Hugging Face's logo

eleferrand
/
Yupik_ASR

PyTorch
wav2vec2
Model card Files
xet
Community
1
eleferrand commited on
Commit
e1448e2
·
verified ·
1 Parent(s): a04f115

Upload 11 files

Browse files
Files changed (11) hide show
  1. Yupik.arpa +0 -0
  2. config.json +117 -0
  3. optimizer.pt +3 -0
  4. preprocessor_config.json +9 -0
  5. pytorch_model.bin +3 -0
  6. rng_state.pth +3 -0
  7. scheduler.pt +3 -0
  8. tokenizer_config.json +13 -0
  9. trainer_state.json +271 -0
  10. training_args.bin +3 -0
  11. vocab.json +45 -0
Yupik.arpa ADDED
The diff for this file is too large to render. See raw diff
 
config.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
3
+ "activation_dropout": 0.0,
4
+ "adapter_attn_dim": null,
5
+ "adapter_kernel_size": 3,
6
+ "adapter_stride": 2,
7
+ "add_adapter": false,
8
+ "apply_spec_augment": true,
9
+ "architectures": [
10
+ "Wav2Vec2ForCTC"
11
+ ],
12
+ "attention_dropout": 0.0,
13
+ "bos_token_id": 1,
14
+ "classifier_proj_size": 256,
15
+ "codevector_dim": 768,
16
+ "contrastive_logits_temperature": 0.1,
17
+ "conv_bias": true,
18
+ "conv_dim": [
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512,
25
+ 512
26
+ ],
27
+ "conv_kernel": [
28
+ 10,
29
+ 3,
30
+ 3,
31
+ 3,
32
+ 3,
33
+ 2,
34
+ 2
35
+ ],
36
+ "conv_stride": [
37
+ 5,
38
+ 2,
39
+ 2,
40
+ 2,
41
+ 2,
42
+ 2,
43
+ 2
44
+ ],
45
+ "ctc_loss_reduction": "mean",
46
+ "ctc_zero_infinity": true,
47
+ "diversity_loss_weight": 0.1,
48
+ "do_stable_layer_norm": true,
49
+ "eos_token_id": 2,
50
+ "feat_extract_activation": "gelu",
51
+ "feat_extract_dropout": 0.0,
52
+ "feat_extract_norm": "layer",
53
+ "feat_proj_dropout": 0.0,
54
+ "feat_quantizer_dropout": 0.0,
55
+ "final_dropout": 0.0,
56
+ "gradient_checkpointing": false,
57
+ "hidden_act": "gelu",
58
+ "hidden_dropout": 0.0,
59
+ "hidden_size": 1024,
60
+ "initializer_range": 0.02,
61
+ "intermediate_size": 4096,
62
+ "layer_norm_eps": 1e-05,
63
+ "layerdrop": 0.0,
64
+ "mask_channel_length": 10,
65
+ "mask_channel_min_space": 1,
66
+ "mask_channel_other": 0.0,
67
+ "mask_channel_prob": 0.0,
68
+ "mask_channel_selection": "static",
69
+ "mask_feature_length": 10,
70
+ "mask_feature_min_masks": 0,
71
+ "mask_feature_prob": 0.0,
72
+ "mask_time_length": 10,
73
+ "mask_time_min_masks": 2,
74
+ "mask_time_min_space": 1,
75
+ "mask_time_other": 0.0,
76
+ "mask_time_prob": 0.05,
77
+ "mask_time_selection": "static",
78
+ "model_type": "wav2vec2",
79
+ "num_adapter_layers": 3,
80
+ "num_attention_heads": 16,
81
+ "num_codevector_groups": 2,
82
+ "num_codevectors_per_group": 320,
83
+ "num_conv_pos_embedding_groups": 16,
84
+ "num_conv_pos_embeddings": 128,
85
+ "num_feat_extract_layers": 7,
86
+ "num_hidden_layers": 24,
87
+ "num_negatives": 100,
88
+ "output_hidden_size": 1024,
89
+ "pad_token_id": 45,
90
+ "proj_codevector_dim": 768,
91
+ "tdnn_dilation": [
92
+ 1,
93
+ 2,
94
+ 3,
95
+ 1,
96
+ 1
97
+ ],
98
+ "tdnn_dim": [
99
+ 512,
100
+ 512,
101
+ 512,
102
+ 512,
103
+ 1500
104
+ ],
105
+ "tdnn_kernel": [
106
+ 5,
107
+ 3,
108
+ 3,
109
+ 1,
110
+ 1
111
+ ],
112
+ "torch_dtype": "float32",
113
+ "transformers_version": "4.31.0",
114
+ "use_weighted_layer_sum": false,
115
+ "vocab_size": 46,
116
+ "xvector_output_dim": 512
117
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f136ff1e529fb21f1e8422c693d0c75b2ae856116bf087422a5d4741a059eb38
3
+ size 2524136442
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
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81676f80d2e3f56a93b6f72f36aecc9ecf35ade2b7520a9c63c1311872b54ed2
3
+ size 1262090922
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3f8d01fdd65ae1e8137da33f3ccf6be1cb597ad3b1584cd4c2a6c5503d6c57f
3
+ size 14308
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a0d4137898c094db183e9c15e57ef34629455254a589eda8badb04a77bf0cb5
3
+ size 1064
tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "clean_up_tokenization_spaces": true,
4
+ "do_lower_case": false,
5
+ "eos_token": "</s>",
6
+ "model_max_length": 1000000000000000019884624838656,
7
+ "pad_token": "[PAD]",
8
+ "replace_word_delimiter_char": " ",
9
+ "target_lang": null,
10
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
11
+ "unk_token": "[UNK]",
12
+ "word_delimiter_token": "|"
13
+ }
trainer_state.json ADDED
@@ -0,0 +1,271 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 25.12562814070352,
5
+ "global_step": 2500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.99,
12
+ "eval_loss": 2.7764203548431396,
13
+ "eval_runtime": 34.2112,
14
+ "eval_samples_per_second": 24.612,
15
+ "eval_steps_per_second": 3.098,
16
+ "eval_wer": 1.0,
17
+ "step": 99
18
+ },
19
+ {
20
+ "epoch": 2.0,
21
+ "eval_loss": 2.762573003768921,
22
+ "eval_runtime": 33.9987,
23
+ "eval_samples_per_second": 24.766,
24
+ "eval_steps_per_second": 3.118,
25
+ "eval_wer": 1.0,
26
+ "step": 199
27
+ },
28
+ {
29
+ "epoch": 2.99,
30
+ "eval_loss": 1.5485457181930542,
31
+ "eval_runtime": 34.1904,
32
+ "eval_samples_per_second": 24.627,
33
+ "eval_steps_per_second": 3.1,
34
+ "eval_wer": 0.9983126781870018,
35
+ "step": 298
36
+ },
37
+ {
38
+ "epoch": 4.0,
39
+ "eval_loss": 0.4536263048648834,
40
+ "eval_runtime": 34.3008,
41
+ "eval_samples_per_second": 24.548,
42
+ "eval_steps_per_second": 3.09,
43
+ "eval_wer": 0.6440914644789667,
44
+ "step": 398
45
+ },
46
+ {
47
+ "epoch": 4.99,
48
+ "eval_loss": 0.32973894476890564,
49
+ "eval_runtime": 34.0188,
50
+ "eval_samples_per_second": 24.751,
51
+ "eval_steps_per_second": 3.116,
52
+ "eval_wer": 0.5054983417699423,
53
+ "step": 497
54
+ },
55
+ {
56
+ "epoch": 5.03,
57
+ "learning_rate": 0.0002497979797979798,
58
+ "loss": 2.1303,
59
+ "step": 500
60
+ },
61
+ {
62
+ "epoch": 6.0,
63
+ "eval_loss": 0.27073708176612854,
64
+ "eval_runtime": 33.9387,
65
+ "eval_samples_per_second": 24.809,
66
+ "eval_steps_per_second": 3.123,
67
+ "eval_wer": 0.3777273520684238,
68
+ "step": 597
69
+ },
70
+ {
71
+ "epoch": 6.99,
72
+ "eval_loss": 0.2596016526222229,
73
+ "eval_runtime": 33.9444,
74
+ "eval_samples_per_second": 24.805,
75
+ "eval_steps_per_second": 3.123,
76
+ "eval_wer": 0.3815092802699715,
77
+ "step": 696
78
+ },
79
+ {
80
+ "epoch": 8.0,
81
+ "eval_loss": 0.25216805934906006,
82
+ "eval_runtime": 33.8461,
83
+ "eval_samples_per_second": 24.877,
84
+ "eval_steps_per_second": 3.132,
85
+ "eval_wer": 0.2978413917495782,
86
+ "step": 796
87
+ },
88
+ {
89
+ "epoch": 8.99,
90
+ "eval_loss": 0.22613081336021423,
91
+ "eval_runtime": 33.8906,
92
+ "eval_samples_per_second": 24.845,
93
+ "eval_steps_per_second": 3.128,
94
+ "eval_wer": 0.30965264444056556,
95
+ "step": 895
96
+ },
97
+ {
98
+ "epoch": 10.0,
99
+ "eval_loss": 0.23950138688087463,
100
+ "eval_runtime": 33.9858,
101
+ "eval_samples_per_second": 24.775,
102
+ "eval_steps_per_second": 3.119,
103
+ "eval_wer": 0.2646767905975447,
104
+ "step": 995
105
+ },
106
+ {
107
+ "epoch": 10.05,
108
+ "learning_rate": 0.00019929292929292926,
109
+ "loss": 0.2057,
110
+ "step": 1000
111
+ },
112
+ {
113
+ "epoch": 10.99,
114
+ "eval_loss": 0.21791674196720123,
115
+ "eval_runtime": 33.9224,
116
+ "eval_samples_per_second": 24.821,
117
+ "eval_steps_per_second": 3.125,
118
+ "eval_wer": 0.2613603304823413,
119
+ "step": 1094
120
+ },
121
+ {
122
+ "epoch": 12.0,
123
+ "eval_loss": 0.21110226213932037,
124
+ "eval_runtime": 33.8358,
125
+ "eval_samples_per_second": 24.885,
126
+ "eval_steps_per_second": 3.133,
127
+ "eval_wer": 0.25903299005061964,
128
+ "step": 1194
129
+ },
130
+ {
131
+ "epoch": 12.99,
132
+ "eval_loss": 0.22310343384742737,
133
+ "eval_runtime": 33.7767,
134
+ "eval_samples_per_second": 24.928,
135
+ "eval_steps_per_second": 3.138,
136
+ "eval_wer": 0.25368010705765986,
137
+ "step": 1293
138
+ },
139
+ {
140
+ "epoch": 14.0,
141
+ "eval_loss": 0.2454575151205063,
142
+ "eval_runtime": 33.8201,
143
+ "eval_samples_per_second": 24.896,
144
+ "eval_steps_per_second": 3.134,
145
+ "eval_wer": 0.2483854075754931,
146
+ "step": 1393
147
+ },
148
+ {
149
+ "epoch": 14.99,
150
+ "eval_loss": 0.24834834039211273,
151
+ "eval_runtime": 33.7913,
152
+ "eval_samples_per_second": 24.918,
153
+ "eval_steps_per_second": 3.137,
154
+ "eval_wer": 0.23663233839529876,
155
+ "step": 1492
156
+ },
157
+ {
158
+ "epoch": 15.08,
159
+ "learning_rate": 0.00014878787878787875,
160
+ "loss": 0.0945,
161
+ "step": 1500
162
+ },
163
+ {
164
+ "epoch": 16.0,
165
+ "eval_loss": 0.259757936000824,
166
+ "eval_runtime": 33.8499,
167
+ "eval_samples_per_second": 24.875,
168
+ "eval_steps_per_second": 3.131,
169
+ "eval_wer": 0.22348286495607145,
170
+ "step": 1592
171
+ },
172
+ {
173
+ "epoch": 16.99,
174
+ "eval_loss": 0.26361608505249023,
175
+ "eval_runtime": 33.9334,
176
+ "eval_samples_per_second": 24.813,
177
+ "eval_steps_per_second": 3.124,
178
+ "eval_wer": 0.21935183568976552,
179
+ "step": 1691
180
+ },
181
+ {
182
+ "epoch": 18.0,
183
+ "eval_loss": 0.2598336637020111,
184
+ "eval_runtime": 33.8712,
185
+ "eval_samples_per_second": 24.859,
186
+ "eval_steps_per_second": 3.13,
187
+ "eval_wer": 0.21964275324373073,
188
+ "step": 1791
189
+ },
190
+ {
191
+ "epoch": 18.99,
192
+ "eval_loss": 0.24791987240314484,
193
+ "eval_runtime": 33.8138,
194
+ "eval_samples_per_second": 24.901,
195
+ "eval_steps_per_second": 3.135,
196
+ "eval_wer": 0.20882062023622505,
197
+ "step": 1890
198
+ },
199
+ {
200
+ "epoch": 20.0,
201
+ "eval_loss": 0.24469724297523499,
202
+ "eval_runtime": 33.9448,
203
+ "eval_samples_per_second": 24.805,
204
+ "eval_steps_per_second": 3.123,
205
+ "eval_wer": 0.2091697213009833,
206
+ "step": 1990
207
+ },
208
+ {
209
+ "epoch": 20.1,
210
+ "learning_rate": 9.828282828282828e-05,
211
+ "loss": 0.0567,
212
+ "step": 2000
213
+ },
214
+ {
215
+ "epoch": 20.99,
216
+ "eval_loss": 0.25630325078964233,
217
+ "eval_runtime": 33.677,
218
+ "eval_samples_per_second": 25.002,
219
+ "eval_steps_per_second": 3.148,
220
+ "eval_wer": 0.20463140745912609,
221
+ "step": 2089
222
+ },
223
+ {
224
+ "epoch": 22.0,
225
+ "eval_loss": 0.2675672173500061,
226
+ "eval_runtime": 33.7312,
227
+ "eval_samples_per_second": 24.962,
228
+ "eval_steps_per_second": 3.142,
229
+ "eval_wer": 0.198871239890615,
230
+ "step": 2189
231
+ },
232
+ {
233
+ "epoch": 22.99,
234
+ "eval_loss": 0.25741901993751526,
235
+ "eval_runtime": 33.6669,
236
+ "eval_samples_per_second": 25.01,
237
+ "eval_steps_per_second": 3.148,
238
+ "eval_wer": 0.19951125850933846,
239
+ "step": 2288
240
+ },
241
+ {
242
+ "epoch": 24.0,
243
+ "eval_loss": 0.2615770101547241,
244
+ "eval_runtime": 33.8091,
245
+ "eval_samples_per_second": 24.905,
246
+ "eval_steps_per_second": 3.135,
247
+ "eval_wer": 0.1961947983941351,
248
+ "step": 2388
249
+ },
250
+ {
251
+ "epoch": 24.99,
252
+ "eval_loss": 0.257935494184494,
253
+ "eval_runtime": 33.8705,
254
+ "eval_samples_per_second": 24.859,
255
+ "eval_steps_per_second": 3.13,
256
+ "eval_wer": 0.1956129632862047,
257
+ "step": 2487
258
+ },
259
+ {
260
+ "epoch": 25.13,
261
+ "learning_rate": 4.777777777777778e-05,
262
+ "loss": 0.0418,
263
+ "step": 2500
264
+ }
265
+ ],
266
+ "max_steps": 2970,
267
+ "num_train_epochs": 30,
268
+ "total_flos": 6.454758338809764e+19,
269
+ "trial_name": null,
270
+ "trial_params": null
271
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11daf2ebfefc3263ad06dececf93fca76549403b3c6a9e6e868c5e1becd6fe54
3
+ size 4408
vocab.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "0": 1,
3
+ "1": 2,
4
+ "2": 3,
5
+ "3": 4,
6
+ "5": 5,
7
+ "8": 6,
8
+ "a": 7,
9
+ "b": 8,
10
+ "c": 9,
11
+ "d": 10,
12
+ "e": 11,
13
+ "f": 12,
14
+ "g": 13,
15
+ "h": 14,
16
+ "i": 15,
17
+ "j": 16,
18
+ "k": 17,
19
+ "l": 18,
20
+ "m": 19,
21
+ "n": 20,
22
+ "o": 21,
23
+ "p": 22,
24
+ "q": 23,
25
+ "r": 24,
26
+ "s": 25,
27
+ "t": 26,
28
+ "u": 27,
29
+ "v": 28,
30
+ "w": 29,
31
+ "x": 30,
32
+ "y": 31,
33
+ "z": 32,
34
+ "|": 0,
35
+ "ü": 33,
36
+ "ā": 34,
37
+ "ē": 35,
38
+ "ī": 36,
39
+ "ō": 37,
40
+ "ʼ": 38,
41
+ "‑": 39,
42
+ "‘": 40,
43
+ "’": 41,
44
+ "ꞌ": 42
45
+ }