Hugging Face's logo

bkh6722
/
bach-arb

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
Model card Files
xet
 Metrics Community
bkh6722 commited on
Commit
a404b97
·
1 Parent(s): 321e144

End of training

Browse files
Files changed (11) hide show
  1. runs/{May15_01-47-58_f8d43c95907a → wav2vec-vlbg-DL}/1652579289.244684/events.out.tfevents.1652579289.f8d43c95907a.7669.1 +0 -0
  2. runs/{May15_01-47-58_f8d43c95907a → wav2vec-vlbg-DL}/events.out.tfevents.1652579289.f8d43c95907a.7669.0 +0 -0
  3. wav2vec-vlbg-DL/config.json +80 -0
  4. wav2vec-vlbg-DL/optimizer.pt +3 -0
  5. wav2vec-vlbg-DL/preprocessor_config.json +9 -0
  6. wav2vec-vlbg-DL/pytorch_model.bin +3 -0
  7. wav2vec-vlbg-DL/rng_state.pth +3 -0
  8. wav2vec-vlbg-DL/scaler.pt +3 -0
  9. wav2vec-vlbg-DL/scheduler.pt +3 -0
  10. wav2vec-vlbg-DL/trainer_state.json +256 -0
  11. wav2vec-vlbg-DL/training_args.bin +3 -0
runs/{May15_01-47-58_f8d43c95907a → wav2vec-vlbg-DL}/1652579289.244684/events.out.tfevents.1652579289.f8d43c95907a.7669.1 RENAMED
File without changes
runs/{May15_01-47-58_f8d43c95907a → wav2vec-vlbg-DL}/events.out.tfevents.1652579289.f8d43c95907a.7669.0 RENAMED
File without changes
wav2vec-vlbg-DL/config.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-base-960h",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "Wav2Vec2ForCTC"
7
+ ],
8
+ "attention_dropout": 0.5,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "codevector_dim": 256,
12
+ "contrastive_logits_temperature": 0.1,
13
+ "conv_bias": false,
14
+ "conv_dim": [
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512
22
+ ],
23
+ "conv_kernel": [
24
+ 10,
25
+ 3,
26
+ 3,
27
+ 3,
28
+ 3,
29
+ 2,
30
+ 2
31
+ ],
32
+ "conv_stride": [
33
+ 5,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2,
38
+ 2,
39
+ 2
40
+ ],
41
+ "ctc_loss_reduction": "mean",
42
+ "ctc_zero_infinity": false,
43
+ "diversity_loss_weight": 0.1,
44
+ "do_stable_layer_norm": false,
45
+ "eos_token_id": 2,
46
+ "feat_extract_activation": "gelu",
47
+ "feat_extract_dropout": 0.0,
48
+ "feat_extract_norm": "group",
49
+ "feat_proj_dropout": 0.0,
50
+ "feat_quantizer_dropout": 0.0,
51
+ "final_dropout": 0.1,
52
+ "gradient_checkpointing": false,
53
+ "hidden_act": "gelu",
54
+ "hidden_dropout": 0.0,
55
+ "hidden_dropout_prob": 0.1,
56
+ "hidden_size": 768,
57
+ "initializer_range": 0.02,
58
+ "intermediate_size": 3072,
59
+ "layer_norm_eps": 1e-05,
60
+ "layerdrop": 0.0,
61
+ "mask_feature_length": 10,
62
+ "mask_feature_prob": 0.0,
63
+ "mask_time_length": 10,
64
+ "mask_time_prob": 0.05,
65
+ "model_type": "wav2vec2",
66
+ "num_attention_heads": 12,
67
+ "num_codevector_groups": 2,
68
+ "num_codevectors_per_group": 320,
69
+ "num_conv_pos_embedding_groups": 16,
70
+ "num_conv_pos_embeddings": 128,
71
+ "num_feat_extract_layers": 7,
72
+ "num_hidden_layers": 12,
73
+ "num_negatives": 100,
74
+ "pad_token_id": 26,
75
+ "proj_codevector_dim": 256,
76
+ "torch_dtype": "float32",
77
+ "transformers_version": "4.11.3",
78
+ "use_weighted_layer_sum": false,
79
+ "vocab_size": 32
80
+ }
wav2vec-vlbg-DL/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26670ef7ae9979d2b63e7d48d5555b53ac4951fee4af78dfb24214932fea0fbe
3
+ size 721685265
wav2vec-vlbg-DL/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": true,
8
+ "sampling_rate": 16000
9
+ }
wav2vec-vlbg-DL/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b665b6cd1874671521f7efdc1efdd814f13f7cfa37a709e704d0c94d89a77c8a
3
+ size 377670039
wav2vec-vlbg-DL/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b577a860d084e7f549e0aab9ecb55d1b4a8cc6a837d423f61c7660757705370d
3
+ size 14503
wav2vec-vlbg-DL/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85064bbb7f0fe2aa0d30bacc3b089f44e411de9ebcc84ca0a6adea1cddc62b8d
3
+ size 559
wav2vec-vlbg-DL/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:501b8c1db8a1e604b70a87a72165753d4acb6f83afdd825a61f20fea16cbbb7c
3
+ size 623
wav2vec-vlbg-DL/trainer_state.json ADDED
@@ -0,0 +1,256 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 114.27586206896552,
5
+ "global_step": 1600,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 7.14,
12
+ "learning_rate": 5.6399999999999995e-05,
13
+ "loss": 21.573,
14
+ "step": 100
15
+ },
16
+ {
17
+ "epoch": 7.14,
18
+ "eval_loss": 7.376168727874756,
19
+ "eval_runtime": 2.8359,
20
+ "eval_samples_per_second": 40.904,
21
+ "eval_steps_per_second": 5.289,
22
+ "eval_wer": 1.0,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 14.28,
27
+ "learning_rate": 0.0001164,
28
+ "loss": 3.9531,
29
+ "step": 200
30
+ },
31
+ {
32
+ "epoch": 14.28,
33
+ "eval_loss": 4.114346981048584,
34
+ "eval_runtime": 2.724,
35
+ "eval_samples_per_second": 42.584,
36
+ "eval_steps_per_second": 5.507,
37
+ "eval_wer": 1.0,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 21.41,
42
+ "learning_rate": 0.00017639999999999998,
43
+ "loss": 3.2227,
44
+ "step": 300
45
+ },
46
+ {
47
+ "epoch": 21.41,
48
+ "eval_loss": 3.1481151580810547,
49
+ "eval_runtime": 2.7023,
50
+ "eval_samples_per_second": 42.926,
51
+ "eval_steps_per_second": 5.551,
52
+ "eval_wer": 1.0,
53
+ "step": 300
54
+ },
55
+ {
56
+ "epoch": 28.55,
57
+ "learning_rate": 0.0002364,
58
+ "loss": 3.1492,
59
+ "step": 400
60
+ },
61
+ {
62
+ "epoch": 28.55,
63
+ "eval_loss": 3.134784460067749,
64
+ "eval_runtime": 2.7133,
65
+ "eval_samples_per_second": 42.752,
66
+ "eval_steps_per_second": 5.528,
67
+ "eval_wer": 1.0,
68
+ "step": 400
69
+ },
70
+ {
71
+ "epoch": 35.69,
72
+ "learning_rate": 0.0002964,
73
+ "loss": 3.2487,
74
+ "step": 500
75
+ },
76
+ {
77
+ "epoch": 35.69,
78
+ "eval_loss": 3.365130662918091,
79
+ "eval_runtime": 2.7317,
80
+ "eval_samples_per_second": 42.464,
81
+ "eval_steps_per_second": 5.491,
82
+ "eval_wer": 1.0,
83
+ "step": 500
84
+ },
85
+ {
86
+ "epoch": 42.83,
87
+ "learning_rate": 0.00027459459459459456,
88
+ "loss": 3.3441,
89
+ "step": 600
90
+ },
91
+ {
92
+ "epoch": 42.83,
93
+ "eval_loss": 3.296278238296509,
94
+ "eval_runtime": 2.7086,
95
+ "eval_samples_per_second": 42.826,
96
+ "eval_steps_per_second": 5.538,
97
+ "eval_wer": 1.0,
98
+ "step": 600
99
+ },
100
+ {
101
+ "epoch": 49.97,
102
+ "learning_rate": 0.00024756756756756757,
103
+ "loss": 3.3207,
104
+ "step": 700
105
+ },
106
+ {
107
+ "epoch": 49.97,
108
+ "eval_loss": 3.2949626445770264,
109
+ "eval_runtime": 2.7298,
110
+ "eval_samples_per_second": 42.495,
111
+ "eval_steps_per_second": 5.495,
112
+ "eval_wer": 1.0,
113
+ "step": 700
114
+ },
115
+ {
116
+ "epoch": 57.14,
117
+ "learning_rate": 0.00022054054054054052,
118
+ "loss": 3.338,
119
+ "step": 800
120
+ },
121
+ {
122
+ "epoch": 57.14,
123
+ "eval_loss": 3.310981035232544,
124
+ "eval_runtime": 2.7592,
125
+ "eval_samples_per_second": 42.042,
126
+ "eval_steps_per_second": 5.436,
127
+ "eval_wer": 1.0,
128
+ "step": 800
129
+ },
130
+ {
131
+ "epoch": 64.28,
132
+ "learning_rate": 0.00019351351351351348,
133
+ "loss": 3.3142,
134
+ "step": 900
135
+ },
136
+ {
137
+ "epoch": 64.28,
138
+ "eval_loss": 3.3037407398223877,
139
+ "eval_runtime": 2.7299,
140
+ "eval_samples_per_second": 42.492,
141
+ "eval_steps_per_second": 5.495,
142
+ "eval_wer": 1.0,
143
+ "step": 900
144
+ },
145
+ {
146
+ "epoch": 71.41,
147
+ "learning_rate": 0.00016648648648648646,
148
+ "loss": 3.315,
149
+ "step": 1000
150
+ },
151
+ {
152
+ "epoch": 71.41,
153
+ "eval_loss": 3.3011434078216553,
154
+ "eval_runtime": 2.7098,
155
+ "eval_samples_per_second": 42.808,
156
+ "eval_steps_per_second": 5.536,
157
+ "eval_wer": 1.0,
158
+ "step": 1000
159
+ },
160
+ {
161
+ "epoch": 78.55,
162
+ "learning_rate": 0.00013945945945945945,
163
+ "loss": 3.306,
164
+ "step": 1100
165
+ },
166
+ {
167
+ "epoch": 78.55,
168
+ "eval_loss": 3.3198962211608887,
169
+ "eval_runtime": 2.7346,
170
+ "eval_samples_per_second": 42.42,
171
+ "eval_steps_per_second": 5.485,
172
+ "eval_wer": 1.0,
173
+ "step": 1100
174
+ },
175
+ {
176
+ "epoch": 85.69,
177
+ "learning_rate": 0.00011243243243243243,
178
+ "loss": 3.313,
179
+ "step": 1200
180
+ },
181
+ {
182
+ "epoch": 85.69,
183
+ "eval_loss": 3.299196720123291,
184
+ "eval_runtime": 2.7252,
185
+ "eval_samples_per_second": 42.566,
186
+ "eval_steps_per_second": 5.504,
187
+ "eval_wer": 1.0,
188
+ "step": 1200
189
+ },
190
+ {
191
+ "epoch": 92.83,
192
+ "learning_rate": 8.540540540540539e-05,
193
+ "loss": 3.3082,
194
+ "step": 1300
195
+ },
196
+ {
197
+ "epoch": 92.83,
198
+ "eval_loss": 3.300992488861084,
199
+ "eval_runtime": 2.7138,
200
+ "eval_samples_per_second": 42.744,
201
+ "eval_steps_per_second": 5.527,
202
+ "eval_wer": 1.0,
203
+ "step": 1300
204
+ },
205
+ {
206
+ "epoch": 99.97,
207
+ "learning_rate": 5.8378378378378376e-05,
208
+ "loss": 3.3045,
209
+ "step": 1400
210
+ },
211
+ {
212
+ "epoch": 99.97,
213
+ "eval_loss": 3.3033885955810547,
214
+ "eval_runtime": 2.7091,
215
+ "eval_samples_per_second": 42.818,
216
+ "eval_steps_per_second": 5.537,
217
+ "eval_wer": 1.0,
218
+ "step": 1400
219
+ },
220
+ {
221
+ "epoch": 107.14,
222
+ "learning_rate": 3.135135135135135e-05,
223
+ "loss": 3.317,
224
+ "step": 1500
225
+ },
226
+ {
227
+ "epoch": 107.14,
228
+ "eval_loss": 3.30355167388916,
229
+ "eval_runtime": 2.726,
230
+ "eval_samples_per_second": 42.554,
231
+ "eval_steps_per_second": 5.503,
232
+ "eval_wer": 1.0,
233
+ "step": 1500
234
+ },
235
+ {
236
+ "epoch": 114.28,
237
+ "learning_rate": 4.324324324324324e-06,
238
+ "loss": 3.3014,
239
+ "step": 1600
240
+ },
241
+ {
242
+ "epoch": 114.28,
243
+ "eval_loss": 3.303847551345825,
244
+ "eval_runtime": 2.7533,
245
+ "eval_samples_per_second": 42.131,
246
+ "eval_steps_per_second": 5.448,
247
+ "eval_wer": 1.0,
248
+ "step": 1600
249
+ }
250
+ ],
251
+ "max_steps": 1610,
252
+ "num_train_epochs": 115,
253
+ "total_flos": 1.3523871543627226e+18,
254
+ "trial_name": null,
255
+ "trial_params": null
256
+ }
wav2vec-vlbg-DL/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef4328cc21c59f95e099af016248146e2421b3db2bb6287a3c2f6de49b0e8f70
3
+ size 2799