AndrewMcDowell commited on
Commit
ef2237f
1 Parent(s): b2eecfc

End of training

Browse files
.ipynb_checkpoints/resume_training-checkpoint.sh ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ python run_speech_recognition_ctc.py \
2
+ --dataset_name="mozilla-foundation/common_voice_7_0" \
3
+ --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
4
+ --dataset_config_name="ar" \
5
+ --output_dir="./" \
6
+ --num_train_epochs="5" \
7
+ --per_device_train_batch_size="8" \
8
+ --per_device_eval_batch_size="8" \
9
+ --gradient_accumulation_steps="4" \
10
+ --learning_rate="7.5e-5" \
11
+ --warmup_steps="2000" \
12
+ --length_column_name="input_length" \
13
+ --evaluation_strategy="steps" \
14
+ --text_column_name="sentence" \
15
+ --chars_to_ignore , ? . ! \- \; \: \" “ % ‘ ” � — ’ … – \
16
+ --save_steps="500" \
17
+ --eval_steps="500" \
18
+ --logging_steps="100" \
19
+ --layerdrop="0.0" \
20
+ --activation_dropout="0.1" \
21
+ --save_total_limit="3" \
22
+ --freeze_feature_encoder \
23
+ --feat_proj_dropout="0.0" \
24
+ --mask_time_prob="0.75" \
25
+ --mask_time_length="10" \
26
+ --mask_feature_prob="0.25" \
27
+ --mask_feature_length="64" \
28
+ --gradient_checkpointing \
29
+ --use_auth_token \
30
+ --fp16 \
31
+ --group_by_length \
32
+ --do_train --do_eval \
33
+ --push_to_hub
all_results.json CHANGED
@@ -1,14 +1,14 @@
1
  {
2
- "epoch": 10.0,
3
- "eval_loss": 0.4525286555290222,
4
- "eval_runtime": 319.7474,
5
  "eval_samples": 10284,
6
- "eval_samples_per_second": 32.163,
7
- "eval_steps_per_second": 4.022,
8
- "eval_wer": 0.4801227197959599,
9
- "train_loss": 2.2287146272217635,
10
- "train_runtime": 23537.8544,
11
  "train_samples": 37318,
12
- "train_samples_per_second": 15.854,
13
- "train_steps_per_second": 0.495
14
  }
 
1
  {
2
+ "epoch": 9.86,
3
+ "eval_loss": 0.45016008615493774,
4
+ "eval_runtime": 332.1791,
5
  "eval_samples": 10284,
6
+ "eval_samples_per_second": 30.959,
7
+ "eval_steps_per_second": 3.871,
8
+ "eval_wer": 0.4782560482007873,
9
+ "train_loss": 0.0,
10
+ "train_runtime": 3.4176,
11
  "train_samples": 37318,
12
+ "train_samples_per_second": 54596.703,
13
+ "train_steps_per_second": 1705.873
14
  }
eval_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 10.0,
3
- "eval_loss": 0.4525286555290222,
4
- "eval_runtime": 319.7474,
5
  "eval_samples": 10284,
6
- "eval_samples_per_second": 32.163,
7
- "eval_steps_per_second": 4.022,
8
- "eval_wer": 0.4801227197959599
9
  }
 
1
  {
2
+ "epoch": 9.86,
3
+ "eval_loss": 0.45016008615493774,
4
+ "eval_runtime": 332.1791,
5
  "eval_samples": 10284,
6
+ "eval_samples_per_second": 30.959,
7
+ "eval_steps_per_second": 3.871,
8
+ "eval_wer": 0.4782560482007873
9
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:32a6c4cdbb67e5027c4796ff7b91aac984f0b5ba33cdc497ef815a547b4926cc
3
  size 1262313137
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:023ddf68990793400757e39049470fe7511fb9501eb8e50e4a024fc333169ede
3
  size 1262313137
resume_training.sh ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ python run_speech_recognition_ctc.py \
2
+ --dataset_name="mozilla-foundation/common_voice_7_0" \
3
+ --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
4
+ --dataset_config_name="ar" \
5
+ --output_dir="./" \
6
+ --num_train_epochs="5" \
7
+ --per_device_train_batch_size="8" \
8
+ --per_device_eval_batch_size="8" \
9
+ --gradient_accumulation_steps="4" \
10
+ --learning_rate="7.5e-5" \
11
+ --warmup_steps="2000" \
12
+ --length_column_name="input_length" \
13
+ --evaluation_strategy="steps" \
14
+ --text_column_name="sentence" \
15
+ --chars_to_ignore , ? . ! \- \; \: \" “ % ‘ ” � — ’ … – \
16
+ --save_steps="500" \
17
+ --eval_steps="500" \
18
+ --logging_steps="100" \
19
+ --layerdrop="0.0" \
20
+ --activation_dropout="0.1" \
21
+ --save_total_limit="3" \
22
+ --freeze_feature_encoder \
23
+ --feat_proj_dropout="0.0" \
24
+ --mask_time_prob="0.75" \
25
+ --mask_time_length="10" \
26
+ --mask_feature_prob="0.25" \
27
+ --mask_feature_length="64" \
28
+ --gradient_checkpointing \
29
+ --use_auth_token \
30
+ --fp16 \
31
+ --group_by_length \
32
+ --do_train --do_eval \
33
+ --push_to_hub
special_tokens_map.json CHANGED
@@ -1 +1 @@
1
- {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 10.0,
3
- "train_loss": 2.2287146272217635,
4
- "train_runtime": 23537.8544,
5
  "train_samples": 37318,
6
- "train_samples_per_second": 15.854,
7
- "train_steps_per_second": 0.495
8
  }
 
1
  {
2
+ "epoch": 9.86,
3
+ "train_loss": 0.0,
4
+ "train_runtime": 3.4176,
5
  "train_samples": 37318,
6
+ "train_samples_per_second": 54596.703,
7
+ "train_steps_per_second": 1705.873
8
  }
trainer_state.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 9.99978563772776,
5
- "global_step": 11660,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
@@ -905,24 +905,18 @@
905
  "step": 11500
906
  },
907
  {
908
- "epoch": 9.95,
909
- "learning_rate": 5.124223602484471e-07,
910
- "loss": 1.5295,
911
- "step": 11600
912
- },
913
- {
914
- "epoch": 10.0,
915
- "step": 11660,
916
- "total_flos": 3.1970153172049187e+19,
917
- "train_loss": 2.2287146272217635,
918
- "train_runtime": 23537.8544,
919
- "train_samples_per_second": 15.854,
920
- "train_steps_per_second": 0.495
921
  }
922
  ],
923
- "max_steps": 11660,
924
- "num_train_epochs": 10,
925
- "total_flos": 3.1970153172049187e+19,
926
  "trial_name": null,
927
  "trial_params": null
928
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 9.862593783494106,
5
+ "global_step": 11500,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
 
905
  "step": 11500
906
  },
907
  {
908
+ "epoch": 9.86,
909
+ "step": 11500,
910
+ "total_flos": 3.154318018894781e+19,
911
+ "train_loss": 0.0,
912
+ "train_runtime": 3.4176,
913
+ "train_samples_per_second": 54596.703,
914
+ "train_steps_per_second": 1705.873
 
 
 
 
 
 
915
  }
916
  ],
917
+ "max_steps": 5830,
918
+ "num_train_epochs": 5,
919
+ "total_flos": 3.154318018894781e+19,
920
  "trial_name": null,
921
  "trial_params": null
922
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8344c1dd84c70d7463dac7514ce6652d1a3d3c39cd05da2c42ca2ebd3a3a2e0b
3
  size 2991
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:796dd8075596c17751204717a11d24106a793cf53118e126c45cc4460fd7b264
3
  size 2991