteed to be 100% cleaned, they may be noisy actual.** If you intend to manually train models using this dataset, we recommend performing necessary preprocessing on the downloaded dataset to eliminate potential noisy samples (approximately 1% probability).
Here is the characters' preview:
| # | Images | Download | Preview 1 | Preview 2 | Preview 3 | Preview 4 | Preview 5 | Preview 6 | Preview 7 | Preview 8 |
|:------|---------:|:---------------------------|:-------------------------------|:-------------------------------|:-------------------------------|:-------------------------------|:-------------------------------|:-------------------------------|:-------------------------------|:-------------------------------|
| 0 | 1392 | [Download](0/dataset.zip) |  |  |  |  |  |  |  |  |
| 1 | 1467 | [Download](1/dataset.zip) |  |  |  |  |  |  |  |  |
| 2 | 248 | [Download](2/dataset.zip) |  |  |  |  |  |  |  |  |
| 3 | 48 | [Download](3/dataset.zip) |  |  |  |  |  |  |  |  |
| 4 | 326 | [Download](4/dataset.zip) |  |  |  |  |  |  |  |  |
| 5 | 38 | [Download](5/dataset.zip) |  |  |  |  |  |  |  |  |
| 6 | 934 | [Download](6/dataset.zip) |  |  |  |  |  |  |  |  |
| 7 | 147 | [Download](7/dataset.zip) |  |  |  |  |  |  |  |  |
| 8 | 130 | [Download](8/dataset.zip) |  |  |  |  |  |  |  |  |
| 9 | 117 | [Download](9/dataset.zip) |  |  |  |  |  |  |  |  |
| 10 | 38 | [Download](10/dataset.zip) |  |  |  |  |  |  |  |  |
| 11 | 19 | [Download](11/dataset.zip) |  |  |  |  |  |  |  |  |
| 12 | 26 | [Download](12/dataset.zip) |  |  |  |  |  |  |  |  |
| 13 | 1185 | [Download](13/dataset.zip) |  |  |  |  |  |  |  |  |
| 14 | 35 | [Download](14/dataset.zip) |  |  |  |  |  |  |  |  |
| 15 | 241 | [Download](15/dataset.zip) |  |  |  |  |  |  |  |  |
| 16 | 45 | [Download](16/dataset.zip) |  |  |  |  |  |  |  |  |
| 17 | 90 | [Download](17/dataset.zip) |  |  |  |  |  |  |  |  |
| 18 | 119 | [Download](18/dataset.zip) |  |  |  |  |  |  |  |  |
| 19 | 11 | [Download](19/dataset.zip) |  |  |  |  |  |  |  |  |
| 20 | 15 | [Download](20/dataset.zip) |  |  |  |  |  |  |  |  |
| 21 | 208 | [Download](21/dataset.zip) |  |  |  |  |  |  |  |  |
| 22 | 30 | [Download](22/dataset.zip) |  |  |  |  |  |  |  |  |
| 23 | 28 | [Download](23/dataset.zip) |  |  |  |  |  |  |  |  |
| 24 | 26 | [Download](24/dataset.zip) |  |  |  |  |  |  |  |  |
| 25 | 949 | [Download](25/dataset.zip) |  |  |  |  |  |  |  |  |
| 26 | 23 | [Download](26/dataset.zip) |  |  |  |  |  |  |  |  |
| 27 | 51 | [Download](27/dataset.zip) |  |  |  |  |  |  |  |  |
| 28 | 37 | [Download](28/dataset.zip) |  |  |  |  |  |  |  |  |
| 29 | 25 | [Download](29/dataset.zip) |  |  |  |  |  |  |  |  |
| 30 | 20 | [Download](30/dataset.zip) |  |  |  |  |  |  |  |  |
| 31 | 37 | [Download](31/dataset.zip) |  |  |  |  |  |  |  |  |
| 32 | 19 | [Download](32/dataset.zip) |  |  |  |  |  |  |  |  |
| 33 | 26 | [Download](33/dataset.zip) |  |  |  |  |  |  |  |  |
| 34 | 11 | [Download](34/dataset.zip) |  |  |  |  |  |  |  |  |
| 35 | 17 | [Download](35/dataset.zip) |  |  |  |  |  |  |  |  |
| 36 | 16 | [Download](36/dataset.zip) |  |  |  |  |  |  |  |  |
| 37 | 13 | [Download](37/dataset.zip) |  |  |  |  |  |  |  |  |
| 38 | 29 | [Download](38/dataset.zip) |  |  |  |  |  |  |  |  |
| 39 | 27 | [Download](39/dataset.zip) |  |  |  |  |  |  |  |  |
| 40 | 22 | [Download](40/dataset.zip) |  |  |  |  |  |  |  |  |
| 41 | 13 | [Download](41/dataset.zip) |  |  |  |  |  |  |  |  |
| 42 | 29 | [Download](42/dataset.zip) |  |  |  |  |  |  |  |  |
| 43 | 15 | [Download](43/dataset.zip) |  |  |  |  |  |  |  |  |
| 44 | 7 | [Download](44/dataset.zip) |  |  |  |  |  |  |  | N/A |
| 45 | 10 | [Download](45/dataset.zip) |  |  |  |  |  |  |  |  |
| 46 | 11 | [Download](46/dataset.zip) |  |  |  |  |  |  |  |  |
| 47 | 11 | [Download](47/dataset.zip) |  |  |  |  |  |  |  |  |
| 48 | 6 | [Download](48/dataset.zip) |  |  |  |  |  |  | N/A | N/A |
| 49 | 13 | [Download](49/dataset.zip) |  |  |  |  |  |  |  |  |
| noise | 331 | [Download](-1/dataset.zip) |  |  |  |  |  |  |  |  |
|
BangumiBase/kon
|
[
"size_categories:1K<n<10K",
"license:mit",
"art",
"region:us"
] |
2023-10-24T16:29:17+00:00
|
{"license": "mit", "size_categories": ["1K<n<10K"], "tags": ["art"]}
|
2023-10-25T09:02:33+00:00
|
[] |
[] |
TAGS
#size_categories-1K<n<10K #license-mit #art #region-us
|
Bangumi Image Base of K-on!
===========================
This is the image base of bangumi K-ON!, we detected 51 characters, 8731 images in total. The full dataset is here.
Please note that these image bases are not guaranteed to be 100% cleaned, they may be noisy actual. If you intend to manually train models using this dataset, we recommend performing necessary preprocessing on the downloaded dataset to eliminate potential noisy samples (approximately 1% probability).
Here is the characters' preview:
|
[] |
[
"TAGS\n#size_categories-1K<n<10K #license-mit #art #region-us \n"
] |
[
25
] |
[
"passage: TAGS\n#size_categories-1K<n<10K #license-mit #art #region-us \n"
] |
802fa25263f36a3f758b53a440c5961769d7aa9e
|
# CartPole-v1 - Imitation Learning Datasets
This is a dataset created by [Imitation Learning Datasets](https://github.com/NathanGavenski/IL-Datasets) project.
It was created by using Stable Baselines weights from a PPO policy from [HuggingFace](https://huggingface.co/sb3/ppo-CartPole-v1).
## Description
The dataset consists of 1,000 episodes with an average episodic reward of 500.
Each entry consists of:
```
obs (list): observation with length 4.
action (int): action (0 or 1).
reward (float): reward point for that timestep.
episode_returns (bool): if that state was the initial timestep for an episode.
```
## Usage
Feel free to download and use the `teacher.jsonl` dataset as you please.
If you are interested in using our PyTorch Dataset implementation, feel free to check the [IL Datasets](https://github.com/NathanGavenski/IL-Datasets/blob/main/src/imitation_datasets/dataset/dataset.py) project.
There, we implement a base Dataset that downloads this dataset and all other datasets directly from HuggingFace.
The Baseline Dataset also allows for more control over train and test splits and how many episodes you want to use (in cases where the 1k episodes are not necessary).
## Citation
Coming soon.
|
NathanGavenski/CartPole-v1
|
[
"size_categories:10M<n<100M",
"license:mit",
"Imitation Learning",
"Expert Trajectory",
"region:us"
] |
2023-10-24T16:30:02+00:00
|
{"license": "mit", "size_categories": ["10M<n<100M"], "pretty_name": "CartPole-v1 Expert Dataset", "tags": ["Imitation Learning", "Expert Trajectory"]}
|
2023-11-01T18:24:38+00:00
|
[] |
[] |
TAGS
#size_categories-10M<n<100M #license-mit #Imitation Learning #Expert Trajectory #region-us
|
# CartPole-v1 - Imitation Learning Datasets
This is a dataset created by Imitation Learning Datasets project.
It was created by using Stable Baselines weights from a PPO policy from HuggingFace.
## Description
The dataset consists of 1,000 episodes with an average episodic reward of 500.
Each entry consists of:
## Usage
Feel free to download and use the 'URL' dataset as you please.
If you are interested in using our PyTorch Dataset implementation, feel free to check the IL Datasets project.
There, we implement a base Dataset that downloads this dataset and all other datasets directly from HuggingFace.
The Baseline Dataset also allows for more control over train and test splits and how many episodes you want to use (in cases where the 1k episodes are not necessary).
Coming soon.
|
[
"# CartPole-v1 - Imitation Learning Datasets\n\nThis is a dataset created by Imitation Learning Datasets project. \nIt was created by using Stable Baselines weights from a PPO policy from HuggingFace.",
"## Description\n\nThe dataset consists of 1,000 episodes with an average episodic reward of 500.\nEach entry consists of:",
"## Usage\n\nFeel free to download and use the 'URL' dataset as you please.\nIf you are interested in using our PyTorch Dataset implementation, feel free to check the IL Datasets project.\nThere, we implement a base Dataset that downloads this dataset and all other datasets directly from HuggingFace.\nThe Baseline Dataset also allows for more control over train and test splits and how many episodes you want to use (in cases where the 1k episodes are not necessary).\n\nComing soon."
] |
[
"TAGS\n#size_categories-10M<n<100M #license-mit #Imitation Learning #Expert Trajectory #region-us \n",
"# CartPole-v1 - Imitation Learning Datasets\n\nThis is a dataset created by Imitation Learning Datasets project. \nIt was created by using Stable Baselines weights from a PPO policy from HuggingFace.",
"## Description\n\nThe dataset consists of 1,000 episodes with an average episodic reward of 500.\nEach entry consists of:",
"## Usage\n\nFeel free to download and use the 'URL' dataset as you please.\nIf you are interested in using our PyTorch Dataset implementation, feel free to check the IL Datasets project.\nThere, we implement a base Dataset that downloads this dataset and all other datasets directly from HuggingFace.\nThe Baseline Dataset also allows for more control over train and test splits and how many episodes you want to use (in cases where the 1k episodes are not necessary).\n\nComing soon."
] |
[
33,
51,
26,
113
] |
[
"passage: TAGS\n#size_categories-10M<n<100M #license-mit #Imitation Learning #Expert Trajectory #region-us \n# CartPole-v1 - Imitation Learning Datasets\n\nThis is a dataset created by Imitation Learning Datasets project. \nIt was created by using Stable Baselines weights from a PPO policy from HuggingFace.## Description\n\nThe dataset consists of 1,000 episodes with an average episodic reward of 500.\nEach entry consists of:## Usage\n\nFeel free to download and use the 'URL' dataset as you please.\nIf you are interested in using our PyTorch Dataset implementation, feel free to check the IL Datasets project.\nThere, we implement a base Dataset that downloads this dataset and all other datasets directly from HuggingFace.\nThe Baseline Dataset also allows for more control over train and test splits and how many episodes you want to use (in cases where the 1k episodes are not necessary).\n\nComing soon."
] |
80716da91358191f51c32fc0b461a7492368cb30
|
# Dataset Card for "combined_toxicity_profanity_v2_eval_only"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
Intuit-GenSRF/combined_toxicity_profanity_v2_eval_only
|
[
"region:us"
] |
2023-10-24T16:44:15+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}], "dataset_info": {"features": [{"name": "text", "dtype": "string"}, {"name": "labels", "sequence": "string"}, {"name": "encoded_labels", "sequence": "int64"}], "splits": [{"name": "train", "num_bytes": 313551093, "num_examples": 710497}], "download_size": 161511342, "dataset_size": 313551093}}
|
2023-10-24T16:44:29+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "combined_toxicity_profanity_v2_eval_only"
More Information needed
|
[
"# Dataset Card for \"combined_toxicity_profanity_v2_eval_only\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"combined_toxicity_profanity_v2_eval_only\"\n\nMore Information needed"
] |
[
6,
29
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"combined_toxicity_profanity_v2_eval_only\"\n\nMore Information needed"
] |
03abc84364009226820e2433f33ef5e1eb963f4a
|
# Dataset Card for "musdb_mp3"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
danjacobellis/musdb_mp3
|
[
"region:us"
] |
2023-10-24T16:56:21+00:00
|
{"dataset_info": {"features": [{"name": "name", "dtype": "string"}, {"name": "mixture", "dtype": {"audio": {"sampling_rate": 44100, "mono": false, "decode": false}}}, {"name": "bps", "dtype": "float32"}], "splits": [{"name": "test", "num_bytes": 1437.0, "num_examples": 15}], "download_size": 3180, "dataset_size": 1437.0}, "configs": [{"config_name": "default", "data_files": [{"split": "test", "path": "data/test-*"}]}]}
|
2023-10-24T16:56:22+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "musdb_mp3"
More Information needed
|
[
"# Dataset Card for \"musdb_mp3\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"musdb_mp3\"\n\nMore Information needed"
] |
[
6,
15
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"musdb_mp3\"\n\nMore Information needed"
] |
b43d71d525ee7777fe9b45e32af7a2c3571f44e7
|
# Dataset Card for "emailchaser-llm-subject-data-v0.0.3"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
gmaijoe-emailchaser/emailchaser-llm-subject-data-v0.0.1
|
[
"region:us"
] |
2023-10-24T17:03:40+00:00
|
{"dataset_info": {"features": [{"name": "text", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 512963, "num_examples": 713}], "download_size": 70894, "dataset_size": 512963}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T17:03:42+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "emailchaser-llm-subject-data-v0.0.3"
More Information needed
|
[
"# Dataset Card for \"emailchaser-llm-subject-data-v0.0.3\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"emailchaser-llm-subject-data-v0.0.3\"\n\nMore Information needed"
] |
[
6,
26
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"emailchaser-llm-subject-data-v0.0.3\"\n\nMore Information needed"
] |
b98da75e763c364e976864442c96a94ebe82be63
|
# Schema-Guided Dialogue dataset - Dialogue State Tracking
This dataset contains the Schema-Guided Dialogue Dataset, formatted according to the prompt formats from the following two dialogue state tracking papers:
- Description-Driven Dialogue State Tracking (D3ST) ([Zhao et al., 2022](https://arxiv.org/abs/2201.08904))
- Show, Don't Tell (SDT) ([Gupta et al., 2022](https://arxiv.org/abs/2204.04327))
Data processing code: https://github.com/google-research/task-oriented-dialogue
Original dataset: https://github.com/google-research-datasets/dstc8-schema-guided-dialogue
|
shermansiu/sgd_dst
|
[
"language:en",
"license:cc-by-sa-4.0",
"natural-language-processing",
"dialogue-state-tracking",
"arxiv:2201.08904",
"arxiv:2204.04327",
"region:us"
] |
2023-10-24T17:04:56+00:00
|
{"language": ["en"], "license": "cc-by-sa-4.0", "pretty_name": "Schema-Guided Dialogue dataset - Dialogue State Tracking", "tags": ["natural-language-processing", "dialogue-state-tracking"], "configs": [{"config_name": "sgd-sdt", "data_files": [{"split": "train", "path": "sdt/sgd_sdt_v0_train.tsv"}, {"split": "dev", "path": "sdt/sgd_sdt_v0_dev.tsv"}, {"split": "test", "path": "sdt/sgd_sdt_v0_test.tsv"}]}, {"config_name": "sgd-d3st", "data_files": [{"split": "train", "path": "d3st/sgd_d3st_v0_train.tsv"}, {"split": "dev", "path": "d3st/sgd_d3st_v0_dev.tsv"}, {"split": "test", "path": "d3st/sgd_d3st_v0_test.tsv"}]}], "dataset_info": [{"config_name": "sgd-sdt", "features": [{"name": "prompt", "dtype": "string"}, {"name": "target", "dtype": "string"}, {"name": "dialogue_id", "dtype": "string"}, {"name": "turn_id", "dtype": "string"}, {"name": "frame_id", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 269690581, "num_examples": 175780}, {"name": "dev", "num_bytes": 38376392, "num_examples": 26077}, {"name": "test", "num_bytes": 72134551, "num_examples": 46116}]}, {"config_name": "sgd-d3st", "features": [{"name": "prompt", "dtype": "string"}, {"name": "target", "dtype": "string"}, {"name": "dialogue_id", "dtype": "string"}, {"name": "turn_id", "dtype": "string"}, {"name": "frame_id", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 226090375, "num_examples": 175780}, {"name": "dev", "num_bytes": 32170435, "num_examples": 26077}, {"name": "test", "num_bytes": 59997238, "num_examples": 46116}]}]}
|
2023-12-27T07:42:20+00:00
|
[
"2201.08904",
"2204.04327"
] |
[
"en"
] |
TAGS
#language-English #license-cc-by-sa-4.0 #natural-language-processing #dialogue-state-tracking #arxiv-2201.08904 #arxiv-2204.04327 #region-us
|
# Schema-Guided Dialogue dataset - Dialogue State Tracking
This dataset contains the Schema-Guided Dialogue Dataset, formatted according to the prompt formats from the following two dialogue state tracking papers:
- Description-Driven Dialogue State Tracking (D3ST) (Zhao et al., 2022)
- Show, Don't Tell (SDT) (Gupta et al., 2022)
Data processing code: URL
Original dataset: URL
|
[
"# Schema-Guided Dialogue dataset - Dialogue State Tracking\n\nThis dataset contains the Schema-Guided Dialogue Dataset, formatted according to the prompt formats from the following two dialogue state tracking papers:\n- Description-Driven Dialogue State Tracking (D3ST) (Zhao et al., 2022)\n- Show, Don't Tell (SDT) (Gupta et al., 2022)\n\nData processing code: URL \nOriginal dataset: URL"
] |
[
"TAGS\n#language-English #license-cc-by-sa-4.0 #natural-language-processing #dialogue-state-tracking #arxiv-2201.08904 #arxiv-2204.04327 #region-us \n",
"# Schema-Guided Dialogue dataset - Dialogue State Tracking\n\nThis dataset contains the Schema-Guided Dialogue Dataset, formatted according to the prompt formats from the following two dialogue state tracking papers:\n- Description-Driven Dialogue State Tracking (D3ST) (Zhao et al., 2022)\n- Show, Don't Tell (SDT) (Gupta et al., 2022)\n\nData processing code: URL \nOriginal dataset: URL"
] |
[
55,
109
] |
[
"passage: TAGS\n#language-English #license-cc-by-sa-4.0 #natural-language-processing #dialogue-state-tracking #arxiv-2201.08904 #arxiv-2204.04327 #region-us \n# Schema-Guided Dialogue dataset - Dialogue State Tracking\n\nThis dataset contains the Schema-Guided Dialogue Dataset, formatted according to the prompt formats from the following two dialogue state tracking papers:\n- Description-Driven Dialogue State Tracking (D3ST) (Zhao et al., 2022)\n- Show, Don't Tell (SDT) (Gupta et al., 2022)\n\nData processing code: URL \nOriginal dataset: URL"
] |
043f27a39fa77b2d29360c06e82b8aa3431cc6a6
|
# Dataset Card for "Title_CC"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
dumyy/Title_CC
|
[
"region:us"
] |
2023-10-24T17:39:47+00:00
|
{"configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}, {"split": "test", "path": "data/test-*"}]}], "dataset_info": {"features": [{"name": "tokens", "sequence": "string"}, {"name": "ner_tags", "sequence": "int64"}], "splits": [{"name": "train", "num_bytes": 6364154, "num_examples": 1846}, {"name": "test", "num_bytes": 6381296, "num_examples": 1500}], "download_size": 1320245, "dataset_size": 12745450}}
|
2023-10-24T17:39:55+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "Title_CC"
More Information needed
|
[
"# Dataset Card for \"Title_CC\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"Title_CC\"\n\nMore Information needed"
] |
[
6,
14
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"Title_CC\"\n\nMore Information needed"
] |
d621e3b51eb797eb9c3fefed5048906c88d56182
|
# Navigating News Narratives: A Media Bias Analysis Dataset
## Introduction
The growing prevalence of bias in news media critically influences public perceptions on various imperative subjects, encompassing areas like political standpoints, health, resource allocation, racial perspectives, age, gender biases, and climate change. In light of this, it's crucial for media outlets to uphold the ethics of providing precise information and heightening public cognizance about these pivotal issues. The necessity for a mechanism that could counteract the spread of fallacious or misleading details is thus underscored, aiming to rejuvenate public confidence in the media.
## Dataset Description
This dataset encompasses multiple dimensions of biases in news media, such as political inclinations, hate speech, toxicity, sexism, ageism, and more, establishing its distinctiveness in the realm of similar datasets. It's noteworthy that the dataset explicitly refrains from including any personally identifiable information (PII).
### Format
The data structure is tabulated as follows:
- **Text**: The main content.
- **Dimension**: Descriptive category of the text.
- **Biased_Words**: A compilation of words regarded as biased.
- **Aspect**: Specific sub-topic within the main content.
- **Label**: Indicates the presence (True) or absence (False) of bias. The label is ternary - highly biased, slightly biased and neutral
- **Toxicity**: Indicates the presence (True) or absence (False) of bias.
- **Identity_mention**: Mention of any identity based on words match.
### Annotation Scheme
The labels and annotations in the dataset are generated through a system of Active Learning, cycling through:
- Manual Labeling
- Semi-Supervised Learning
- Human Verification
The scheme comprises:
- **Bias Label**: Specifies the degree of bias (e.g., no bias, mild, or strong).
- **Words/Phrases Level Biases**: Pinpoints specific biased terms or phrases.
- **Subjective Bias (Aspect)**: Highlights biases pertinent to content dimensions.
*Due to the nuances of semantic match algorithms, certain labels such as 'identity' and 'aspect' may appear distinctively different.*
## Datasets Utilized
For a comprehensive perspective, various news categories like climate crisis summaries, occupational, spiritual/faith, etc., have been curated using RSS. Active learning has been harnessed to classify sentences based on their neutrality or degree of bias and identify biased terms.
Moreover, this compilation incorporates data from several reputable sources:
- **MBIC (media bias)**: [Source](https://zenodo.org/records/4474336)
- **Hyperpartisan news**: [Source](https://huggingface.co/datasets/hyperpartisan_news_detection)
- **Toxic comment classification**: [Kaggle Source](https://kaggle.com/competitions/jigsaw-toxic-comment-classification-challenge)
- **Jigsaw Unintended Bias**: [Kaggle Source](https://kaggle.com/competitions/jigsaw-unintended-bias-in-toxicity-classification)
- **Age Bias**: [Harvard Source](https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/L4OAKN)
- **Multi-dimensional news (Ukraine)**: [Source](https://zenodo.org/records/3885351#.ZF0KoxHMLtV)
- **Social biases**: [Source](https://maartensap.com/social-bias-frames/)
## Objective
This dataset aims to provide unrestricted, complimentary access, striving to facilitate its adoption among global AI researchers and practitioners. Emphasis is placed on ensuring easy accessibility and usage of the dataset.
https://arxiv.org/abs/2312.00168
## Licensing & Citation
If you leverage this dataset in your work, we kindly request you to cite us:
> Navigating News Narratives: A Media Bias Analysis Dataset © 2023 by Shaina Raza, Vector Institute is licensed under [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/).
For any other datasets utilized in this project, please ensure appropriate attribution as detailed in the aforementioned list.
[Hugging Face Source](https://huggingface.co/datasets/newsmediabias/news-bias-full-data/edit/main/README.md) - Accessed on 10/24/2023.
## Citation
```bibtex
@misc{Raza2023MediaBiasDataset,
title={Navigating News Narratives: A Media Bias Analysis Dataset},
author={Shaina Raza},
year={2023},
note={Licensed under CC BY-NC 4.0},
url={https://creativecommons.org/licenses/by-nc/4.0/}
}
```
```bibtex
@article{raza2023navigating,
title={Navigating News Narratives: A Media Bias Analysis Dataset},
author={Raza, Shaina},
journal={arXiv preprint arXiv:2312.00168},
year={2023}
}
```
|
newsmediabias/news-bias-full-data
|
[
"task_categories:text-classification",
"task_categories:token-classification",
"language:en",
"license:cc-by-nc-4.0",
"arxiv:2312.00168",
"region:us"
] |
2023-10-24T17:51:24+00:00
|
{"language": ["en"], "license": "cc-by-nc-4.0", "task_categories": ["text-classification", "token-classification"], "pretty_name": "Navigating News Narratives"}
|
2024-01-08T18:35:47+00:00
|
[
"2312.00168"
] |
[
"en"
] |
TAGS
#task_categories-text-classification #task_categories-token-classification #language-English #license-cc-by-nc-4.0 #arxiv-2312.00168 #region-us
|
# Navigating News Narratives: A Media Bias Analysis Dataset
## Introduction
The growing prevalence of bias in news media critically influences public perceptions on various imperative subjects, encompassing areas like political standpoints, health, resource allocation, racial perspectives, age, gender biases, and climate change. In light of this, it's crucial for media outlets to uphold the ethics of providing precise information and heightening public cognizance about these pivotal issues. The necessity for a mechanism that could counteract the spread of fallacious or misleading details is thus underscored, aiming to rejuvenate public confidence in the media.
## Dataset Description
This dataset encompasses multiple dimensions of biases in news media, such as political inclinations, hate speech, toxicity, sexism, ageism, and more, establishing its distinctiveness in the realm of similar datasets. It's noteworthy that the dataset explicitly refrains from including any personally identifiable information (PII).
### Format
The data structure is tabulated as follows:
- Text: The main content.
- Dimension: Descriptive category of the text.
- Biased_Words: A compilation of words regarded as biased.
- Aspect: Specific sub-topic within the main content.
- Label: Indicates the presence (True) or absence (False) of bias. The label is ternary - highly biased, slightly biased and neutral
- Toxicity: Indicates the presence (True) or absence (False) of bias.
- Identity_mention: Mention of any identity based on words match.
### Annotation Scheme
The labels and annotations in the dataset are generated through a system of Active Learning, cycling through:
- Manual Labeling
- Semi-Supervised Learning
- Human Verification
The scheme comprises:
- Bias Label: Specifies the degree of bias (e.g., no bias, mild, or strong).
- Words/Phrases Level Biases: Pinpoints specific biased terms or phrases.
- Subjective Bias (Aspect): Highlights biases pertinent to content dimensions.
*Due to the nuances of semantic match algorithms, certain labels such as 'identity' and 'aspect' may appear distinctively different.*
## Datasets Utilized
For a comprehensive perspective, various news categories like climate crisis summaries, occupational, spiritual/faith, etc., have been curated using RSS. Active learning has been harnessed to classify sentences based on their neutrality or degree of bias and identify biased terms.
Moreover, this compilation incorporates data from several reputable sources:
- MBIC (media bias): Source
- Hyperpartisan news: Source
- Toxic comment classification: Kaggle Source
- Jigsaw Unintended Bias: Kaggle Source
- Age Bias: Harvard Source
- Multi-dimensional news (Ukraine): Source
- Social biases: Source
## Objective
This dataset aims to provide unrestricted, complimentary access, striving to facilitate its adoption among global AI researchers and practitioners. Emphasis is placed on ensuring easy accessibility and usage of the dataset.
URL
## Licensing & Citation
If you leverage this dataset in your work, we kindly request you to cite us:
> Navigating News Narratives: A Media Bias Analysis Dataset © 2023 by Shaina Raza, Vector Institute is licensed under CC BY-NC 4.0.
For any other datasets utilized in this project, please ensure appropriate attribution as detailed in the aforementioned list.
Hugging Face Source - Accessed on 10/24/2023.
|
[
"# Navigating News Narratives: A Media Bias Analysis Dataset",
"## Introduction\nThe growing prevalence of bias in news media critically influences public perceptions on various imperative subjects, encompassing areas like political standpoints, health, resource allocation, racial perspectives, age, gender biases, and climate change. In light of this, it's crucial for media outlets to uphold the ethics of providing precise information and heightening public cognizance about these pivotal issues. The necessity for a mechanism that could counteract the spread of fallacious or misleading details is thus underscored, aiming to rejuvenate public confidence in the media.",
"## Dataset Description\nThis dataset encompasses multiple dimensions of biases in news media, such as political inclinations, hate speech, toxicity, sexism, ageism, and more, establishing its distinctiveness in the realm of similar datasets. It's noteworthy that the dataset explicitly refrains from including any personally identifiable information (PII).",
"### Format\nThe data structure is tabulated as follows:\n- Text: The main content.\n- Dimension: Descriptive category of the text.\n- Biased_Words: A compilation of words regarded as biased.\n- Aspect: Specific sub-topic within the main content.\n- Label: Indicates the presence (True) or absence (False) of bias. The label is ternary - highly biased, slightly biased and neutral\n- Toxicity: Indicates the presence (True) or absence (False) of bias. \n- Identity_mention: Mention of any identity based on words match.",
"### Annotation Scheme\nThe labels and annotations in the dataset are generated through a system of Active Learning, cycling through:\n- Manual Labeling \n- Semi-Supervised Learning \n- Human Verification\n\nThe scheme comprises:\n- Bias Label: Specifies the degree of bias (e.g., no bias, mild, or strong).\n- Words/Phrases Level Biases: Pinpoints specific biased terms or phrases.\n- Subjective Bias (Aspect): Highlights biases pertinent to content dimensions.\n\n*Due to the nuances of semantic match algorithms, certain labels such as 'identity' and 'aspect' may appear distinctively different.*",
"## Datasets Utilized\nFor a comprehensive perspective, various news categories like climate crisis summaries, occupational, spiritual/faith, etc., have been curated using RSS. Active learning has been harnessed to classify sentences based on their neutrality or degree of bias and identify biased terms. \n\nMoreover, this compilation incorporates data from several reputable sources:\n- MBIC (media bias): Source\n- Hyperpartisan news: Source\n- Toxic comment classification: Kaggle Source\n- Jigsaw Unintended Bias: Kaggle Source\n- Age Bias: Harvard Source\n- Multi-dimensional news (Ukraine): Source\n- Social biases: Source",
"## Objective\nThis dataset aims to provide unrestricted, complimentary access, striving to facilitate its adoption among global AI researchers and practitioners. Emphasis is placed on ensuring easy accessibility and usage of the dataset.\n\nURL",
"## Licensing & Citation\nIf you leverage this dataset in your work, we kindly request you to cite us:\n\n> Navigating News Narratives: A Media Bias Analysis Dataset © 2023 by Shaina Raza, Vector Institute is licensed under CC BY-NC 4.0.\n\nFor any other datasets utilized in this project, please ensure appropriate attribution as detailed in the aforementioned list. \n\nHugging Face Source - Accessed on 10/24/2023."
] |
[
"TAGS\n#task_categories-text-classification #task_categories-token-classification #language-English #license-cc-by-nc-4.0 #arxiv-2312.00168 #region-us \n",
"# Navigating News Narratives: A Media Bias Analysis Dataset",
"## Introduction\nThe growing prevalence of bias in news media critically influences public perceptions on various imperative subjects, encompassing areas like political standpoints, health, resource allocation, racial perspectives, age, gender biases, and climate change. In light of this, it's crucial for media outlets to uphold the ethics of providing precise information and heightening public cognizance about these pivotal issues. The necessity for a mechanism that could counteract the spread of fallacious or misleading details is thus underscored, aiming to rejuvenate public confidence in the media.",
"## Dataset Description\nThis dataset encompasses multiple dimensions of biases in news media, such as political inclinations, hate speech, toxicity, sexism, ageism, and more, establishing its distinctiveness in the realm of similar datasets. It's noteworthy that the dataset explicitly refrains from including any personally identifiable information (PII).",
"### Format\nThe data structure is tabulated as follows:\n- Text: The main content.\n- Dimension: Descriptive category of the text.\n- Biased_Words: A compilation of words regarded as biased.\n- Aspect: Specific sub-topic within the main content.\n- Label: Indicates the presence (True) or absence (False) of bias. The label is ternary - highly biased, slightly biased and neutral\n- Toxicity: Indicates the presence (True) or absence (False) of bias. \n- Identity_mention: Mention of any identity based on words match.",
"### Annotation Scheme\nThe labels and annotations in the dataset are generated through a system of Active Learning, cycling through:\n- Manual Labeling \n- Semi-Supervised Learning \n- Human Verification\n\nThe scheme comprises:\n- Bias Label: Specifies the degree of bias (e.g., no bias, mild, or strong).\n- Words/Phrases Level Biases: Pinpoints specific biased terms or phrases.\n- Subjective Bias (Aspect): Highlights biases pertinent to content dimensions.\n\n*Due to the nuances of semantic match algorithms, certain labels such as 'identity' and 'aspect' may appear distinctively different.*",
"## Datasets Utilized\nFor a comprehensive perspective, various news categories like climate crisis summaries, occupational, spiritual/faith, etc., have been curated using RSS. Active learning has been harnessed to classify sentences based on their neutrality or degree of bias and identify biased terms. \n\nMoreover, this compilation incorporates data from several reputable sources:\n- MBIC (media bias): Source\n- Hyperpartisan news: Source\n- Toxic comment classification: Kaggle Source\n- Jigsaw Unintended Bias: Kaggle Source\n- Age Bias: Harvard Source\n- Multi-dimensional news (Ukraine): Source\n- Social biases: Source",
"## Objective\nThis dataset aims to provide unrestricted, complimentary access, striving to facilitate its adoption among global AI researchers and practitioners. Emphasis is placed on ensuring easy accessibility and usage of the dataset.\n\nURL",
"## Licensing & Citation\nIf you leverage this dataset in your work, we kindly request you to cite us:\n\n> Navigating News Narratives: A Media Bias Analysis Dataset © 2023 by Shaina Raza, Vector Institute is licensed under CC BY-NC 4.0.\n\nFor any other datasets utilized in this project, please ensure appropriate attribution as detailed in the aforementioned list. \n\nHugging Face Source - Accessed on 10/24/2023."
] |
[
52,
16,
138,
86,
143,
157,
147,
58,
104
] |
[
"passage: TAGS\n#task_categories-text-classification #task_categories-token-classification #language-English #license-cc-by-nc-4.0 #arxiv-2312.00168 #region-us \n# Navigating News Narratives: A Media Bias Analysis Dataset## Introduction\nThe growing prevalence of bias in news media critically influences public perceptions on various imperative subjects, encompassing areas like political standpoints, health, resource allocation, racial perspectives, age, gender biases, and climate change. In light of this, it's crucial for media outlets to uphold the ethics of providing precise information and heightening public cognizance about these pivotal issues. The necessity for a mechanism that could counteract the spread of fallacious or misleading details is thus underscored, aiming to rejuvenate public confidence in the media.## Dataset Description\nThis dataset encompasses multiple dimensions of biases in news media, such as political inclinations, hate speech, toxicity, sexism, ageism, and more, establishing its distinctiveness in the realm of similar datasets. It's noteworthy that the dataset explicitly refrains from including any personally identifiable information (PII).### Format\nThe data structure is tabulated as follows:\n- Text: The main content.\n- Dimension: Descriptive category of the text.\n- Biased_Words: A compilation of words regarded as biased.\n- Aspect: Specific sub-topic within the main content.\n- Label: Indicates the presence (True) or absence (False) of bias. The label is ternary - highly biased, slightly biased and neutral\n- Toxicity: Indicates the presence (True) or absence (False) of bias. \n- Identity_mention: Mention of any identity based on words match."
] |
c4949d2acae93dc9795bb7f07317080b9ae882a2
|
# Dataset Card for "test_book"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
justinphan3110/test_book
|
[
"region:us"
] |
2023-10-24T18:29:14+00:00
|
{"dataset_info": {"features": [{"name": "name", "dtype": "string"}, {"name": "minhashes", "sequence": "binary"}, {"name": "behavior", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 1688460457, "num_examples": 24}], "download_size": 161058929, "dataset_size": 1688460457}}
|
2023-10-24T19:21:59+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "test_book"
More Information needed
|
[
"# Dataset Card for \"test_book\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"test_book\"\n\nMore Information needed"
] |
[
6,
13
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"test_book\"\n\nMore Information needed"
] |
5a52250ac6e628780e089a8b2b2cc260de97ff82
|
# Bangumi Image Base of Fairy Tail
This is the image base of bangumi Fairy Tail, we detected 270 characters, 33650 images in total. The full dataset is [here](all.zip).
**Please note that these image bases are not guaranteed to be 100% cleaned, they may be noisy actual.** If you intend to manually train models using this dataset, we recommend performing necessary preprocessing on the downloaded dataset to eliminate potential noisy samples (approximately 1% probability).
Here is the characters' preview:
| # | Images | Download | Preview 1 | Preview 2 | Preview 3 | Preview 4 | Preview 5 | Preview 6 | Preview 7 | Preview 8 |
|:------|---------:|:----------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|
| 0 | 1894 | [Download](0/dataset.zip) |  |  |  |  |  |  |  |  |
| 1 | 83 | [Download](1/dataset.zip) |  |  |  |  |  |  |  |  |
| 2 | 62 | [Download](2/dataset.zip) |  |  |  |  |  |  |  |  |
| 3 | 49 | [Download](3/dataset.zip) |  |  |  |  |  |  |  |  |
| 4 | 36 | [Download](4/dataset.zip) |  |  |  |  |  |  |  |  |
| 5 | 11 | [Download](5/dataset.zip) |  |  |  |  |  |  |  |  |
| 6 | 76 | [Download](6/dataset.zip) |  |  |  |  |  |  |  |  |
| 7 | 162 | [Download](7/dataset.zip) |  |  |  |  |  |  |  |  |
| 8 | 4062 | [Download](8/dataset.zip) |  |  |  |  |  |  |  |  |
| 9 | 52 | [Download](9/dataset.zip) |  |  |  |  |  |  |  |  |
| 10 | 210 | [Download](10/dataset.zip) |  |  |  |  |  |  |  |  |
| 11 | 441 | [Download](11/dataset.zip) |  |  |  |  |  |  |  |  |
| 12 | 481 | [Download](12/dataset.zip) |  |  |  |  |  |  |  |  |
| 13 | 2387 | [Download](13/dataset.zip) |  |  |  |  |  |  |  |  |
| 14 | 108 | [Download](14/dataset.zip) |  |  |  |  |  |  |  |  |
| 15 | 200 | [Download](15/dataset.zip) |  |  |  |  |  |  |  |  |
| 16 | 137 | [Download](16/dataset.zip) |  |  |  |  |  |  |  |  |
| 17 | 481 | [Download](17/dataset.zip) |  |  |  |  |  |  |  |  |
| 18 | 81 | [Download](18/dataset.zip) |  |  |  |  |  |  |  |  |
| 19 | 320 | [Download](19/dataset.zip) |  |  |  |  |  |  |  |  |
| 20 | 225 | [Download](20/dataset.zip) |  |  |  |  |  |  |  |  |
| 21 | 42 | [Download](21/dataset.zip) |  |  |  |  |  |  |  |  |
| 22 | 81 | [Download](22/dataset.zip) |  |  |  |  |  |  |  |  |
| 23 | 167 | [Download](23/dataset.zip) |  |  |  |  |  |  |  |  |
| 24 | 99 | [Download](24/dataset.zip) |  |  |  |  |  |  |  |  |
| 25 | 112 | [Download](25/dataset.zip) |  |  |  |  |  |  |  |  |
| 26 | 81 | [Download](26/dataset.zip) |  |  |  |  |  |  |  |  |
| 27 | 27 | [Download](27/dataset.zip) |  |  |  |  |  |  |  |  |
| 28 | 108 | [Download](28/dataset.zip) |  |  |  |  |  |  |  |  |
| 29 | 717 | [Download](29/dataset.zip) |  |  |  |  |  |  |  |  |
| 30 | 221 | [Download](30/dataset.zip) |  |  |  |  |  |  |  |  |
| 31 | 61 | [Download](31/dataset.zip) |  |  |  |  |  |  |  |  |
| 32 | 37 | [Download](32/dataset.zip) |  |  |  |  |  |  |  |  |
| 33 | 20 | [Download](33/dataset.zip) |  |  |  |  |  |  |  |  |
| 34 | 86 | [Download](34/dataset.zip) |  |  |  |  |  |  |  |  |
| 35 | 55 | [Download](35/dataset.zip) |  |  |  |  |  |  |  |  |
| 36 | 27 | [Download](36/dataset.zip) |  |  |  |  |  |  |  |  |
| 37 | 80 | [Download](37/dataset.zip) |  |  |  |  |  |  |  |  |
| 38 | 50 | [Download](38/dataset.zip) |  |  |  |  |  |  |  |  |
| 39 | 643 | [Download](39/dataset.zip) |  |  |  |  |  |  |  |  |
| 40 | 352 | [Download](40/dataset.zip) |  |  |  |  |  |  |  |  |
| 41 | 35 | [Download](41/dataset.zip) |  |  |  |  |  |  |  |  |
| 42 | 94 | [Download](42/dataset.zip) |  |  |  |  |  |  |  |  |
| 43 | 50 | [Download](43/dataset.zip) |  |  |  |  |  |  |  |  |
| 44 | 50 | [Download](44/dataset.zip) |  |  |  |  |  |  |  |  |
| 45 | 29 | [Download](45/dataset.zip) |  |  |  |  |  |  |  |  |
| 46 | 62 | [Download](46/dataset.zip) |  |  |  |  |  |  |  |  |
| 47 | 21 | [Download](47/dataset.zip) |  |  |  |  |  |  |  |  |
| 48 | 26 | [Download](48/dataset.zip) |  |  |  |  |  |  |  |  |
| 49 | 30 | [Download](49/dataset.zip) |  |  |  |  |  |  |  |  |
| 50 | 253 | [Download](50/dataset.zip) |  |  |  |  |  |  |  |  |
| 51 | 39 | [Download](51/dataset.zip) |  |  |  |  |  |  |  |  |
| 52 | 126 | [Download](52/dataset.zip) |  |  |  |  |  |  |  |  |
| 53 | 127 | [Download](53/dataset.zip) |  |  |  |  |  |  |  |  |
| 54 | 106 | [Download](54/dataset.zip) |  |  |  |  |  |  |  |  |
| 55 | 34 | [Download](55/dataset.zip) |  |  |  |  |  |  |  |  |
| 56 | 46 | [Download](56/dataset.zip) |  |  |  |  |  |  |  |  |
| 57 | 109 | [Download](57/dataset.zip) |  |  |  |  |  |  |  |  |
| 58 | 72 | [Download](58/dataset.zip) |  |  |  |  |  |  |  |  |
| 59 | 45 | [Download](59/dataset.zip) |  |  |  |  |  |  |  |  |
| 60 | 25 | [Download](60/dataset.zip) |  |  |  |  |  |  |  |  |
| 61 | 59 | [Download](61/dataset.zip) |  |  |  |  |  |  |  |  |
| 62 | 53 | [Download](62/dataset.zip) |  |  |  |  |  |  |  |  |
| 63 | 56 | [Download](63/dataset.zip) |  |  |  |  |  |  |  |  |
| 64 | 149 | [Download](64/dataset.zip) |  |  |  |  |  |  |  |  |
| 65 | 311 | [Download](65/dataset.zip) |  |  |  |  |  |  |  |  |
| 66 | 64 | [Download](66/dataset.zip) |  |  |  |  |  |  |  |  |
| 67 | 58 | [Download](67/dataset.zip) |  |  |  |  |  |  |  |  |
| 68 | 53 | [Download](68/dataset.zip) |  |  |  |  |  |  |  |  |
| 69 | 21 | [Download](69/dataset.zip) |  |  |  |  |  |  |  |  |
| 70 | 27 | [Download](70/dataset.zip) |  |  |  |  |  |  |  |  |
| 71 | 57 | [Download](71/dataset.zip) |  |  |  |  |  |  |  |  |
| 72 | 44 | [Download](72/dataset.zip) |  |  |  |  |  |  |  |  |
| 73 | 26 | [Download](73/dataset.zip) |  |  |  |  |  |  |  |  |
| 74 | 8 | [Download](74/dataset.zip) |  |  |  |  |  |  |  |  |
| 75 | 29 | [Download](75/dataset.zip) |  |  |  |  |  |  |  |  |
| 76 | 37 | [Download](76/dataset.zip) |  |  |  |  |  |  |  |  |
| 77 | 379 | [Download](77/dataset.zip) |  |  |  |  |  |  |  |  |
| 78 | 188 | [Download](78/dataset.zip) |  |  |  |  |  |  |  |  |
| 79 | 1209 | [Download](79/dataset.zip) |  |  |  |  |  |  |  |  |
| 80 | 413 | [Download](80/dataset.zip) |  |  |  |  |  |  |  |  |
| 81 | 14 | [Download](81/dataset.zip) |  |  |  |  |  |  |  |  |
| 82 | 19 | [Download](82/dataset.zip) |  |  |  |  |  |  |  |  |
| 83 | 34 | [Download](83/dataset.zip) |  |  |  |  |  |  |  |  |
| 84 | 56 | [Download](84/dataset.zip) |  |  |  |  |  |  |  |  |
| 85 | 20 | [Download](85/dataset.zip) |  |  |  |  |  |  |  |  |
| 86 | 27 | [Download](86/dataset.zip) |  |  |  |  |  |  |  |  |
| 87 | 16 | [Download](87/dataset.zip) |  |  |  |  |  |  |  |  |
| 88 | 28 | [Download](88/dataset.zip) |  |  |  |  |  |  |  |  |
| 89 | 29 | [Download](89/dataset.zip) |  |  |  |  |  |  |  |  |
| 90 | 52 | [Download](90/dataset.zip) |  |  |  |  |  |  |  |  |
| 91 | 30 | [Download](91/dataset.zip) |  |  |  |  |  |  |  |  |
| 92 | 29 | [Download](92/dataset.zip) |  |  |  |  |  |  |  |  |
| 93 | 21 | [Download](93/dataset.zip) |  |  |  |  |  |  |  |  |
| 94 | 36 | [Download](94/dataset.zip) |  |  |  |  |  |  |  |  |
| 95 | 81 | [Download](95/dataset.zip) |  |  |  |  |  |  |  |  |
| 96 | 30 | [Download](96/dataset.zip) |  |  |  |  |  |  |  |  |
| 97 | 85 | [Download](97/dataset.zip) |  |  |  |  |  |  |  |  |
| 98 | 34 | [Download](98/dataset.zip) |  |  |  |  |  |  |  |  |
| 99 | 80 | [Download](99/dataset.zip) |  |  |  |  |  |  |  |  |
| 100 | 96 | [Download](100/dataset.zip) |  |  |  |  |  |  |  |  |
| 101 | 27 | [Download](101/dataset.zip) |  |  |  |  |  |  |  |  |
| 102 | 57 | [Download](102/dataset.zip) |  |  |  |  |  |  |  |  |
| 103 | 33 | [Download](103/dataset.zip) |  |  |  |  |  |  |  |  |
| 104 | 38 | [Download](104/dataset.zip) |  |  |  |  |  |  |  |  |
| 105 | 401 | [Download](105/dataset.zip) |  |  |  |  |  |  |  |  |
| 106 | 35 | [Download](106/dataset.zip) |  |  |  |  |  |  |  |  |
| 107 | 27 | [Download](107/dataset.zip) |  |  |  |  |  |  |  |  |
| 108 | 42 | [Download](108/dataset.zip) |  |  |  |  |  |  |  |  |
| 109 | 27 | [Download](109/dataset.zip) |  |  |  |  |  |  |  |  |
| 110 | 24 | [Download](110/dataset.zip) |  |  |  |  |  |  |  |  |
| 111 | 51 | [Download](111/dataset.zip) |  |  |  |  |  |  |  |  |
| 112 | 33 | [Download](112/dataset.zip) |  |  |  |  |  |  |  |  |
| 113 | 15 | [Download](113/dataset.zip) |  |  |  |  |  |  |  |  |
| 114 | 43 | [Download](114/dataset.zip) |  |  |  |  |  |  |  |  |
| 115 | 64 | [Download](115/dataset.zip) |  |  |  |  |  |  |  |  |
| 116 | 171 | [Download](116/dataset.zip) |  |  |  |  |  |  |  |  |
| 117 | 358 | [Download](117/dataset.zip) |  |  |  |  |  |  |  |  |
| 118 | 99 | [Download](118/dataset.zip) |  |  |  |  |  |  |  |  |
| 119 | 15 | [Download](119/dataset.zip) |  |  |  |  |  |  |  |  |
| 120 | 29 | [Download](120/dataset.zip) |  |  |  |  |  |  |  |  |
| 121 | 33 | [Download](121/dataset.zip) |  |  |  |  |  |  |  |  |
| 122 | 18 | [Download](122/dataset.zip) |  |  |  |  |  |  |  |  |
| 123 | 31 | [Download](123/dataset.zip) |  |  |  |  |  |  |  |  |
| 124 | 16 | [Download](124/dataset.zip) |  |  |  |  |  |  |  |  |
| 125 | 30 | [Download](125/dataset.zip) |  |  |  |  |  |  |  |  |
| 126 | 68 | [Download](126/dataset.zip) |  |  |  |  |  |  |  |  |
| 127 | 75 | [Download](127/dataset.zip) |  |  |  |  |  |  |  |  |
| 128 | 107 | [Download](128/dataset.zip) |  |  |  |  |  |  |  |  |
| 129 | 27 | [Download](129/dataset.zip) |  |  |  |  |  |  |  |  |
| 130 | 39 | [Download](130/dataset.zip) |  |  |  |  |  |  |  |  |
| 131 | 16 | [Download](131/dataset.zip) |  |  |  |  |  |  |  |  |
| 132 | 29 | [Download](132/dataset.zip) |  |  |  |  |  |  |  |  |
| 133 | 527 | [Download](133/dataset.zip) |  |  |  |  |  |  |  |  |
| 134 | 27 | [Download](134/dataset.zip) |  |  |  |  |  |  |  |  |
| 135 | 42 | [Download](135/dataset.zip) |  |  |  |  |  |  |  |  |
| 136 | 57 | [Download](136/dataset.zip) |  |  |  |  |  |  |  |  |
| 137 | 37 | [Download](137/dataset.zip) |  |  |  |  |  |  |  |  |
| 138 | 66 | [Download](138/dataset.zip) |  |  |  |  |  |  |  |  |
| 139 | 2383 | [Download](139/dataset.zip) |  |  |  |  |  |  |  |  |
| 140 | 192 | [Download](140/dataset.zip) |  |  |  |  |  |  |  |  |
| 141 | 826 | [Download](141/dataset.zip) |  |  |  |  |  |  |  |  |
| 142 | 104 | [Download](142/dataset.zip) |  |  |  |  |  |  |  |  |
| 143 | 108 | [Download](143/dataset.zip) |  |  |  |  |  |  |  |  |
| 144 | 67 | [Download](144/dataset.zip) |  |  |  |  |  |  |  |  |
| 145 | 111 | [Download](145/dataset.zip) |  |  |  |  |  |  |  |  |
| 146 | 140 | [Download](146/dataset.zip) |  |  |  |  |  |  |  |  |
| 147 | 59 | [Download](147/dataset.zip) |  |  |  |  |  |  |  |  |
| 148 | 20 | [Download](148/dataset.zip) |  |  |  |  |  |  |  |  |
| 149 | 33 | [Download](149/dataset.zip) |  |  |  |  |  |  |  |  |
| 150 | 38 | [Download](150/dataset.zip) |  |  |  |  |  |  |  |  |
| 151 | 15 | [Download](151/dataset.zip) |  |  |  |  |  |  |  |  |
| 152 | 231 | [Download](152/dataset.zip) |  |  |  |  |  |  |  |  |
| 153 | 20 | [Download](153/dataset.zip) |  |  |  |  |  |  |  |  |
| 154 | 16 | [Download](154/dataset.zip) |  |  |  |  |  |  |  |  |
| 155 | 23 | [Download](155/dataset.zip) |  |  |  |  |  |  |  |  |
| 156 | 16 | [Download](156/dataset.zip) |  |  |  |  |  |  |  |  |
| 157 | 52 | [Download](157/dataset.zip) |  |  |  |  |  |  |  |  |
| 158 | 83 | [Download](158/dataset.zip) |  |  |  |  |  |  |  |  |
| 159 | 41 | [Download](159/dataset.zip) |  |  |  |  |  |  |  |  |
| 160 | 36 | [Download](160/dataset.zip) |  |  |  |  |  |  |  |  |
| 161 | 12 | [Download](161/dataset.zip) |  |  |  |  |  |  |  |  |
| 162 | 108 | [Download](162/dataset.zip) |  |  |  |  |  |  |  |  |
| 163 | 39 | [Download](163/dataset.zip) |  |  |  |  |  |  |  |  |
| 164 | 76 | [Download](164/dataset.zip) |  |  |  |  |  |  |  |  |
| 165 | 19 | [Download](165/dataset.zip) |  |  |  |  |  |  |  |  |
| 166 | 17 | [Download](166/dataset.zip) |  |  |  |  |  |  |  |  |
| 167 | 79 | [Download](167/dataset.zip) |  |  |  |  |  |  |  |  |
| 168 | 115 | [Download](168/dataset.zip) |  |  |  |  |  |  |  |  |
| 169 | 51 | [Download](169/dataset.zip) |  |  |  |  |  |  |  |  |
| 170 | 27 | [Download](170/dataset.zip) |  |  |  |  |  |  |  |  |
| 171 | 30 | [Download](171/dataset.zip) |  |  |  |  |  |  |  |  |
| 172 | 58 | [Download](172/dataset.zip) |  |  |  |  |  |  |  |  |
| 173 | 40 | [Download](173/dataset.zip) |  |  |  |  |  |  |  |  |
| 174 | 28 | [Download](174/dataset.zip) |  |  |  |  |  |  |  |  |
| 175 | 35 | [Download](175/dataset.zip) |  |  |  |  |  |  |  |  |
| 176 | 37 | [Download](176/dataset.zip) |  |  |  |  |  |  |  |  |
| 177 | 24 | [Download](177/dataset.zip) |  |  |  |  |  |  |  |  |
| 178 | 17 | [Download](178/dataset.zip) |  |  |  |  |  |  |  |  |
| 179 | 39 | [Download](179/dataset.zip) |  |  |  |  |  |  |  |  |
| 180 | 16 | [Download](180/dataset.zip) |  |  |  |  |  |  |  |  |
| 181 | 70 | [Download](181/dataset.zip) |  |  |  |  |  |  |  |  |
| 182 | 38 | [Download](182/dataset.zip) |  |  |  |  |  |  |  |  |
| 183 | 427 | [Download](183/dataset.zip) |  |  |  |  |  |  |  |  |
| 184 | 80 | [Download](184/dataset.zip) |  |  |  |  |  |  |  |  |
| 185 | 27 | [Download](185/dataset.zip) |  |  |  |  |  |  |  |  |
| 186 | 68 | [Download](186/dataset.zip) |  |  |  |  |  |  |  |  |
| 187 | 23 | [Download](187/dataset.zip) |  |  |  |  |  |  |  |  |
| 188 | 29 | [Download](188/dataset.zip) |  |  |  |  |  |  |  |  |
| 189 | 13 | [Download](189/dataset.zip) |  |  |  |  |  |  |  |  |
| 190 | 13 | [Download](190/dataset.zip) |  |  |  |  |  |  |  |  |
| 191 | 107 | [Download](191/dataset.zip) |  |  |  |  |  |  |  |  |
| 192 | 21 | [Download](192/dataset.zip) |  |  |  |  |  |  |  |  |
| 193 | 21 | [Download](193/dataset.zip) |  |  |  |  |  |  |  |  |
| 194 | 27 | [Download](194/dataset.zip) |  |  |  |  |  |  |  |  |
| 195 | 29 | [Download](195/dataset.zip) |  |  |  |  |  |  |  |  |
| 196 | 20 | [Download](196/dataset.zip) |  |  |  |  |  |  |  |  |
| 197 | 33 | [Download](197/dataset.zip) |  |  |  |  |  |  |  |  |
| 198 | 45 | [Download](198/dataset.zip) |  |  |  |  |  |  |  |  |
| 199 | 63 | [Download](199/dataset.zip) |  |  |  |  |  |  |  |  |
| 200 | 20 | [Download](200/dataset.zip) |  |  |  |  |  |  |  |  |
| 201 | 26 | [Download](201/dataset.zip) |  |  |  |  |  |  |  |  |
| 202 | 26 | [Download](202/dataset.zip) |  |  |  |  |  |  |  |  |
| 203 | 234 | [Download](203/dataset.zip) |  |  |  |  |  |  |  |  |
| 204 | 313 | [Download](204/dataset.zip) |  |  |  |  |  |  |  |  |
| 205 | 14 | [Download](205/dataset.zip) |  |  |  |  |  |  |  |  |
| 206 | 591 | [Download](206/dataset.zip) |  |  |  |  |  |  |  |  |
| 207 | 37 | [Download](207/dataset.zip) |  |  |  |  |  |  |  |  |
| 208 | 70 | [Download](208/dataset.zip) |  |  |  |  |  |  |  |  |
| 209 | 36 | [Download](209/dataset.zip) |  |  |  |  |  |  |  |  |
| 210 | 22 | [Download](210/dataset.zip) |  |  |  |  |  |  |  |  |
| 211 | 17 | [Download](211/dataset.zip) |  |  |  |  |  |  |  |  |
| 212 | 190 | [Download](212/dataset.zip) |  |  |  |  |  |  |  |  |
| 213 | 40 | [Download](213/dataset.zip) |  |  |  |  |  |  |  |  |
| 214 | 39 | [Download](214/dataset.zip) |  |  |  |  |  |  |  |  |
| 215 | 19 | [Download](215/dataset.zip) |  |  |  |  |  |  |  |  |
| 216 | 32 | [Download](216/dataset.zip) |  |  |  |  |  |  |  |  |
| 217 | 73 | [Download](217/dataset.zip) |  |  |  |  |  |  |  |  |
| 218 | 27 | [Download](218/dataset.zip) |  |  |  |  |  |  |  |  |
| 219 | 76 | [Download](219/dataset.zip) |  |  |  |  |  |  |  |  |
| 220 | 13 | [Download](220/dataset.zip) |  |  |  |  |  |  |  |  |
| 221 | 23 | [Download](221/dataset.zip) |  |  |  |  |  |  |  |  |
| 222 | 33 | [Download](222/dataset.zip) |  |  |  |  |  |  |  |  |
| 223 | 29 | [Download](223/dataset.zip) |  |  |  |  |  |  |  |  |
| 224 | 24 | [Download](224/dataset.zip) |  |  |  |  |  |  |  |  |
| 225 | 254 | [Download](225/dataset.zip) |  |  |  |  |  |  |  |  |
| 226 | 30 | [Download](226/dataset.zip) |  |  |  |  |  |  |  |  |
| 227 | 11 | [Download](227/dataset.zip) |  |  |  |  |  |  |  |  |
| 228 | 101 | [Download](228/dataset.zip) |  |  |  |  |  |  |  |  |
| 229 | 18 | [Download](229/dataset.zip) |  |  |  |  |  |  |  |  |
| 230 | 30 | [Download](230/dataset.zip) |  |  |  |  |  |  |  |  |
| 231 | 41 | [Download](231/dataset.zip) |  |  |  |  |  |  |  |  |
| 232 | 44 | [Download](232/dataset.zip) |  |  |  |  |  |  |  |  |
| 233 | 42 | [Download](233/dataset.zip) |  |  |  |  |  |  |  |  |
| 234 | 121 | [Download](234/dataset.zip) |  |  |  |  |  |  |  |  |
| 235 | 24 | [Download](235/dataset.zip) |  |  |  |  |  |  |  |  |
| 236 | 11 | [Download](236/dataset.zip) |  |  |  |  |  |  |  |  |
| 237 | 14 | [Download](237/dataset.zip) |  |  |  |  |  |  |  |  |
| 238 | 11 | [Download](238/dataset.zip) |  |  |  |  |  |  |  |  |
| 239 | 36 | [Download](239/dataset.zip) |  |  |  |  |  |  |  |  |
| 240 | 150 | [Download](240/dataset.zip) |  |  |  |  |  |  |  |  |
| 241 | 18 | [Download](241/dataset.zip) |  |  |  |  |  |  |  |  |
| 242 | 17 | [Download](242/dataset.zip) |  |  |  |  |  |  |  |  |
| 243 | 37 | [Download](243/dataset.zip) |  |  |  |  |  |  |  |  |
| 244 | 38 | [Download](244/dataset.zip) |  |  |  |  |  |  |  |  |
| 245 | 16 | [Download](245/dataset.zip) |  |  |  |  |  |  |  |  |
| 246 | 13 | [Download](246/dataset.zip) |  |  |  |  |  |  |  |  |
| 247 | 42 | [Download](247/dataset.zip) |  |  |  |  |  |  |  |  |
| 248 | 67 | [Download](248/dataset.zip) |  |  |  |  |  |  |  |  |
| 249 | 14 | [Download](249/dataset.zip) |  |  |  |  |  |  |  |  |
| 250 | 16 | [Download](250/dataset.zip) |  |  |  |  |  |  |  |  |
| 251 | 16 | [Download](251/dataset.zip) |  |  |  |  |  |  |  |  |
| 252 | 32 | [Download](252/dataset.zip) |  |  |  |  |  |  |  |  |
| 253 | 73 | [Download](253/dataset.zip) |  |  |  |  |  |  |  |  |
| 254 | 10 | [Download](254/dataset.zip) |  |  |  |  |  |  |  |  |
| 255 | 36 | [Download](255/dataset.zip) |  |  |  |  |  |  |  |  |
| 256 | 14 | [Download](256/dataset.zip) |  |  |  |  |  |  |  |  |
| 257 | 14 | [Download](257/dataset.zip) |  |  |  |  |  |  |  |  |
| 258 | 13 | [Download](258/dataset.zip) |  |  |  |  |  |  |  |  |
| 259 | 45 | [Download](259/dataset.zip) |  |  |  |  |  |  |  |  |
| 260 | 12 | [Download](260/dataset.zip) |  |  |  |  |  |  |  |  |
| 261 | 18 | [Download](261/dataset.zip) |  |  |  |  |  |  |  |  |
| 262 | 14 | [Download](262/dataset.zip) |  |  |  |  |  |  |  |  |
| 263 | 40 | [Download](263/dataset.zip) |  |  |  |  |  |  |  |  |
| 264 | 13 | [Download](264/dataset.zip) |  |  |  |  |  |  |  |  |
| 265 | 11 | [Download](265/dataset.zip) |  |  |  |  |  |  |  |  |
| 266 | 11 | [Download](266/dataset.zip) |  |  |  |  |  |  |  |  |
| 267 | 6 | [Download](267/dataset.zip) |  |  |  |  |  |  | N/A | N/A |
| 268 | 8 | [Download](268/dataset.zip) |  |  |  |  |  |  |  |  |
| noise | 494 | [Download](-1/dataset.zip) |  |  |  |  |  |  |  |  |
|
BangumiBase/fairytail
|
[
"size_categories:10K<n<100K",
"license:mit",
"art",
"region:us"
] |
2023-10-24T18:45:11+00:00
|
{"license": "mit", "size_categories": ["10K<n<100K"], "tags": ["art"]}
|
2023-10-26T00:32:22+00:00
|
[] |
[] |
TAGS
#size_categories-10K<n<100K #license-mit #art #region-us
|
Bangumi Image Base of Fairy Tail
================================
This is the image base of bangumi Fairy Tail, we detected 270 characters, 33650 images in total. The full dataset is here.
Please note that these image bases are not guaranteed to be 100% cleaned, they may be noisy actual. If you intend to manually train models using this dataset, we recommend performing necessary preprocessing on the downloaded dataset to eliminate potential noisy samples (approximately 1% probability).
Here is the characters' preview:
|
[] |
[
"TAGS\n#size_categories-10K<n<100K #license-mit #art #region-us \n"
] |
[
25
] |
[
"passage: TAGS\n#size_categories-10K<n<100K #license-mit #art #region-us \n"
] |
60736179582811a85fa931ba2dd6e57e031f2607
|
# Dataset Card for "TalkBank_CA_TaiwanHakka"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
NathanRoll/TalkBank_CA_TaiwanHakka
|
[
"region:us"
] |
2023-10-24T18:49:06+00:00
|
{"dataset_info": {"features": [{"name": "audio", "sequence": "float32"}, {"name": "__index_level_0__", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 2411264233, "num_examples": 44}], "download_size": 2412867714, "dataset_size": 2411264233}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T18:50:42+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "TalkBank_CA_TaiwanHakka"
More Information needed
|
[
"# Dataset Card for \"TalkBank_CA_TaiwanHakka\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"TalkBank_CA_TaiwanHakka\"\n\nMore Information needed"
] |
[
6,
19
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"TalkBank_CA_TaiwanHakka\"\n\nMore Information needed"
] |
31ae7161b0b7e1d4bb56462e6ad4cd7fac2a8485
|
This dataset contains question/answer pairs from a French legal protection insurance (https://www.service-public.fr/particuliers/vosdroits/F3049?lang=en).
The objective of this dataset is to contribute to open source research projects aiming to, for instance:
* fine-tune LLMs on high-quality datasets, specializing them in the insurance domain
* develop new question/answer applications using Retrieval Augmented Generation (RAG) for insurance contracts
* assess the knowledge of language models in the insurance field
* more generally, apply LLMs to the insurance domain for better understanding and increased transparency of this industry.
Other datasets of the same kind are also available - or will be available soon - and are part of this research effort. See here: https://huggingface.co/collections/zelros/legal-protection-insurance-6536e8f389dd48faca78447e
Here is an example of usages of this dataset: https://huggingface.co/spaces/zelros/The-legal-protection-insurance-comparator
|
zelros/pj-sg
|
[
"insurance",
"region:us"
] |
2023-10-24T18:49:46+00:00
|
{"tags": ["insurance"]}
|
2023-11-06T00:27:11+00:00
|
[] |
[] |
TAGS
#insurance #region-us
|
This dataset contains question/answer pairs from a French legal protection insurance (URL
The objective of this dataset is to contribute to open source research projects aiming to, for instance:
* fine-tune LLMs on high-quality datasets, specializing them in the insurance domain
* develop new question/answer applications using Retrieval Augmented Generation (RAG) for insurance contracts
* assess the knowledge of language models in the insurance field
* more generally, apply LLMs to the insurance domain for better understanding and increased transparency of this industry.
Other datasets of the same kind are also available - or will be available soon - and are part of this research effort. See here: URL
Here is an example of usages of this dataset: URL
|
[] |
[
"TAGS\n#insurance #region-us \n"
] |
[
9
] |
[
"passage: TAGS\n#insurance #region-us \n"
] |
b6f063ba601a2ca9cc1fc94c2499dfed92c8b432
|
# Dataset Card for "TalkBank_CA_SBCSAE"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
|
NathanRoll/TalkBank_CA_SBCSAE
|
[
"region:us"
] |
2023-10-24T18:55:12+00:00
|
{"dataset_info": {"features": [{"name": "audio", "sequence": "float32"}, {"name": "__index_level_0__", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 5371599988, "num_examples": 60}], "download_size": 5378933953, "dataset_size": 5371599988}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T18:58:43+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "TalkBank_CA_SBCSAE"
More Information needed
|
[
"# Dataset Card for \"TalkBank_CA_SBCSAE\"\n\nMore Information needed"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"TalkBank_CA_SBCSAE\"\n\nMore Information needed"
] |
[
6,
18
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"TalkBank_CA_SBCSAE\"\n\nMore Information needed"
] |
451a947c335fd786dc166b02d9e7c3fe5bdc5689
|
# Dataset Card for "TalkBank_CA_GCSAusE"
This dataset has been extracted, downfolded, and resampled from the TalkBank CA dataset. Please cite the original work:
```Haugh, Michael and Wei-Lin Melody Chang (2013). Collaborative creation of spoken language corpora. In Tim Greer, Yuriko Kite and Donna Tatsuki (eds.),Pragmatics and Language Learning. Volume 13 (pp.133-159), National Foreign Language Resource Center, University of Hawai’i, Honolulu```
|
NathanRoll/TalkBank_CA_GCSAusE
|
[
"region:us"
] |
2023-10-24T19:00:06+00:00
|
{"dataset_info": {"features": [{"name": "audio", "sequence": "float32"}, {"name": "__index_level_0__", "dtype": "string"}], "splits": [{"name": "train", "num_bytes": 832432648, "num_examples": 36}], "download_size": 833471504, "dataset_size": 832432648}, "configs": [{"config_name": "default", "data_files": [{"split": "train", "path": "data/train-*"}]}]}
|
2023-10-24T21:20:01+00:00
|
[] |
[] |
TAGS
#region-us
|
# Dataset Card for "TalkBank_CA_GCSAusE"
This dataset has been extracted, downfolded, and resampled from the TalkBank CA dataset. Please cite the original work:
|
[
"# Dataset Card for \"TalkBank_CA_GCSAusE\"\n\nThis dataset has been extracted, downfolded, and resampled from the TalkBank CA dataset. Please cite the original work:"
] |
[
"TAGS\n#region-us \n",
"# Dataset Card for \"TalkBank_CA_GCSAusE\"\n\nThis dataset has been extracted, downfolded, and resampled from the TalkBank CA dataset. Please cite the original work:"
] |
[
6,
46
] |
[
"passage: TAGS\n#region-us \n# Dataset Card for \"TalkBank_CA_GCSAusE\"\n\nThis dataset has been extracted, downfolded, and resampled from the TalkBank CA dataset. Please cite the original work:"
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.