Hugging Face's logo

madatnlp
/
ke-t5-scratch

Text2Text Generation
Transformers
TensorFlow
t5
generated_from_keras_callback
Model card Files Community
madatnlp commited on
Commit
9ed7f46
·
1 Parent(s): 4f6593b

Training in progress epoch 0

Browse files
Files changed (2) hide show
  1. README.md +5 -32
  2. tf_model.h5 +1 -1
README.md CHANGED
@@ -13,9 +13,9 @@ probably proofread and complete it, then remove this comment. -->
13
 
14
  This model is a fine-tuned version of [madatnlp/ke-t5-math-py](https://huggingface.co/madatnlp/ke-t5-math-py) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
- - Train Loss: 0.2662
17
- - Validation Loss: 0.7210
18
- - Epoch: 27
19
 
20
  ## Model description
21
 
@@ -34,41 +34,14 @@ More information needed
34
  ### Training hyperparameters
35
 
36
  The following hyperparameters were used during training:
37
- - optimizer: {'name': 'Adam', 'learning_rate': 0.001, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
38
  - training_precision: float32
39
 
40
  ### Training results
41
 
42
  | Train Loss | Validation Loss | Epoch |
43
  |:----------:|:---------------:|:-----:|
44
- | 3.9330 | 1.9960 | 0 |
45
- | 1.9578 | 1.4484 | 1 |
46
- | 1.5358 | 1.3202 | 2 |
47
- | 1.3807 | 1.1251 | 3 |
48
- | 1.2885 | 1.0331 | 4 |
49
- | 1.1943 | 1.0004 | 5 |
50
- | 1.1366 | 0.9263 | 6 |
51
- | 1.0507 | 0.8866 | 7 |
52
- | 1.0160 | 0.8788 | 8 |
53
- | 0.9553 | 0.8301 | 9 |
54
- | 0.9149 | 0.8480 | 10 |
55
- | 0.8545 | 0.8021 | 11 |
56
- | 0.8271 | 0.7890 | 12 |
57
- | 0.7783 | 0.7549 | 13 |
58
- | 0.7166 | 0.6960 | 14 |
59
- | 0.6853 | 0.6828 | 15 |
60
- | 0.6142 | 0.7129 | 16 |
61
- | 0.5774 | 0.6368 | 17 |
62
- | 0.5612 | 0.6432 | 18 |
63
- | 0.4980 | 0.6483 | 19 |
64
- | 0.4723 | 0.6485 | 20 |
65
- | 0.4245 | 0.6569 | 21 |
66
- | 0.4040 | 0.6494 | 22 |
67
- | 0.3733 | 0.6970 | 23 |
68
- | 0.3461 | 0.7069 | 24 |
69
- | 0.3166 | 0.6597 | 25 |
70
- | 0.2912 | 0.6372 | 26 |
71
- | 0.2662 | 0.7210 | 27 |
72
 
73
 
74
  ### Framework versions
 
13
 
14
  This model is a fine-tuned version of [madatnlp/ke-t5-math-py](https://huggingface.co/madatnlp/ke-t5-math-py) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
+ - Train Loss: 13.5076
17
+ - Validation Loss: 11.8125
18
+ - Epoch: 0
19
 
20
  ## Model description
21
 
 
34
  ### Training hyperparameters
35
 
36
  The following hyperparameters were used during training:
37
+ - optimizer: {'name': 'Adam', 'learning_rate': 1e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
38
  - training_precision: float32
39
 
40
  ### Training results
41
 
42
  | Train Loss | Validation Loss | Epoch |
43
  |:----------:|:---------------:|:-----:|
44
+ | 13.5076 | 11.8125 | 0 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
45
 
46
 
47
  ### Framework versions
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:75047029361c87147fddf308c342e6c668df9506dcc048984e4b9671edda7d05
3
  size 831509840
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba93083be54c6ab91c48c6dab375b7fb03583825f33c4f2d84d3c75876f8f83a
3
  size 831509840