Model save

.ipynb_checkpoints/run-checkpoint.sh

@@ -1,6 +1,6 @@
 python run_speech_recognition_ctc.py \
 	--dataset_name="common_voice" \
-	--model_name_or_path="facebook/wav2vec2-large-xlsr-53" \
+	--model_name_or_path="hf-test/xls-r-dummy" \
 	--dataset_config_name="ab" \
 	--output_dir="./" \
 	--overwrite_output_dir \

README.md

@@ -1,64 +0,0 @@
----
-language:
-- ab
-license: apache-2.0
-tags:
-- automatic-speech-recognition
-- common_voice
-- generated_from_trainer
-datasets:
-- common_voice
-model-index:
-- name: ''
-  results: []
----
-
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
-
-# 
-
-This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the COMMON_VOICE - AB dataset.
-It achieves the following results on the evaluation set:
-- Loss: 15.1833
-- Wer: 1.0
-
-## Model description
-
-More information needed
-
-## Intended uses & limitations
-
-More information needed
-
-## Training and evaluation data
-
-More information needed
-
-## Training procedure
-
-### Training hyperparameters
-
-The following hyperparameters were used during training:
-- learning_rate: 0.0003
-- train_batch_size: 16
-- eval_batch_size: 8
-- seed: 42
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 32
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 500
-- num_epochs: 2.0
-- mixed_precision_training: Native AMP
-
-### Training results
-
-
-
-### Framework versions
-
-- Transformers 4.17.0.dev0
-- Pytorch 1.10.2+cu102
-- Datasets 1.18.2.dev0
-- Tokenizers 0.11.0

all_results.json

@@ -1,14 +0,0 @@
-{
-    "epoch": 2.0,
-    "eval_loss": 15.183259963989258,
-    "eval_runtime": 0.5122,
-    "eval_samples": 9,
-    "eval_samples_per_second": 17.572,
-    "eval_steps_per_second": 3.905,
-    "eval_wer": 1.0,
-    "train_loss": 21.219526290893555,
-    "train_runtime": 4.3872,
-    "train_samples": 22,
-    "train_samples_per_second": 10.029,
-    "train_steps_per_second": 0.456
-}

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
+  "_name_or_path": "hf-test/xls-r-dummy",
   "activation_dropout": 0.0,
   "adapter_kernel_size": 3,
   "adapter_stride": 2,
@@ -11,35 +11,23 @@
   "attention_dropout": 0.0,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
-  "codevector_dim": 768,
+  "codevector_dim": 256,
   "contrastive_logits_temperature": 0.1,
-  "conv_bias": true,
+  "conv_bias": false,
   "conv_dim": [
-    512,
-    512,
-    512,
-    512,
-    512,
-    512,
-    512
+    32,
+    32,
+    32
   ],
   "conv_kernel": [
-    10,
-    3,
-    3,
-    3,
-    3,
-    2,
-    2
+    8,
+    8,
+    8
   ],
   "conv_stride": [
-    5,
-    2,
-    2,
-    2,
-    2,
-    2,
-    2
+    4,
+    4,
+    4
   ],
   "ctc_loss_reduction": "mean",
   "ctc_zero_infinity": false,
@@ -54,38 +42,31 @@
   "final_dropout": 0.0,
   "hidden_act": "gelu",
   "hidden_dropout": 0.0,
-  "hidden_size": 1024,
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 16,
   "initializer_range": 0.02,
-  "intermediate_size": 4096,
+  "intermediate_size": 20,
   "layer_norm_eps": 1e-05,
   "layerdrop": 0.0,
-  "mask_channel_length": 10,
-  "mask_channel_min_space": 1,
-  "mask_channel_other": 0.0,
-  "mask_channel_prob": 0.0,
-  "mask_channel_selection": "static",
   "mask_feature_length": 10,
   "mask_feature_min_masks": 0,
   "mask_feature_prob": 0.0,
   "mask_time_length": 10,
   "mask_time_min_masks": 2,
-  "mask_time_min_space": 1,
-  "mask_time_other": 0.0,
   "mask_time_prob": 0.05,
-  "mask_time_selection": "static",
   "model_type": "wav2vec2",
   "num_adapter_layers": 3,
-  "num_attention_heads": 16,
+  "num_attention_heads": 2,
   "num_codevector_groups": 2,
   "num_codevectors_per_group": 320,
-  "num_conv_pos_embedding_groups": 16,
-  "num_conv_pos_embeddings": 128,
-  "num_feat_extract_layers": 7,
-  "num_hidden_layers": 24,
-  "num_negatives": 100,
-  "output_hidden_size": 1024,
+  "num_conv_pos_embedding_groups": 2,
+  "num_conv_pos_embeddings": 16,
+  "num_feat_extract_layers": 3,
+  "num_hidden_layers": 4,
+  "num_negatives": 10,
+  "output_hidden_size": 16,
   "pad_token_id": 39,
-  "proj_codevector_dim": 768,
+  "proj_codevector_dim": 256,
   "tdnn_dilation": [
     1,
     2,

eval_results.json

@@ -1,9 +0,0 @@
-{
-    "epoch": 2.0,
-    "eval_loss": 15.183259963989258,
-    "eval_runtime": 0.5122,
-    "eval_samples": 9,
-    "eval_samples_per_second": 17.572,
-    "eval_steps_per_second": 3.905,
-    "eval_wer": 1.0
-}

preprocessor_config.json

@@ -3,7 +3,7 @@
   "feature_extractor_type": "Wav2Vec2FeatureExtractor",
   "feature_size": 1,
   "padding_side": "right",
-  "padding_value": 0,
-  "return_attention_mask": true,
+  "padding_value": 0.0,
+  "return_attention_mask": false,
   "sampling_rate": 16000
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:910c4a1f6ba987dc94174de99e17aea1ba16b0b098742df52036cd0086b2cb63
-size 1262095857
+oid sha256:ac89a4d5306b1b0af49e557b2cf55ff2eae830d8e96f3da9876d516bfe96bf71
+size 143142

run.sh

@@ -1,6 +1,6 @@
 python run_speech_recognition_ctc.py \
 	--dataset_name="common_voice" \
-	--model_name_or_path="facebook/wav2vec2-large-xlsr-53" \
+	--model_name_or_path="hf-test/xls-r-dummy" \
 	--dataset_config_name="ab" \
 	--output_dir="./" \
 	--overwrite_output_dir \

train_results.json

@@ -1,8 +0,0 @@
-{
-    "epoch": 2.0,
-    "train_loss": 21.219526290893555,
-    "train_runtime": 4.3872,
-    "train_samples": 22,
-    "train_samples_per_second": 10.029,
-    "train_steps_per_second": 0.456
-}

trainer_state.json

@@ -1,25 +0,0 @@
-{
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 2.0,
-  "global_step": 2,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 2.0,
-      "step": 2,
-      "total_flos": 1.171423688905728e+16,
-      "train_loss": 21.219526290893555,
-      "train_runtime": 4.3872,
-      "train_samples_per_second": 10.029,
-      "train_steps_per_second": 0.456
-    }
-  ],
-  "max_steps": 2,
-  "num_train_epochs": 2,
-  "total_flos": 1.171423688905728e+16,
-  "trial_name": null,
-  "trial_params": null
-}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac91995bba3613eae578bea40f065315404269cabdcaf2cee979d28ab550d414
+oid sha256:2abf70ee833af8d8c22a2c8a2cbecfd532a51e2b5d92efc5025a369bcd756a19
 size 2991