Hugging Face's logo

wrice
/
wav2vec2-large-robust-ft-timit

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
Generated from Trainer
Model card Files
xet
 Metrics Community
1
will commited on
Commit
212c017
·
1 Parent(s): 44c441c

add tokenizer

Browse files
Files changed (21) hide show
  1. checkpoint-9000/config.json +109 -0
  2. checkpoint-9000/optimizer.pt +3 -0
  3. checkpoint-9000/preprocessor_config.json +9 -0
  4. checkpoint-9000/pytorch_model.bin +3 -0
  5. checkpoint-9000/rng_state.pth +3 -0
  6. checkpoint-9000/scaler.pt +3 -0
  7. checkpoint-9000/scheduler.pt +3 -0
  8. checkpoint-9000/trainer_state.json +286 -0
  9. checkpoint-9000/training_args.bin +3 -0
  10. checkpoint-9500/config.json +109 -0
  11. checkpoint-9500/optimizer.pt +3 -0
  12. checkpoint-9500/preprocessor_config.json +9 -0
  13. checkpoint-9500/pytorch_model.bin +3 -0
  14. checkpoint-9500/rng_state.pth +3 -0
  15. checkpoint-9500/scaler.pt +3 -0
  16. checkpoint-9500/scheduler.pt +3 -0
  17. checkpoint-9500/trainer_state.json +301 -0
  18. checkpoint-9500/training_args.bin +3 -0
  19. runs/May31_12-22-32_dylan/1654014221.6821415/events.out.tfevents.1654014221.dylan.17187.1 +3 -0
  20. runs/May31_12-22-32_dylan/events.out.tfevents.1654014221.dylan.17187.0 +3 -0
  21. vocab.json +1 -1
checkpoint-9000/config.json ADDED
@@ -0,0 +1,109 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-large-robust",
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": 0.1,
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": 0.1,
53
+ "feat_quantizer_dropout": 0.0,
54
+ "final_dropout": 0.1,
55
+ "gradient_checkpointing": false,
56
+ "hidden_act": "gelu",
57
+ "hidden_dropout": 0.1,
58
+ "hidden_dropout_prob": 0.1,
59
+ "hidden_size": 1024,
60
+ "initializer_range": 0.02,
61
+ "intermediate_size": 4096,
62
+ "layer_norm_eps": 1e-05,
63
+ "layerdrop": 0.1,
64
+ "mask_feature_length": 10,
65
+ "mask_feature_min_masks": 0,
66
+ "mask_feature_prob": 0.0,
67
+ "mask_time_length": 10,
68
+ "mask_time_min_masks": 2,
69
+ "mask_time_prob": 0.05,
70
+ "model_type": "wav2vec2",
71
+ "num_adapter_layers": 3,
72
+ "num_attention_heads": 16,
73
+ "num_codevector_groups": 2,
74
+ "num_codevectors_per_group": 320,
75
+ "num_conv_pos_embedding_groups": 16,
76
+ "num_conv_pos_embeddings": 128,
77
+ "num_feat_extract_layers": 7,
78
+ "num_hidden_layers": 24,
79
+ "num_negatives": 100,
80
+ "output_hidden_size": 1024,
81
+ "pad_token_id": 62,
82
+ "proj_codevector_dim": 768,
83
+ "tdnn_dilation": [
84
+ 1,
85
+ 2,
86
+ 3,
87
+ 1,
88
+ 1
89
+ ],
90
+ "tdnn_dim": [
91
+ 512,
92
+ 512,
93
+ 512,
94
+ 512,
95
+ 1500
96
+ ],
97
+ "tdnn_kernel": [
98
+ 5,
99
+ 3,
100
+ 3,
101
+ 1,
102
+ 1
103
+ ],
104
+ "torch_dtype": "float32",
105
+ "transformers_version": "4.19.2",
106
+ "use_weighted_layer_sum": false,
107
+ "vocab_size": 63,
108
+ "xvector_output_dim": 512
109
+ }
checkpoint-9000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86e14d6fd19aad484a7e0d9212e47651fe297f5f7ee23654d2e1bdb477e5f1b8
3
+ size 2490593799
checkpoint-9000/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": false,
8
+ "sampling_rate": 16000
9
+ }
checkpoint-9000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:346448f5f392e81d59d003633950b0eaaca89d725201889963202b2743182857
3
+ size 1262167127
checkpoint-9000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a341c32fb170d35b0bc440b8083a0ec6826b3e10b99a32c48885a5aef5f9bdb
3
+ size 14657
checkpoint-9000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b176226b6b98ae78a063f4ed489e4f139c32bc0eff90ab8a62a0cfc8b518c82
3
+ size 559
checkpoint-9000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f58021fb10a5cb473d60e41d943112cb4c8c585adc16a920f223a8d441ed7810
3
+ size 623
checkpoint-9000/trainer_state.json ADDED
@@ -0,0 +1,286 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 18.072289156626507,
5
+ "global_step": 9000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 1.0,
12
+ "learning_rate": 4.96e-05,
13
+ "loss": 6.6105,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 1.0,
18
+ "eval_loss": 3.314295530319214,
19
+ "eval_runtime": 52.8824,
20
+ "eval_samples_per_second": 31.769,
21
+ "eval_steps_per_second": 3.971,
22
+ "eval_wer": 1.0,
23
+ "step": 500
24
+ },
25
+ {
26
+ "epoch": 2.01,
27
+ "learning_rate": 9.960000000000001e-05,
28
+ "loss": 3.0691,
29
+ "step": 1000
30
+ },
31
+ {
32
+ "epoch": 2.01,
33
+ "eval_loss": 2.9860787391662598,
34
+ "eval_runtime": 53.2337,
35
+ "eval_samples_per_second": 31.559,
36
+ "eval_steps_per_second": 3.945,
37
+ "eval_wer": 1.0,
38
+ "step": 1000
39
+ },
40
+ {
41
+ "epoch": 3.01,
42
+ "learning_rate": 9.644189383070302e-05,
43
+ "loss": 2.4126,
44
+ "step": 1500
45
+ },
46
+ {
47
+ "epoch": 3.01,
48
+ "eval_loss": 0.8802480101585388,
49
+ "eval_runtime": 53.6638,
50
+ "eval_samples_per_second": 31.306,
51
+ "eval_steps_per_second": 3.913,
52
+ "eval_wer": 0.7785507645681223,
53
+ "step": 1500
54
+ },
55
+ {
56
+ "epoch": 4.02,
57
+ "learning_rate": 9.285509325681493e-05,
58
+ "loss": 0.8634,
59
+ "step": 2000
60
+ },
61
+ {
62
+ "epoch": 4.02,
63
+ "eval_loss": 0.4688626825809479,
64
+ "eval_runtime": 53.6829,
65
+ "eval_samples_per_second": 31.295,
66
+ "eval_steps_per_second": 3.912,
67
+ "eval_wer": 0.4637691142030583,
68
+ "step": 2000
69
+ },
70
+ {
71
+ "epoch": 5.02,
72
+ "learning_rate": 8.926829268292683e-05,
73
+ "loss": 0.5694,
74
+ "step": 2500
75
+ },
76
+ {
77
+ "epoch": 5.02,
78
+ "eval_loss": 0.3807709217071533,
79
+ "eval_runtime": 53.5215,
80
+ "eval_samples_per_second": 31.389,
81
+ "eval_steps_per_second": 3.924,
82
+ "eval_wer": 0.4057721449235432,
83
+ "step": 2500
84
+ },
85
+ {
86
+ "epoch": 6.02,
87
+ "learning_rate": 8.568149210903874e-05,
88
+ "loss": 0.4495,
89
+ "step": 3000
90
+ },
91
+ {
92
+ "epoch": 6.02,
93
+ "eval_loss": 0.3306347131729126,
94
+ "eval_runtime": 53.7192,
95
+ "eval_samples_per_second": 31.274,
96
+ "eval_steps_per_second": 3.909,
97
+ "eval_wer": 0.3470863755338201,
98
+ "step": 3000
99
+ },
100
+ {
101
+ "epoch": 7.03,
102
+ "learning_rate": 8.209469153515065e-05,
103
+ "loss": 0.3642,
104
+ "step": 3500
105
+ },
106
+ {
107
+ "epoch": 7.03,
108
+ "eval_loss": 0.2894286811351776,
109
+ "eval_runtime": 53.8042,
110
+ "eval_samples_per_second": 31.224,
111
+ "eval_steps_per_second": 3.903,
112
+ "eval_wer": 0.3156770905083345,
113
+ "step": 3500
114
+ },
115
+ {
116
+ "epoch": 8.03,
117
+ "learning_rate": 7.850789096126256e-05,
118
+ "loss": 0.3088,
119
+ "step": 4000
120
+ },
121
+ {
122
+ "epoch": 8.03,
123
+ "eval_loss": 0.27430233359336853,
124
+ "eval_runtime": 53.8551,
125
+ "eval_samples_per_second": 31.195,
126
+ "eval_steps_per_second": 3.899,
127
+ "eval_wer": 0.2921201267392203,
128
+ "step": 4000
129
+ },
130
+ {
131
+ "epoch": 9.04,
132
+ "learning_rate": 7.492109038737447e-05,
133
+ "loss": 0.2622,
134
+ "step": 4500
135
+ },
136
+ {
137
+ "epoch": 9.04,
138
+ "eval_loss": 0.27223920822143555,
139
+ "eval_runtime": 53.8927,
140
+ "eval_samples_per_second": 31.173,
141
+ "eval_steps_per_second": 3.897,
142
+ "eval_wer": 0.2808238049318088,
143
+ "step": 4500
144
+ },
145
+ {
146
+ "epoch": 10.04,
147
+ "learning_rate": 7.133428981348638e-05,
148
+ "loss": 0.2408,
149
+ "step": 5000
150
+ },
151
+ {
152
+ "epoch": 10.04,
153
+ "eval_loss": 0.2509672939777374,
154
+ "eval_runtime": 54.0435,
155
+ "eval_samples_per_second": 31.086,
156
+ "eval_steps_per_second": 3.886,
157
+ "eval_wer": 0.2630527620884419,
158
+ "step": 5000
159
+ },
160
+ {
161
+ "epoch": 11.04,
162
+ "learning_rate": 6.774748923959828e-05,
163
+ "loss": 0.2156,
164
+ "step": 5500
165
+ },
166
+ {
167
+ "epoch": 11.04,
168
+ "eval_loss": 0.25069141387939453,
169
+ "eval_runtime": 54.2035,
170
+ "eval_samples_per_second": 30.994,
171
+ "eval_steps_per_second": 3.874,
172
+ "eval_wer": 0.25678468108554897,
173
+ "step": 5500
174
+ },
175
+ {
176
+ "epoch": 12.05,
177
+ "learning_rate": 6.416068866571019e-05,
178
+ "loss": 0.1796,
179
+ "step": 6000
180
+ },
181
+ {
182
+ "epoch": 12.05,
183
+ "eval_loss": 0.2650609612464905,
184
+ "eval_runtime": 54.1749,
185
+ "eval_samples_per_second": 31.011,
186
+ "eval_steps_per_second": 3.876,
187
+ "eval_wer": 0.2605730816916931,
188
+ "step": 6000
189
+ },
190
+ {
191
+ "epoch": 13.05,
192
+ "learning_rate": 6.05738880918221e-05,
193
+ "loss": 0.1621,
194
+ "step": 6500
195
+ },
196
+ {
197
+ "epoch": 13.05,
198
+ "eval_loss": 0.2567823827266693,
199
+ "eval_runtime": 54.068,
200
+ "eval_samples_per_second": 31.072,
201
+ "eval_steps_per_second": 3.884,
202
+ "eval_wer": 0.25299628047940487,
203
+ "step": 6500
204
+ },
205
+ {
206
+ "epoch": 14.06,
207
+ "learning_rate": 5.6987087517934e-05,
208
+ "loss": 0.157,
209
+ "step": 7000
210
+ },
211
+ {
212
+ "epoch": 14.06,
213
+ "eval_loss": 0.2659182548522949,
214
+ "eval_runtime": 54.2391,
215
+ "eval_samples_per_second": 30.974,
216
+ "eval_steps_per_second": 3.872,
217
+ "eval_wer": 0.24707259953161592,
218
+ "step": 7000
219
+ },
220
+ {
221
+ "epoch": 15.06,
222
+ "learning_rate": 5.340746054519369e-05,
223
+ "loss": 0.1427,
224
+ "step": 7500
225
+ },
226
+ {
227
+ "epoch": 15.06,
228
+ "eval_loss": 0.257994145154953,
229
+ "eval_runtime": 54.005,
230
+ "eval_samples_per_second": 31.108,
231
+ "eval_steps_per_second": 3.889,
232
+ "eval_wer": 0.2492078798732608,
233
+ "step": 7500
234
+ },
235
+ {
236
+ "epoch": 16.06,
237
+ "learning_rate": 4.98206599713056e-05,
238
+ "loss": 0.1333,
239
+ "step": 8000
240
+ },
241
+ {
242
+ "epoch": 16.06,
243
+ "eval_loss": 0.258346825838089,
244
+ "eval_runtime": 54.2656,
245
+ "eval_samples_per_second": 30.959,
246
+ "eval_steps_per_second": 3.87,
247
+ "eval_wer": 0.24390411902465906,
248
+ "step": 8000
249
+ },
250
+ {
251
+ "epoch": 17.07,
252
+ "learning_rate": 4.6233859397417504e-05,
253
+ "loss": 0.1329,
254
+ "step": 8500
255
+ },
256
+ {
257
+ "epoch": 17.07,
258
+ "eval_loss": 0.2447778582572937,
259
+ "eval_runtime": 54.2728,
260
+ "eval_samples_per_second": 30.955,
261
+ "eval_steps_per_second": 3.869,
262
+ "eval_wer": 0.24411075905772145,
263
+ "step": 8500
264
+ },
265
+ {
266
+ "epoch": 18.07,
267
+ "learning_rate": 4.2647058823529415e-05,
268
+ "loss": 0.1208,
269
+ "step": 9000
270
+ },
271
+ {
272
+ "epoch": 18.07,
273
+ "eval_loss": 0.2728850245475769,
274
+ "eval_runtime": 54.156,
275
+ "eval_samples_per_second": 31.022,
276
+ "eval_steps_per_second": 3.878,
277
+ "eval_wer": 0.24321531891445103,
278
+ "step": 9000
279
+ }
280
+ ],
281
+ "max_steps": 14940,
282
+ "num_train_epochs": 30,
283
+ "total_flos": 6.196911507727277e+18,
284
+ "trial_name": null,
285
+ "trial_params": null
286
+ }
checkpoint-9000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34346bdfec37881c161519ed6a92e1e3029fd9a29dc35e37cc06364b169d3b65
3
+ size 3247
checkpoint-9500/config.json ADDED
@@ -0,0 +1,109 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-large-robust",
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": 0.1,
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": 0.1,
53
+ "feat_quantizer_dropout": 0.0,
54
+ "final_dropout": 0.1,
55
+ "gradient_checkpointing": false,
56
+ "hidden_act": "gelu",
57
+ "hidden_dropout": 0.1,
58
+ "hidden_dropout_prob": 0.1,
59
+ "hidden_size": 1024,
60
+ "initializer_range": 0.02,
61
+ "intermediate_size": 4096,
62
+ "layer_norm_eps": 1e-05,
63
+ "layerdrop": 0.1,
64
+ "mask_feature_length": 10,
65
+ "mask_feature_min_masks": 0,
66
+ "mask_feature_prob": 0.0,
67
+ "mask_time_length": 10,
68
+ "mask_time_min_masks": 2,
69
+ "mask_time_prob": 0.05,
70
+ "model_type": "wav2vec2",
71
+ "num_adapter_layers": 3,
72
+ "num_attention_heads": 16,
73
+ "num_codevector_groups": 2,
74
+ "num_codevectors_per_group": 320,
75
+ "num_conv_pos_embedding_groups": 16,
76
+ "num_conv_pos_embeddings": 128,
77
+ "num_feat_extract_layers": 7,
78
+ "num_hidden_layers": 24,
79
+ "num_negatives": 100,
80
+ "output_hidden_size": 1024,
81
+ "pad_token_id": 62,
82
+ "proj_codevector_dim": 768,
83
+ "tdnn_dilation": [
84
+ 1,
85
+ 2,
86
+ 3,
87
+ 1,
88
+ 1
89
+ ],
90
+ "tdnn_dim": [
91
+ 512,
92
+ 512,
93
+ 512,
94
+ 512,
95
+ 1500
96
+ ],
97
+ "tdnn_kernel": [
98
+ 5,
99
+ 3,
100
+ 3,
101
+ 1,
102
+ 1
103
+ ],
104
+ "torch_dtype": "float32",
105
+ "transformers_version": "4.19.2",
106
+ "use_weighted_layer_sum": false,
107
+ "vocab_size": 63,
108
+ "xvector_output_dim": 512
109
+ }
checkpoint-9500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2c0bc9a7b05dd5736d5da90998063f1a90de0a0230f9f24e2fc1680720eeceb
3
+ size 2490593799
checkpoint-9500/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": false,
8
+ "sampling_rate": 16000
9
+ }
checkpoint-9500/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9780af54a189d2a6c97bf2f8bb1bcf68b4d6275b1be0f0996fad18c192e78422
3
+ size 1262167127
checkpoint-9500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2341c2bf215f93c23da99a537f5dcbb4248ae8c13908c65abb81eb37d0d14d1
3
+ size 14657
checkpoint-9500/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81b6f67580029562af36f68e07cc85b4f244b6560fce0bd9ffe531684d1cf07b
3
+ size 559
checkpoint-9500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a60bef062df58c031db84acebb46e2a7cc1c252578654ade390c41a4dd16aa6e
3
+ size 623
checkpoint-9500/trainer_state.json ADDED
@@ -0,0 +1,301 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 19.076305220883533,
5
+ "global_step": 9500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 1.0,
12
+ "learning_rate": 4.96e-05,
13
+ "loss": 6.6105,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 1.0,
18
+ "eval_loss": 3.314295530319214,
19
+ "eval_runtime": 52.8824,
20
+ "eval_samples_per_second": 31.769,
21
+ "eval_steps_per_second": 3.971,
22
+ "eval_wer": 1.0,
23
+ "step": 500
24
+ },
25
+ {
26
+ "epoch": 2.01,
27
+ "learning_rate": 9.960000000000001e-05,
28
+ "loss": 3.0691,
29
+ "step": 1000
30
+ },
31
+ {
32
+ "epoch": 2.01,
33
+ "eval_loss": 2.9860787391662598,
34
+ "eval_runtime": 53.2337,
35
+ "eval_samples_per_second": 31.559,
36
+ "eval_steps_per_second": 3.945,
37
+ "eval_wer": 1.0,
38
+ "step": 1000
39
+ },
40
+ {
41
+ "epoch": 3.01,
42
+ "learning_rate": 9.644189383070302e-05,
43
+ "loss": 2.4126,
44
+ "step": 1500
45
+ },
46
+ {
47
+ "epoch": 3.01,
48
+ "eval_loss": 0.8802480101585388,
49
+ "eval_runtime": 53.6638,
50
+ "eval_samples_per_second": 31.306,
51
+ "eval_steps_per_second": 3.913,
52
+ "eval_wer": 0.7785507645681223,
53
+ "step": 1500
54
+ },
55
+ {
56
+ "epoch": 4.02,
57
+ "learning_rate": 9.285509325681493e-05,
58
+ "loss": 0.8634,
59
+ "step": 2000
60
+ },
61
+ {
62
+ "epoch": 4.02,
63
+ "eval_loss": 0.4688626825809479,
64
+ "eval_runtime": 53.6829,
65
+ "eval_samples_per_second": 31.295,
66
+ "eval_steps_per_second": 3.912,
67
+ "eval_wer": 0.4637691142030583,
68
+ "step": 2000
69
+ },
70
+ {
71
+ "epoch": 5.02,
72
+ "learning_rate": 8.926829268292683e-05,
73
+ "loss": 0.5694,
74
+ "step": 2500
75
+ },
76
+ {
77
+ "epoch": 5.02,
78
+ "eval_loss": 0.3807709217071533,
79
+ "eval_runtime": 53.5215,
80
+ "eval_samples_per_second": 31.389,
81
+ "eval_steps_per_second": 3.924,
82
+ "eval_wer": 0.4057721449235432,
83
+ "step": 2500
84
+ },
85
+ {
86
+ "epoch": 6.02,
87
+ "learning_rate": 8.568149210903874e-05,
88
+ "loss": 0.4495,
89
+ "step": 3000
90
+ },
91
+ {
92
+ "epoch": 6.02,
93
+ "eval_loss": 0.3306347131729126,
94
+ "eval_runtime": 53.7192,
95
+ "eval_samples_per_second": 31.274,
96
+ "eval_steps_per_second": 3.909,
97
+ "eval_wer": 0.3470863755338201,
98
+ "step": 3000
99
+ },
100
+ {
101
+ "epoch": 7.03,
102
+ "learning_rate": 8.209469153515065e-05,
103
+ "loss": 0.3642,
104
+ "step": 3500
105
+ },
106
+ {
107
+ "epoch": 7.03,
108
+ "eval_loss": 0.2894286811351776,
109
+ "eval_runtime": 53.8042,
110
+ "eval_samples_per_second": 31.224,
111
+ "eval_steps_per_second": 3.903,
112
+ "eval_wer": 0.3156770905083345,
113
+ "step": 3500
114
+ },
115
+ {
116
+ "epoch": 8.03,
117
+ "learning_rate": 7.850789096126256e-05,
118
+ "loss": 0.3088,
119
+ "step": 4000
120
+ },
121
+ {
122
+ "epoch": 8.03,
123
+ "eval_loss": 0.27430233359336853,
124
+ "eval_runtime": 53.8551,
125
+ "eval_samples_per_second": 31.195,
126
+ "eval_steps_per_second": 3.899,
127
+ "eval_wer": 0.2921201267392203,
128
+ "step": 4000
129
+ },
130
+ {
131
+ "epoch": 9.04,
132
+ "learning_rate": 7.492109038737447e-05,
133
+ "loss": 0.2622,
134
+ "step": 4500
135
+ },
136
+ {
137
+ "epoch": 9.04,
138
+ "eval_loss": 0.27223920822143555,
139
+ "eval_runtime": 53.8927,
140
+ "eval_samples_per_second": 31.173,
141
+ "eval_steps_per_second": 3.897,
142
+ "eval_wer": 0.2808238049318088,
143
+ "step": 4500
144
+ },
145
+ {
146
+ "epoch": 10.04,
147
+ "learning_rate": 7.133428981348638e-05,
148
+ "loss": 0.2408,
149
+ "step": 5000
150
+ },
151
+ {
152
+ "epoch": 10.04,
153
+ "eval_loss": 0.2509672939777374,
154
+ "eval_runtime": 54.0435,
155
+ "eval_samples_per_second": 31.086,
156
+ "eval_steps_per_second": 3.886,
157
+ "eval_wer": 0.2630527620884419,
158
+ "step": 5000
159
+ },
160
+ {
161
+ "epoch": 11.04,
162
+ "learning_rate": 6.774748923959828e-05,
163
+ "loss": 0.2156,
164
+ "step": 5500
165
+ },
166
+ {
167
+ "epoch": 11.04,
168
+ "eval_loss": 0.25069141387939453,
169
+ "eval_runtime": 54.2035,
170
+ "eval_samples_per_second": 30.994,
171
+ "eval_steps_per_second": 3.874,
172
+ "eval_wer": 0.25678468108554897,
173
+ "step": 5500
174
+ },
175
+ {
176
+ "epoch": 12.05,
177
+ "learning_rate": 6.416068866571019e-05,
178
+ "loss": 0.1796,
179
+ "step": 6000
180
+ },
181
+ {
182
+ "epoch": 12.05,
183
+ "eval_loss": 0.2650609612464905,
184
+ "eval_runtime": 54.1749,
185
+ "eval_samples_per_second": 31.011,
186
+ "eval_steps_per_second": 3.876,
187
+ "eval_wer": 0.2605730816916931,
188
+ "step": 6000
189
+ },
190
+ {
191
+ "epoch": 13.05,
192
+ "learning_rate": 6.05738880918221e-05,
193
+ "loss": 0.1621,
194
+ "step": 6500
195
+ },
196
+ {
197
+ "epoch": 13.05,
198
+ "eval_loss": 0.2567823827266693,
199
+ "eval_runtime": 54.068,
200
+ "eval_samples_per_second": 31.072,
201
+ "eval_steps_per_second": 3.884,
202
+ "eval_wer": 0.25299628047940487,
203
+ "step": 6500
204
+ },
205
+ {
206
+ "epoch": 14.06,
207
+ "learning_rate": 5.6987087517934e-05,
208
+ "loss": 0.157,
209
+ "step": 7000
210
+ },
211
+ {
212
+ "epoch": 14.06,
213
+ "eval_loss": 0.2659182548522949,
214
+ "eval_runtime": 54.2391,
215
+ "eval_samples_per_second": 30.974,
216
+ "eval_steps_per_second": 3.872,
217
+ "eval_wer": 0.24707259953161592,
218
+ "step": 7000
219
+ },
220
+ {
221
+ "epoch": 15.06,
222
+ "learning_rate": 5.340746054519369e-05,
223
+ "loss": 0.1427,
224
+ "step": 7500
225
+ },
226
+ {
227
+ "epoch": 15.06,
228
+ "eval_loss": 0.257994145154953,
229
+ "eval_runtime": 54.005,
230
+ "eval_samples_per_second": 31.108,
231
+ "eval_steps_per_second": 3.889,
232
+ "eval_wer": 0.2492078798732608,
233
+ "step": 7500
234
+ },
235
+ {
236
+ "epoch": 16.06,
237
+ "learning_rate": 4.98206599713056e-05,
238
+ "loss": 0.1333,
239
+ "step": 8000
240
+ },
241
+ {
242
+ "epoch": 16.06,
243
+ "eval_loss": 0.258346825838089,
244
+ "eval_runtime": 54.2656,
245
+ "eval_samples_per_second": 30.959,
246
+ "eval_steps_per_second": 3.87,
247
+ "eval_wer": 0.24390411902465906,
248
+ "step": 8000
249
+ },
250
+ {
251
+ "epoch": 17.07,
252
+ "learning_rate": 4.6233859397417504e-05,
253
+ "loss": 0.1329,
254
+ "step": 8500
255
+ },
256
+ {
257
+ "epoch": 17.07,
258
+ "eval_loss": 0.2447778582572937,
259
+ "eval_runtime": 54.2728,
260
+ "eval_samples_per_second": 30.955,
261
+ "eval_steps_per_second": 3.869,
262
+ "eval_wer": 0.24411075905772145,
263
+ "step": 8500
264
+ },
265
+ {
266
+ "epoch": 18.07,
267
+ "learning_rate": 4.2647058823529415e-05,
268
+ "loss": 0.1208,
269
+ "step": 9000
270
+ },
271
+ {
272
+ "epoch": 18.07,
273
+ "eval_loss": 0.2728850245475769,
274
+ "eval_runtime": 54.156,
275
+ "eval_samples_per_second": 31.022,
276
+ "eval_steps_per_second": 3.878,
277
+ "eval_wer": 0.24321531891445103,
278
+ "step": 9000
279
+ },
280
+ {
281
+ "epoch": 19.08,
282
+ "learning_rate": 3.9060258249641326e-05,
283
+ "loss": 0.1153,
284
+ "step": 9500
285
+ },
286
+ {
287
+ "epoch": 19.08,
288
+ "eval_loss": 0.2608482241630554,
289
+ "eval_runtime": 54.3748,
290
+ "eval_samples_per_second": 30.897,
291
+ "eval_steps_per_second": 3.862,
292
+ "eval_wer": 0.242526518804243,
293
+ "step": 9500
294
+ }
295
+ ],
296
+ "max_steps": 14940,
297
+ "num_train_epochs": 30,
298
+ "total_flos": 6.541015365721004e+18,
299
+ "trial_name": null,
300
+ "trial_params": null
301
+ }
checkpoint-9500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34346bdfec37881c161519ed6a92e1e3029fd9a29dc35e37cc06364b169d3b65
3
+ size 3247
runs/May31_12-22-32_dylan/1654014221.6821415/events.out.tfevents.1654014221.dylan.17187.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99e82f7ac09b0c8e3af6ca52683b9f5582e0736801c24ad3702e5883daeb2145
3
+ size 5206
runs/May31_12-22-32_dylan/events.out.tfevents.1654014221.dylan.17187.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:868d576a249c600d5fe99032c5ff341c8921b1aa999e4fd11b5ec1d4ecd75dc1
3
+ size 14039
vocab.json CHANGED
@@ -1 +1 @@
1
- {"M": 0, "'": 1, "Q": 2, "g": 3, "j": 4, ",": 5, "\"": 6, "a": 7, "H": 8, "-": 9, "m": 10, "T": 11, "q": 12, "k": 13, "F": 14, "o": 15, "A": 16, "z": 17, "r": 18, "l": 19, "G": 20, "U": 21, ".": 22, "w": 23, "Y": 24, "R": 25, "S": 26, "N": 27, "i": 28, "B": 29, "u": 30, ";": 32, "x": 33, "I": 34, "t": 35, "K": 36, "J": 37, "!": 38, "D": 39, "y": 40, "h": 41, "f": 42, "X": 43, "d": 44, "s": 45, "C": 46, "p": 47, "E": 48, "n": 49, ":": 50, "v": 51, "O": 52, "V": 53, "b": 54, "c": 55, "W": 56, "?": 57, "e": 58, "P": 59, "L": 60, "|": 31, "[UNK]": 61, "[PAD]": 62}
 
1
+ {"y": 0, "s": 1, "?": 2, "m": 3, "S": 4, "x": 5, "I": 6, "E": 7, "l": 8, "T": 9, "R": 10, ":": 11, "g": 12, "v": 13, "K": 14, "r": 15, ",": 16, "n": 17, ";": 18, "!": 19, "X": 20, "H": 21, "j": 22, ".": 23, "h": 24, "i": 25, "u": 26, "a": 27, "P": 28, "w": 29, "N": 30, "'": 31, "L": 32, "O": 33, "W": 34, "C": 35, "U": 36, "z": 37, "V": 38, "A": 39, "t": 40, "J": 41, "Q": 42, "o": 43, "e": 45, "d": 46, "Y": 47, "\"": 48, "b": 49, "f": 50, "k": 51, "-": 52, "q": 53, "F": 54, "p": 55, "B": 56, "D": 57, "c": 58, "G": 59, "M": 60, "|": 44, "[UNK]": 61, "[PAD]": 62}