Plim commited on
Commit
e8b3151
1 Parent(s): 505559b

clean start

Browse files
Files changed (34) hide show
  1. .gitignore +3 -1
  2. .ipynb_checkpoints/README-checkpoint.md +0 -70
  3. .ipynb_checkpoints/run-checkpoint.sh +3 -5
  4. README.md +0 -62
  5. all_results.json +0 -14
  6. config.json +0 -107
  7. eval_results.json +0 -9
  8. preprocessor_config.json +0 -9
  9. pytorch_model.bin +0 -3
  10. run.sh +3 -5
  11. train_results.json +0 -8
  12. trainer_state.json +0 -25
  13. training_args.bin +0 -3
  14. wandb/debug-internal.log +0 -1
  15. wandb/debug.log +0 -1
  16. wandb/latest-run +0 -1
  17. wandb/run-20220130_224738-2uzt3kt1/files/conda-environment.yaml +0 -0
  18. wandb/run-20220130_224738-2uzt3kt1/files/config.yaml +0 -698
  19. wandb/run-20220130_224738-2uzt3kt1/files/output.log +0 -66
  20. wandb/run-20220130_224738-2uzt3kt1/files/requirements.txt +0 -180
  21. wandb/run-20220130_224738-2uzt3kt1/files/wandb-metadata.json +0 -63
  22. wandb/run-20220130_224738-2uzt3kt1/files/wandb-summary.json +0 -1
  23. wandb/run-20220130_224738-2uzt3kt1/logs/debug-internal.log +0 -210
  24. wandb/run-20220130_224738-2uzt3kt1/logs/debug.log +0 -146
  25. wandb/run-20220130_224738-2uzt3kt1/run-2uzt3kt1.wandb +0 -0
  26. wandb/run-20220130_230018-ktkg6ghu/files/conda-environment.yaml +0 -0
  27. wandb/run-20220130_230018-ktkg6ghu/files/config.yaml +0 -692
  28. wandb/run-20220130_230018-ktkg6ghu/files/output.log +0 -62
  29. wandb/run-20220130_230018-ktkg6ghu/files/requirements.txt +0 -180
  30. wandb/run-20220130_230018-ktkg6ghu/files/wandb-metadata.json +0 -63
  31. wandb/run-20220130_230018-ktkg6ghu/files/wandb-summary.json +0 -1
  32. wandb/run-20220130_230018-ktkg6ghu/logs/debug-internal.log +0 -110
  33. wandb/run-20220130_230018-ktkg6ghu/logs/debug.log +0 -24
  34. wandb/run-20220130_230018-ktkg6ghu/run-ktkg6ghu.wandb +0 -0
.gitignore CHANGED
@@ -1 +1,3 @@
1
- checkpoint-*/
 
 
1
+ checkpoint-*/
2
+
3
+ wandb
.ipynb_checkpoints/README-checkpoint.md DELETED
@@ -1,70 +0,0 @@
1
- ---
2
- language:
3
- - fr
4
- license: apache-2.0
5
- tags:
6
- - automatic-speech-recognition
7
- - mozilla-foundation/common_voice_7_0
8
- - generated_from_trainer
9
- datasets:
10
- - common_voice
11
- model-index:
12
- - name: ''
13
- results: []
14
- ---
15
-
16
- <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
- should probably proofread and complete it, then remove this comment. -->
18
-
19
- #
20
-
21
- This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - FR dataset.
22
- It achieves the following results on the evaluation set:
23
- - Loss: 0.5417
24
- - Wer: 0.4479
25
-
26
- ## Model description
27
-
28
- More information needed
29
-
30
- ## Intended uses & limitations
31
-
32
- More information needed
33
-
34
- ## Training and evaluation data
35
-
36
- More information needed
37
-
38
- ## Training procedure
39
-
40
- ### Training hyperparameters
41
-
42
- The following hyperparameters were used during training:
43
- - learning_rate: 7.5e-05
44
- - train_batch_size: 8
45
- - eval_batch_size: 8
46
- - seed: 42
47
- - gradient_accumulation_steps: 4
48
- - total_train_batch_size: 32
49
- - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
- - lr_scheduler_type: linear
51
- - lr_scheduler_warmup_steps: 2000
52
- - num_epochs: 0.2
53
- - mixed_precision_training: Native AMP
54
-
55
- ### Training results
56
-
57
- | Training Loss | Epoch | Step | Validation Loss | Wer |
58
- |:-------------:|:-----:|:----:|:---------------:|:------:|
59
- | 6.9106 | 0.04 | 500 | 6.7171 | 1.0 |
60
- | 3.0034 | 0.08 | 1000 | 3.0126 | 1.0 |
61
- | 2.8699 | 0.12 | 1500 | 2.8509 | 0.9817 |
62
- | 1.629 | 0.16 | 2000 | 0.7787 | 0.5861 |
63
-
64
-
65
- ### Framework versions
66
-
67
- - Transformers 4.17.0.dev0
68
- - Pytorch 1.10.2+cu102
69
- - Datasets 1.18.2.dev0
70
- - Tokenizers 0.11.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
.ipynb_checkpoints/run-checkpoint.sh CHANGED
@@ -20,14 +20,12 @@ python run_speech_recognition_ctc.py \
20
  --mask_feature_prob="0.25" \
21
  --mask_time_length="10" \
22
  --mask_time_prob="0.75" \
23
- --max_train_samples="1000" \
24
- --max_eval_samples="200" \
25
  --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
26
- --num_train_epochs="0.4" \
27
  --output_dir="./" \
28
  --overwrite_output_dir \
29
- --per_device_train_batch_size="8" \
30
- --per_device_eval_batch_size="8" \
31
  --preprocessing_num_workers="4" \
32
  --push_to_hub \
33
  --report_to="wandb" \
20
  --mask_feature_prob="0.25" \
21
  --mask_time_length="10" \
22
  --mask_time_prob="0.75" \
 
 
23
  --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
24
+ --num_train_epochs="2.0" \
25
  --output_dir="./" \
26
  --overwrite_output_dir \
27
+ --per_device_train_batch_size="16" \
28
+ --per_device_eval_batch_size="16" \
29
  --preprocessing_num_workers="4" \
30
  --push_to_hub \
31
  --report_to="wandb" \
README.md DELETED
@@ -1,62 +0,0 @@
1
- ---
2
- language:
3
- - fr
4
- license: apache-2.0
5
- tags:
6
- - automatic-speech-recognition
7
- - mozilla-foundation/common_voice_7_0
8
- - generated_from_trainer
9
- model-index:
10
- - name: ''
11
- results: []
12
- ---
13
-
14
- <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
- should probably proofread and complete it, then remove this comment. -->
16
-
17
- #
18
-
19
- This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - FR dataset.
20
- It achieves the following results on the evaluation set:
21
- - Loss: 16.9129
22
- - Wer: 2.3789
23
-
24
- ## Model description
25
-
26
- More information needed
27
-
28
- ## Intended uses & limitations
29
-
30
- More information needed
31
-
32
- ## Training and evaluation data
33
-
34
- More information needed
35
-
36
- ## Training procedure
37
-
38
- ### Training hyperparameters
39
-
40
- The following hyperparameters were used during training:
41
- - learning_rate: 7.5e-05
42
- - train_batch_size: 8
43
- - eval_batch_size: 8
44
- - seed: 42
45
- - gradient_accumulation_steps: 8
46
- - total_train_batch_size: 64
47
- - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
- - lr_scheduler_type: linear
49
- - lr_scheduler_warmup_steps: 2000
50
- - num_epochs: 0.4
51
- - mixed_precision_training: Native AMP
52
-
53
- ### Training results
54
-
55
-
56
-
57
- ### Framework versions
58
-
59
- - Transformers 4.17.0.dev0
60
- - Pytorch 1.10.2+cu102
61
- - Datasets 1.18.2.dev0
62
- - Tokenizers 0.11.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
all_results.json DELETED
@@ -1,14 +0,0 @@
1
- {
2
- "epoch": 0.38,
3
- "eval_loss": 16.912879943847656,
4
- "eval_runtime": 8.6337,
5
- "eval_samples": 200,
6
- "eval_samples_per_second": 23.165,
7
- "eval_steps_per_second": 2.896,
8
- "eval_wer": 2.3789039481437833,
9
- "train_loss": 13.584136962890625,
10
- "train_runtime": 23.2007,
11
- "train_samples": 1000,
12
- "train_samples_per_second": 17.241,
13
- "train_steps_per_second": 0.259
14
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
config.json DELETED
@@ -1,107 +0,0 @@
1
- {
2
- "_name_or_path": "facebook/wav2vec2-xls-r-300m",
3
- "activation_dropout": 0.1,
4
- "adapter_kernel_size": 3,
5
- "adapter_stride": 2,
6
- "add_adapter": false,
7
- "apply_spec_augment": true,
8
- "architectures": [
9
- "Wav2Vec2ForCTC"
10
- ],
11
- "attention_dropout": 0.0,
12
- "bos_token_id": 1,
13
- "classifier_proj_size": 256,
14
- "codevector_dim": 768,
15
- "contrastive_logits_temperature": 0.1,
16
- "conv_bias": true,
17
- "conv_dim": [
18
- 512,
19
- 512,
20
- 512,
21
- 512,
22
- 512,
23
- 512,
24
- 512
25
- ],
26
- "conv_kernel": [
27
- 10,
28
- 3,
29
- 3,
30
- 3,
31
- 3,
32
- 2,
33
- 2
34
- ],
35
- "conv_stride": [
36
- 5,
37
- 2,
38
- 2,
39
- 2,
40
- 2,
41
- 2,
42
- 2
43
- ],
44
- "ctc_loss_reduction": "mean",
45
- "ctc_zero_infinity": false,
46
- "diversity_loss_weight": 0.1,
47
- "do_stable_layer_norm": true,
48
- "eos_token_id": 2,
49
- "feat_extract_activation": "gelu",
50
- "feat_extract_dropout": 0.0,
51
- "feat_extract_norm": "layer",
52
- "feat_proj_dropout": 0.0,
53
- "feat_quantizer_dropout": 0.0,
54
- "final_dropout": 0.0,
55
- "hidden_act": "gelu",
56
- "hidden_dropout": 0.0,
57
- "hidden_size": 1024,
58
- "initializer_range": 0.02,
59
- "intermediate_size": 4096,
60
- "layer_norm_eps": 1e-05,
61
- "layerdrop": 0.0,
62
- "mask_feature_length": 64,
63
- "mask_feature_min_masks": 0,
64
- "mask_feature_prob": 0.25,
65
- "mask_time_length": 10,
66
- "mask_time_min_masks": 2,
67
- "mask_time_prob": 0.75,
68
- "model_type": "wav2vec2",
69
- "num_adapter_layers": 3,
70
- "num_attention_heads": 16,
71
- "num_codevector_groups": 2,
72
- "num_codevectors_per_group": 320,
73
- "num_conv_pos_embedding_groups": 16,
74
- "num_conv_pos_embeddings": 128,
75
- "num_feat_extract_layers": 7,
76
- "num_hidden_layers": 24,
77
- "num_negatives": 100,
78
- "output_hidden_size": 1024,
79
- "pad_token_id": 40,
80
- "proj_codevector_dim": 768,
81
- "tdnn_dilation": [
82
- 1,
83
- 2,
84
- 3,
85
- 1,
86
- 1
87
- ],
88
- "tdnn_dim": [
89
- 512,
90
- 512,
91
- 512,
92
- 512,
93
- 1500
94
- ],
95
- "tdnn_kernel": [
96
- 5,
97
- 3,
98
- 3,
99
- 1,
100
- 1
101
- ],
102
- "torch_dtype": "float32",
103
- "transformers_version": "4.17.0.dev0",
104
- "use_weighted_layer_sum": false,
105
- "vocab_size": 41,
106
- "xvector_output_dim": 512
107
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
eval_results.json DELETED
@@ -1,9 +0,0 @@
1
- {
2
- "epoch": 0.38,
3
- "eval_loss": 16.912879943847656,
4
- "eval_runtime": 8.6337,
5
- "eval_samples": 200,
6
- "eval_samples_per_second": 23.165,
7
- "eval_steps_per_second": 2.896,
8
- "eval_wer": 2.3789039481437833
9
- }
 
 
 
 
 
 
 
 
 
preprocessor_config.json DELETED
@@ -1,9 +0,0 @@
1
- {
2
- "do_normalize": true,
3
- "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
- "feature_size": 1,
5
- "padding_side": "right",
6
- "padding_value": 0,
7
- "return_attention_mask": true,
8
- "sampling_rate": 16000
9
- }
 
 
 
 
 
 
 
 
 
pytorch_model.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:d1d7bd7ffd2ed6a01faf9f143c79eb1c4a1e163dd1a62f92c26af28850511bcd
3
- size 1262091761
 
 
 
run.sh CHANGED
@@ -20,14 +20,12 @@ python run_speech_recognition_ctc.py \
20
  --mask_feature_prob="0.25" \
21
  --mask_time_length="10" \
22
  --mask_time_prob="0.75" \
23
- --max_train_samples="1000" \
24
- --max_eval_samples="200" \
25
  --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
26
- --num_train_epochs="0.4" \
27
  --output_dir="./" \
28
  --overwrite_output_dir \
29
- --per_device_train_batch_size="8" \
30
- --per_device_eval_batch_size="8" \
31
  --preprocessing_num_workers="4" \
32
  --push_to_hub \
33
  --report_to="wandb" \
20
  --mask_feature_prob="0.25" \
21
  --mask_time_length="10" \
22
  --mask_time_prob="0.75" \
 
 
23
  --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
24
+ --num_train_epochs="2.0" \
25
  --output_dir="./" \
26
  --overwrite_output_dir \
27
+ --per_device_train_batch_size="16" \
28
+ --per_device_eval_batch_size="16" \
29
  --preprocessing_num_workers="4" \
30
  --push_to_hub \
31
  --report_to="wandb" \
train_results.json DELETED
@@ -1,8 +0,0 @@
1
- {
2
- "epoch": 0.38,
3
- "train_loss": 13.584136962890625,
4
- "train_runtime": 23.2007,
5
- "train_samples": 1000,
6
- "train_samples_per_second": 17.241,
7
- "train_steps_per_second": 0.259
8
- }
 
 
 
 
 
 
 
 
trainer_state.json DELETED
@@ -1,25 +0,0 @@
1
- {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 0.384,
5
- "global_step": 6,
6
- "is_hyper_param_search": false,
7
- "is_local_process_zero": true,
8
- "is_world_process_zero": true,
9
- "log_history": [
10
- {
11
- "epoch": 0.38,
12
- "step": 6,
13
- "total_flos": 5.41371015650304e+16,
14
- "train_loss": 13.584136962890625,
15
- "train_runtime": 23.2007,
16
- "train_samples_per_second": 17.241,
17
- "train_steps_per_second": 0.259
18
- }
19
- ],
20
- "max_steps": 6,
21
- "num_train_epochs": 1,
22
- "total_flos": 5.41371015650304e+16,
23
- "trial_name": null,
24
- "trial_params": null
25
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
training_args.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:a4f6b5e530f353910710ef95150e349ddf6f70545b8095619803d7da46090983
3
- size 2991
 
 
 
wandb/debug-internal.log DELETED
@@ -1 +0,0 @@
1
- run-20220130_230018-ktkg6ghu/logs/debug-internal.log
 
wandb/debug.log DELETED
@@ -1 +0,0 @@
1
- run-20220130_230018-ktkg6ghu/logs/debug.log
 
wandb/latest-run DELETED
@@ -1 +0,0 @@
1
- run-20220130_230018-ktkg6ghu
 
wandb/run-20220130_224738-2uzt3kt1/files/conda-environment.yaml DELETED
File without changes
wandb/run-20220130_224738-2uzt3kt1/files/config.yaml DELETED
@@ -1,698 +0,0 @@
1
- wandb_version: 1
2
-
3
- _n_gpu:
4
- desc: null
5
- value: 1
6
- _name_or_path:
7
- desc: null
8
- value: facebook/wav2vec2-xls-r-300m
9
- _wandb:
10
- desc: null
11
- value:
12
- cli_version: 0.12.9
13
- framework: huggingface
14
- huggingface_version: 4.17.0.dev0
15
- is_jupyter_run: false
16
- is_kaggle_kernel: false
17
- m:
18
- - 1: train/global_step
19
- 6:
20
- - 3
21
- - 1: train/train_runtime
22
- 5: 1
23
- 6:
24
- - 1
25
- - 1: train/train_samples_per_second
26
- 5: 1
27
- 6:
28
- - 1
29
- - 1: train/train_steps_per_second
30
- 5: 1
31
- 6:
32
- - 1
33
- - 1: train/total_flos
34
- 5: 1
35
- 6:
36
- - 1
37
- - 1: train/train_loss
38
- 5: 1
39
- 6:
40
- - 1
41
- - 1: train/epoch
42
- 5: 1
43
- 6:
44
- - 1
45
- - 1: eval/loss
46
- 5: 1
47
- 6:
48
- - 1
49
- - 1: eval/wer
50
- 5: 1
51
- 6:
52
- - 1
53
- - 1: eval/runtime
54
- 5: 1
55
- 6:
56
- - 1
57
- - 1: eval/samples_per_second
58
- 5: 1
59
- 6:
60
- - 1
61
- - 1: eval/steps_per_second
62
- 5: 1
63
- 6:
64
- - 1
65
- python_version: 3.8.8
66
- start_time: 1643582858
67
- t:
68
- 1:
69
- - 1
70
- - 5
71
- - 11
72
- 2:
73
- - 1
74
- - 5
75
- - 11
76
- 3:
77
- - 1
78
- - 7
79
- - 13
80
- 4: 3.8.8
81
- 5: 0.12.9
82
- 6: 4.17.0.dev0
83
- 8:
84
- - 5
85
- activation_dropout:
86
- desc: null
87
- value: 0.1
88
- adafactor:
89
- desc: null
90
- value: false
91
- adam_beta1:
92
- desc: null
93
- value: 0.9
94
- adam_beta2:
95
- desc: null
96
- value: 0.999
97
- adam_epsilon:
98
- desc: null
99
- value: 1.0e-08
100
- adapter_kernel_size:
101
- desc: null
102
- value: 3
103
- adapter_stride:
104
- desc: null
105
- value: 2
106
- add_adapter:
107
- desc: null
108
- value: false
109
- add_cross_attention:
110
- desc: null
111
- value: false
112
- apply_spec_augment:
113
- desc: null
114
- value: true
115
- architectures:
116
- desc: null
117
- value:
118
- - Wav2Vec2ForPreTraining
119
- attention_dropout:
120
- desc: null
121
- value: 0.0
122
- bad_words_ids:
123
- desc: null
124
- value: null
125
- bf16:
126
- desc: null
127
- value: false
128
- bf16_full_eval:
129
- desc: null
130
- value: false
131
- bos_token_id:
132
- desc: null
133
- value: 1
134
- chunk_size_feed_forward:
135
- desc: null
136
- value: 0
137
- classifier_proj_size:
138
- desc: null
139
- value: 256
140
- codevector_dim:
141
- desc: null
142
- value: 768
143
- contrastive_logits_temperature:
144
- desc: null
145
- value: 0.1
146
- conv_bias:
147
- desc: null
148
- value: true
149
- conv_dim:
150
- desc: null
151
- value:
152
- - 512
153
- - 512
154
- - 512
155
- - 512
156
- - 512
157
- - 512
158
- - 512
159
- conv_kernel:
160
- desc: null
161
- value:
162
- - 10
163
- - 3
164
- - 3
165
- - 3
166
- - 3
167
- - 2
168
- - 2
169
- conv_stride:
170
- desc: null
171
- value:
172
- - 5
173
- - 2
174
- - 2
175
- - 2
176
- - 2
177
- - 2
178
- - 2
179
- cross_attention_hidden_size:
180
- desc: null
181
- value: null
182
- ctc_loss_reduction:
183
- desc: null
184
- value: mean
185
- ctc_zero_infinity:
186
- desc: null
187
- value: false
188
- dataloader_drop_last:
189
- desc: null
190
- value: false
191
- dataloader_num_workers:
192
- desc: null
193
- value: 0
194
- dataloader_pin_memory:
195
- desc: null
196
- value: true
197
- ddp_bucket_cap_mb:
198
- desc: null
199
- value: None
200
- ddp_find_unused_parameters:
201
- desc: null
202
- value: None
203
- debug:
204
- desc: null
205
- value: '[]'
206
- decoder_start_token_id:
207
- desc: null
208
- value: null
209
- deepspeed:
210
- desc: null
211
- value: None
212
- disable_tqdm:
213
- desc: null
214
- value: false
215
- diversity_loss_weight:
216
- desc: null
217
- value: 0.1
218
- diversity_penalty:
219
- desc: null
220
- value: 0.0
221
- do_eval:
222
- desc: null
223
- value: true
224
- do_predict:
225
- desc: null
226
- value: false
227
- do_sample:
228
- desc: null
229
- value: false
230
- do_stable_layer_norm:
231
- desc: null
232
- value: true
233
- do_train:
234
- desc: null
235
- value: true
236
- early_stopping:
237
- desc: null
238
- value: false
239
- encoder_no_repeat_ngram_size:
240
- desc: null
241
- value: 0
242
- eos_token_id:
243
- desc: null
244
- value: 2
245
- eval_accumulation_steps:
246
- desc: null
247
- value: None
248
- eval_batch_size:
249
- desc: null
250
- value: 8
251
- eval_steps:
252
- desc: null
253
- value: 500
254
- evaluation_strategy:
255
- desc: null
256
- value: steps
257
- feat_extract_activation:
258
- desc: null
259
- value: gelu
260
- feat_extract_dropout:
261
- desc: null
262
- value: 0.0
263
- feat_extract_norm:
264
- desc: null
265
- value: layer
266
- feat_proj_dropout:
267
- desc: null
268
- value: 0.0
269
- feat_quantizer_dropout:
270
- desc: null
271
- value: 0.0
272
- final_dropout:
273
- desc: null
274
- value: 0.0
275
- finetuning_task:
276
- desc: null
277
- value: null
278
- forced_bos_token_id:
279
- desc: null
280
- value: null
281
- forced_eos_token_id:
282
- desc: null
283
- value: null
284
- fp16:
285
- desc: null
286
- value: true
287
- fp16_backend:
288
- desc: null
289
- value: auto
290
- fp16_full_eval:
291
- desc: null
292
- value: false
293
- fp16_opt_level:
294
- desc: null
295
- value: O1
296
- gradient_accumulation_steps:
297
- desc: null
298
- value: 8
299
- gradient_checkpointing:
300
- desc: null
301
- value: true
302
- greater_is_better:
303
- desc: null
304
- value: false
305
- group_by_length:
306
- desc: null
307
- value: true
308
- half_precision_backend:
309
- desc: null
310
- value: amp
311
- hidden_act:
312
- desc: null
313
- value: gelu
314
- hidden_dropout:
315
- desc: null
316
- value: 0.0
317
- hidden_size:
318
- desc: null
319
- value: 1024
320
- hub_model_id:
321
- desc: null
322
- value: None
323
- hub_strategy:
324
- desc: null
325
- value: every_save
326
- hub_token:
327
- desc: null
328
- value: <HUB_TOKEN>
329
- id2label:
330
- desc: null
331
- value:
332
- '0': LABEL_0
333
- '1': LABEL_1
334
- ignore_data_skip:
335
- desc: null
336
- value: false
337
- initializer_range:
338
- desc: null
339
- value: 0.02
340
- intermediate_size:
341
- desc: null
342
- value: 4096
343
- is_decoder:
344
- desc: null
345
- value: false
346
- is_encoder_decoder:
347
- desc: null
348
- value: false
349
- label2id:
350
- desc: null
351
- value:
352
- LABEL_0: 0
353
- LABEL_1: 1
354
- label_names:
355
- desc: null
356
- value: None
357
- label_smoothing_factor:
358
- desc: null
359
- value: 0.0
360
- layer_norm_eps:
361
- desc: null
362
- value: 1.0e-05
363
- layerdrop:
364
- desc: null
365
- value: 0.0
366
- learning_rate:
367
- desc: null
368
- value: 7.5e-05
369
- length_column_name:
370
- desc: null
371
- value: input_length
372
- length_penalty:
373
- desc: null
374
- value: 1.0
375
- load_best_model_at_end:
376
- desc: null
377
- value: true
378
- local_rank:
379
- desc: null
380
- value: -1
381
- log_level:
382
- desc: null
383
- value: -1
384
- log_level_replica:
385
- desc: null
386
- value: -1
387
- log_on_each_node:
388
- desc: null
389
- value: true
390
- logging_dir:
391
- desc: null
392
- value: ./runs/Jan30_22-46-41_job-3261699b-76eb-4c28-8419-66a66c5c9199
393
- logging_first_step:
394
- desc: null
395
- value: false
396
- logging_nan_inf_filter:
397
- desc: null
398
- value: true
399
- logging_steps:
400
- desc: null
401
- value: 100
402
- logging_strategy:
403
- desc: null
404
- value: steps
405
- lr_scheduler_type:
406
- desc: null
407
- value: linear
408
- mask_feature_length:
409
- desc: null
410
- value: 64
411
- mask_feature_min_masks:
412
- desc: null
413
- value: 0
414
- mask_feature_prob:
415
- desc: null
416
- value: 0.25
417
- mask_time_length:
418
- desc: null
419
- value: 10
420
- mask_time_min_masks:
421
- desc: null
422
- value: 2
423
- mask_time_prob:
424
- desc: null
425
- value: 0.75
426
- max_grad_norm:
427
- desc: null
428
- value: 1.0
429
- max_length:
430
- desc: null
431
- value: 20
432
- max_steps:
433
- desc: null
434
- value: -1
435
- metric_for_best_model:
436
- desc: null
437
- value: loss
438
- min_length:
439
- desc: null
440
- value: 0
441
- model_type:
442
- desc: null
443
- value: wav2vec2
444
- mp_parameters:
445
- desc: null
446
- value: ''
447
- no_cuda:
448
- desc: null
449
- value: false
450
- no_repeat_ngram_size:
451
- desc: null
452
- value: 0
453
- num_adapter_layers:
454
- desc: null
455
- value: 3
456
- num_attention_heads:
457
- desc: null
458
- value: 16
459
- num_beam_groups:
460
- desc: null
461
- value: 1
462
- num_beams:
463
- desc: null
464
- value: 1
465
- num_codevector_groups:
466
- desc: null
467
- value: 2
468
- num_codevectors_per_group:
469
- desc: null
470
- value: 320
471
- num_conv_pos_embedding_groups:
472
- desc: null
473
- value: 16
474
- num_conv_pos_embeddings:
475
- desc: null
476
- value: 128
477
- num_feat_extract_layers:
478
- desc: null
479
- value: 7
480
- num_hidden_layers:
481
- desc: null
482
- value: 24
483
- num_negatives:
484
- desc: null
485
- value: 100
486
- num_return_sequences:
487
- desc: null
488
- value: 1
489
- num_train_epochs:
490
- desc: null
491
- value: 0.2
492
- optim:
493
- desc: null
494
- value: adamw_hf
495
- output_attentions:
496
- desc: null
497
- value: false
498
- output_dir:
499
- desc: null
500
- value: ./
501
- output_hidden_size:
502
- desc: null
503
- value: 1024
504
- output_hidden_states:
505
- desc: null
506
- value: false
507
- output_scores:
508
- desc: null
509
- value: false
510
- overwrite_output_dir:
511
- desc: null
512
- value: true
513
- pad_token_id:
514
- desc: null
515
- value: 40
516
- past_index:
517
- desc: null
518
- value: -1
519
- per_device_eval_batch_size:
520
- desc: null
521
- value: 8
522
- per_device_train_batch_size:
523
- desc: null
524
- value: 8
525
- per_gpu_eval_batch_size:
526
- desc: null
527
- value: None
528
- per_gpu_train_batch_size:
529
- desc: null
530
- value: None
531
- prediction_loss_only:
532
- desc: null
533
- value: false
534
- prefix:
535
- desc: null
536
- value: null
537
- problem_type:
538
- desc: null
539
- value: null
540
- proj_codevector_dim:
541
- desc: null
542
- value: 768
543
- pruned_heads:
544
- desc: null
545
- value: {}
546
- push_to_hub:
547
- desc: null
548
- value: true
549
- push_to_hub_model_id:
550
- desc: null
551
- value: None
552
- push_to_hub_organization:
553
- desc: null
554
- value: None
555
- push_to_hub_token:
556
- desc: null
557
- value: <PUSH_TO_HUB_TOKEN>
558
- remove_invalid_values:
559
- desc: null
560
- value: false
561
- remove_unused_columns:
562
- desc: null
563
- value: true
564
- repetition_penalty:
565
- desc: null
566
- value: 1.0
567
- report_to:
568
- desc: null
569
- value: '[''wandb'']'
570
- resume_from_checkpoint:
571
- desc: null
572
- value: None
573
- return_dict:
574
- desc: null
575
- value: true
576
- return_dict_in_generate:
577
- desc: null
578
- value: false
579
- run_name:
580
- desc: null
581
- value: ./
582
- save_on_each_node:
583
- desc: null
584
- value: false
585
- save_steps:
586
- desc: null
587
- value: 500
588
- save_strategy:
589
- desc: null
590
- value: steps
591
- save_total_limit:
592
- desc: null
593
- value: 3
594
- seed:
595
- desc: null
596
- value: 42
597
- sep_token_id:
598
- desc: null
599
- value: null
600
- sharded_ddp:
601
- desc: null
602
- value: '[]'
603
- skip_memory_metrics:
604
- desc: null
605
- value: true
606
- task_specific_params:
607
- desc: null
608
- value: null
609
- tdnn_dilation:
610
- desc: null
611
- value:
612
- - 1
613
- - 2
614
- - 3
615
- - 1
616
- - 1
617
- tdnn_dim:
618
- desc: null
619
- value:
620
- - 512
621
- - 512
622
- - 512
623
- - 512
624
- - 1500
625
- tdnn_kernel:
626
- desc: null
627
- value:
628
- - 5
629
- - 3
630
- - 3
631
- - 1
632
- - 1
633
- temperature:
634
- desc: null
635
- value: 1.0
636
- tf32:
637
- desc: null
638
- value: None
639
- tie_encoder_decoder:
640
- desc: null
641
- value: false
642
- tie_word_embeddings:
643
- desc: null
644
- value: true
645
- tokenizer_class:
646
- desc: null
647
- value: null
648
- top_k:
649
- desc: null
650
- value: 50
651
- top_p:
652
- desc: null
653
- value: 1.0
654
- torch_dtype:
655
- desc: null
656
- value: float32
657
- torchscript:
658
- desc: null
659
- value: false
660
- tpu_metrics_debug:
661
- desc: null
662
- value: false
663
- tpu_num_cores:
664
- desc: null
665
- value: None
666
- train_batch_size:
667
- desc: null
668
- value: 8
669
- transformers_version:
670
- desc: null
671
- value: 4.17.0.dev0
672
- use_bfloat16:
673
- desc: null
674
- value: false
675
- use_legacy_prediction_loop:
676
- desc: null
677
- value: false
678
- use_weighted_layer_sum:
679
- desc: null
680
- value: false
681
- vocab_size:
682
- desc: null
683
- value: 41
684
- warmup_ratio:
685
- desc: null
686
- value: 0.0
687
- warmup_steps:
688
- desc: null
689
- value: 2000
690
- weight_decay:
691
- desc: null
692
- value: 0.0
693
- xpu_backend:
694
- desc: null
695
- value: None
696
- xvector_output_dim:
697
- desc: null
698
- value: 512
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
wandb/run-20220130_224738-2uzt3kt1/files/output.log DELETED
@@ -1,66 +0,0 @@
1
-
2
-
3
-
4
- 67%|██████████████████████████████████████████████████████████████████████████████████████████ | 2/3 [00:07<00:03, 3.88s/it]
5
- 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [00:10<00:00, 3.23s/it]
6
- Training completed. Do not forget to share your model on huggingface.co/models =)
7
- 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [00:10<00:00, 3.48s/it]
8
- Saving model checkpoint to ./
9
- Configuration saved in ./config.json
10
- Model weights saved in ./pytorch_model.bin
11
- Configuration saved in ./preprocessor_config.json
12
- Saving model checkpoint to ./
13
- Configuration saved in ./config.json
14
- Model weights saved in ./pytorch_model.bin
15
- Configuration saved in ./preprocessor_config.json
16
- Upload file pytorch_model.bin: 0%| | 3.39k/1.18G [00:00<?, ?B/s]
17
- Upload file training_args.bin: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████| 2.92k/2.92k [00:00<?, ?B/s]
18
- 01/30/2022 22:49:01 - WARNING - huggingface_hub.repository - To https://huggingface.co/Plim/xls-r-300m-fr
19
- 1d17287..8ac44c4 main -> main0%|█████████████████████████████████████████████████████████████████████████████████████████████████████████| 2.92k/2.92k [00:00<?, ?B/s]
20
- Upload file pytorch_model.bin: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████| 1.18G/1.18G [00:42<00:00, 29.7MB/s]
21
- Upload file training_args.bin: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████| 2.92k/2.92k [00:42<?, ?B/s]
22
- Dropping the following result as it does not have all the necessary fields:██████████████████████████████████████████████████████████████████| 2.92k/2.92k [00:42<?, ?B/s]
23
- {}
24
- 01/30/2022 22:49:07 - WARNING - huggingface_hub.repository - To https://huggingface.co/Plim/xls-r-300m-fr
25
- 8ac44c4..77260d3 main -> main
26
- To https://huggingface.co/Plim/xls-r-300m-fr
27
- 8ac44c4..77260d3 main -> main
28
- The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length.
29
- ***** Running Evaluation *****
30
- Num examples = 200
31
- Batch size = 8
32
- 0%| | 0/25 [00:00<?, ?it/s]
33
- ***** train metrics *****
34
- epoch = 0.19
35
- train_loss = 12.4969
36
- train_runtime = 0:00:12.89
37
- train_samples = 1000
38
- train_samples_per_second = 15.512
39
- train_steps_per_second = 0.233
40
-
41
-
42
-
43
-
44
- 96%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▋ | 24/25 [00:07<00:00, 2.74it/s]
45
- ***** eval metrics *****
46
- epoch = 0.19
47
- eval_loss = 16.9132
48
- eval_runtime = 0:00:08.67
49
- eval_samples = 200
50
- eval_samples_per_second = 23.067
51
- eval_steps_per_second = 2.883
52
- 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 25/25 [00:08<00:00, 3.02it/s]
53
- Saving model checkpoint to ./
54
- Configuration saved in ./config.json
55
- Model weights saved in ./pytorch_model.bin
56
- Configuration saved in ./preprocessor_config.json
57
- 01/30/2022 22:49:41 - WARNING - huggingface_hub.repository - To https://huggingface.co/Plim/xls-r-300m-fr
58
- 77260d3..45cb5d4 main -> main
59
- To https://huggingface.co/Plim/xls-r-300m-fr
60
- 77260d3..45cb5d4 main -> main
61
- Dropping the following result as it does not have all the necessary fields:
62
- {}
63
- 01/30/2022 22:49:47 - WARNING - huggingface_hub.repository - To https://huggingface.co/Plim/xls-r-300m-fr
64
- 45cb5d4..1fb68dc main -> main
65
- To https://huggingface.co/Plim/xls-r-300m-fr
66
- 45cb5d4..1fb68dc main -> main
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
wandb/run-20220130_224738-2uzt3kt1/files/requirements.txt DELETED
@@ -1,180 +0,0 @@
1
- aiohttp==3.8.1
2
- aiosignal==1.2.0
3
- analytics-python==1.4.0
4
- anyio==3.5.0
5
- appdirs==1.4.4
6
- argon2-cffi-bindings==21.2.0
7
- argon2-cffi==21.3.0
8
- asgiref==3.5.0
9
- asttokens==2.0.5
10
- async-timeout==4.0.2
11
- attrs==21.4.0
12
- audioread==2.1.9
13
- backcall==0.2.0
14
- backoff==1.10.0
15
- bcrypt==3.2.0
16
- beautifulsoup4==4.9.3
17
- black==21.12b0
18
- bleach==4.1.0
19
- brotlipy==0.7.0
20
- certifi==2020.12.5
21
- cffi==1.14.3
22
- chardet==3.0.4
23
- charset-normalizer==2.0.10
24
- click==8.0.3
25
- conda-build==3.21.4
26
- conda-package-handling==1.7.2
27
- conda==4.9.2
28
- configparser==5.2.0
29
- cryptography==3.2.1
30
- cycler==0.11.0
31
- datasets==1.18.2.dev0
32
- debugpy==1.5.1
33
- decorator==4.4.2
34
- defusedxml==0.7.1
35
- dill==0.3.4
36
- dnspython==2.1.0
37
- docker-pycreds==0.4.0
38
- entrypoints==0.3
39
- executing==0.8.2
40
- fastapi==0.73.0
41
- ffmpy==0.3.0
42
- filelock==3.0.12
43
- fonttools==4.29.0
44
- frozenlist==1.3.0
45
- fsspec==2022.1.0
46
- gitdb==4.0.9
47
- gitpython==3.1.26
48
- glob2==0.7
49
- gradio==2.7.5.2
50
- h11==0.13.0
51
- huggingface-hub==0.4.0
52
- idna==2.10
53
- importlib-resources==5.4.0
54
- ipykernel==6.7.0
55
- ipython-genutils==0.2.0
56
- ipython==8.0.1
57
- ipywidgets==7.6.3
58
- jedi==0.17.0
59
- jinja2==2.11.3
60
- jiwer==2.3.0
61
- joblib==1.1.0
62
- json5==0.9.6
63
- jsonschema==4.4.0
64
- jupyter-client==7.1.2
65
- jupyter-core==4.9.1
66
- jupyterlab-pygments==0.1.2
67
- jupyterlab-server==1.2.0
68
- jupyterlab-widgets==1.0.2
69
- jupyterlab==2.2.9
70
- kiwisolver==1.3.2
71
- libarchive-c==2.9
72
- librosa==0.8.1
73
- llvmlite==0.38.0
74
- markdown2==2.4.2
75
- markupsafe==1.1.1
76
- matplotlib-inline==0.1.3
77
- matplotlib==3.5.1
78
- mistune==0.8.4
79
- mkl-fft==1.3.0
80
- mkl-random==1.1.1
81
- mkl-service==2.3.0
82
- monotonic==1.6
83
- multidict==6.0.2
84
- multiprocess==0.70.12.2
85
- mypy-extensions==0.4.3
86
- nano==0.10.0
87
- nbclient==0.5.10
88
- nbconvert==6.4.1
89
- nbformat==5.1.3
90
- nest-asyncio==1.5.4
91
- notebook==6.4.8
92
- numba==0.55.1
93
- numpy==1.19.2
94
- olefile==0.46
95
- packaging==21.3
96
- pandas==1.4.0
97
- pandocfilters==1.5.0
98
- paramiko==2.9.2
99
- parso==0.8.1
100
- pathspec==0.9.0
101
- pathtools==0.1.2
102
- pexpect==4.8.0
103
- pickleshare==0.7.5
104
- pillow==8.1.2
105
- pip==21.3.1
106
- pkginfo==1.7.0
107
- platformdirs==2.4.1
108
- pooch==1.6.0
109
- prometheus-client==0.13.0
110
- promise==2.3
111
- prompt-toolkit==3.0.8
112
- protobuf==3.19.4
113
- psutil==5.8.0
114
- ptyprocess==0.7.0
115
- pure-eval==0.2.2
116
- pyarrow==6.0.1
117
- pycosat==0.6.3
118
- pycparser==2.20
119
- pycryptodome==3.13.0
120
- pydantic==1.9.0
121
- pydub==0.25.1
122
- pygments==2.8.0
123
- pynacl==1.5.0
124
- pyopenssl==19.1.0
125
- pyparsing==3.0.7
126
- pyrsistent==0.18.1
127
- pysocks==1.7.1
128
- python-dateutil==2.8.2
129
- python-etcd==0.4.5
130
- python-levenshtein==0.12.2
131
- python-multipart==0.0.5
132
- pytz==2021.1
133
- pyyaml==5.4.1
134
- pyzmq==22.3.0
135
- regex==2022.1.18
136
- requests==2.24.0
137
- resampy==0.2.2
138
- ruamel-yaml==0.15.87
139
- sacremoses==0.0.47
140
- scikit-learn==1.0.2
141
- scipy==1.7.3
142
- send2trash==1.8.0
143
- sentry-sdk==1.5.4
144
- setuptools==50.3.1.post20201107
145
- shortuuid==1.0.8
146
- six==1.15.0
147
- smmap==5.0.0
148
- sniffio==1.2.0
149
- soundfile==0.10.3.post1
150
- soupsieve==2.2
151
- stack-data==0.1.4
152
- starlette==0.17.1
153
- subprocess32==3.5.4
154
- termcolor==1.1.0
155
- terminado==0.13.1
156
- testpath==0.5.0
157
- threadpoolctl==3.0.0
158
- tokenizers==0.11.4
159
- tomli==1.2.3
160
- torch==1.10.2
161
- torchaudio==0.10.2
162
- torchelastic==0.2.2
163
- torchtext==0.9.1
164
- torchvision==0.9.1
165
- tornado==6.1
166
- tqdm==4.62.3
167
- traitlets==5.1.1
168
- transformers==4.17.0.dev0
169
- typing-extensions==4.0.1
170
- urllib3==1.25.11
171
- uvicorn==0.17.1
172
- wandb==0.12.9
173
- wcwidth==0.2.5
174
- webencodings==0.5.1
175
- wheel==0.35.1
176
- widgetsnbextension==3.5.2
177
- xxhash==2.0.2
178
- yarl==1.7.2
179
- yaspin==2.1.0
180
- zipp==3.7.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
wandb/run-20220130_224738-2uzt3kt1/files/wandb-metadata.json DELETED
@@ -1,63 +0,0 @@
1
- {
2
- "os": "Linux-4.15.0-151-generic-x86_64-with-glibc2.10",
3
- "python": "3.8.8",
4
- "heartbeatAt": "2022-01-30T22:47:39.607019",
5
- "startedAt": "2022-01-30T22:47:38.310593",
6
- "docker": null,
7
- "gpu": "Tesla V100S-PCIE-32GB",
8
- "gpu_count": 1,
9
- "cpu_count": 60,
10
- "cuda": null,
11
- "args": [
12
- "--activation_dropout=0.1",
13
- "--dataset_name=mozilla-foundation/common_voice_7_0",
14
- "--dataset_config_name=fr",
15
- "--eval_steps=500",
16
- "--evaluation_strategy=steps",
17
- "--feat_proj_dropout=0.0",
18
- "--freeze_feature_encoder",
19
- "--fp16",
20
- "--gradient_accumulation_steps=8",
21
- "--gradient_checkpointing",
22
- "--group_by_length",
23
- "--layerdrop=0.0",
24
- "--learning_rate=7.5e-5",
25
- "--length_column_name=input_length",
26
- "--load_best_model_at_end",
27
- "--logging_steps=100",
28
- "--mask_feature_length=64",
29
- "--mask_feature_prob=0.25",
30
- "--mask_time_length=10",
31
- "--mask_time_prob=0.75",
32
- "--max_train_samples=1000",
33
- "--max_eval_samples=200",
34
- "--model_name_or_path=facebook/wav2vec2-xls-r-300m",
35
- "--num_train_epochs=0.2",
36
- "--output_dir=./",
37
- "--overwrite_output_dir",
38
- "--per_device_train_batch_size=8",
39
- "--per_device_eval_batch_size=8",
40
- "--preprocessing_num_workers=4",
41
- "--push_to_hub",
42
- "--report_to=wandb",
43
- "--save_steps=500",
44
- "--save_total_limit=3",
45
- "--text_column_name=sentence",
46
- "--use_auth_token",
47
- "--warmup_steps=2000",
48
- "--do_train",
49
- "--do_eval"
50
- ],
51
- "state": "running",
52
- "program": "run_speech_recognition_ctc.py",
53
- "codePath": "run_speech_recognition_ctc.py",
54
- "git": {
55
- "remote": "https://huggingface.co/Plim/xls-r-300m-fr",
56
- "commit": "1d172876193bf100999c8d09d283f8d0894252f2"
57
- },
58
- "email": "lim.pascal93@gmail.com",
59
- "root": "/workspace/xls-r-300m-fr",
60
- "host": "job-3261699b-76eb-4c28-8419-66a66c5c9199",
61
- "username": "ovh",
62
- "executable": "/opt/conda/bin/python"
63
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
wandb/run-20220130_224738-2uzt3kt1/files/wandb-summary.json DELETED
@@ -1 +0,0 @@
1
- {"train/train_runtime": 12.893, "train/train_samples_per_second": 15.512, "train/train_steps_per_second": 0.233, "train/total_flos": 2.67196543170048e+16, "train/train_loss": 12.496875762939453, "train/epoch": 0.19, "train/global_step": 3, "_runtime": 100, "_timestamp": 1643582958, "_step": 1, "eval/loss": 16.913198471069336, "eval/wer": 2.3629935179728934, "eval/runtime": 8.6705, "eval/samples_per_second": 23.067, "eval/steps_per_second": 2.883, "_wandb": {"runtime": 133}}
 
wandb/run-20220130_224738-2uzt3kt1/logs/debug-internal.log DELETED