Hugging Face's logo

sirajul116
/
wav2vec2-base-finetuned-digit

Audio Classification
Transformers
TensorBoard
Safetensors
wav2vec2
Inference Endpoints
Model card Files Metrics Community
sirajul116 commited on
Commit
5dedb4e
•
1 Parent(s): bfad258

Training in progress, epoch 1

Browse files
Files changed (5) hide show
  1. config.json +324 -0
  2. model.safetensors +3 -0
  3. preprocessor_config.json +9 -0
  4. runs/May20_11-43-51_7cb4956f5d3d/events.out.tfevents.1716205447.7cb4956f5d3d.34.1 +3 -0
  5. training_args.bin +3 -0
config.json ADDED
@@ -0,0 +1,324 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-base",
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
+ "Wav2Vec2ForSequenceClassification"
11
+ ],
12
+ "attention_dropout": 0.1,
13
+ "bos_token_id": 1,
14
+ "classifier_proj_size": 256,
15
+ "codevector_dim": 256,
16
+ "contrastive_logits_temperature": 0.1,
17
+ "conv_bias": false,
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": "sum",
46
+ "ctc_zero_infinity": false,
47
+ "diversity_loss_weight": 0.1,
48
+ "do_stable_layer_norm": false,
49
+ "eos_token_id": 2,
50
+ "feat_extract_activation": "gelu",
51
+ "feat_extract_norm": "group",
52
+ "feat_proj_dropout": 0.1,
53
+ "feat_quantizer_dropout": 0.0,
54
+ "final_dropout": 0.0,
55
+ "freeze_feat_extract_train": true,
56
+ "hidden_act": "gelu",
57
+ "hidden_dropout": 0.1,
58
+ "hidden_size": 768,
59
+ "id2label": {
60
+ "0": 0,
61
+ "1": 1,
62
+ "2": 2,
63
+ "3": 3,
64
+ "4": 4,
65
+ "5": 5,
66
+ "6": 6,
67
+ "7": 7,
68
+ "8": 8,
69
+ "9": 9,
70
+ "10": 10,
71
+ "11": 11,
72
+ "12": 12,
73
+ "13": 13,
74
+ "14": 14,
75
+ "15": 15,
76
+ "16": 16,
77
+ "17": 17,
78
+ "18": 18,
79
+ "19": 19,
80
+ "20": 20,
81
+ "21": 21,
82
+ "22": 22,
83
+ "23": 23,
84
+ "24": 24,
85
+ "25": 25,
86
+ "26": 26,
87
+ "27": 27,
88
+ "28": 28,
89
+ "29": 29,
90
+ "30": 30,
91
+ "31": 31,
92
+ "32": 32,
93
+ "33": 33,
94
+ "34": 34,
95
+ "35": 35,
96
+ "36": 36,
97
+ "37": 37,
98
+ "38": 38,
99
+ "39": 39,
100
+ "40": 40,
101
+ "41": 41,
102
+ "42": 42,
103
+ "43": 43,
104
+ "44": 44,
105
+ "45": 45,
106
+ "46": 46,
107
+ "47": 47,
108
+ "48": 48,
109
+ "49": 49,
110
+ "50": 50,
111
+ "51": 51,
112
+ "52": 52,
113
+ "53": 53,
114
+ "54": 54,
115
+ "55": 55,
116
+ "56": 56,
117
+ "57": 57,
118
+ "58": 58,
119
+ "59": 59,
120
+ "60": 60,
121
+ "61": 61,
122
+ "62": 62,
123
+ "63": 63,
124
+ "64": 64,
125
+ "65": 65,
126
+ "66": 66,
127
+ "67": 67,
128
+ "68": 68,
129
+ "69": 69,
130
+ "70": 70,
131
+ "71": 71,
132
+ "72": 72,
133
+ "73": 73,
134
+ "74": 74,
135
+ "75": 75,
136
+ "76": 76,
137
+ "77": 77,
138
+ "78": 78,
139
+ "79": 79,
140
+ "80": 80,
141
+ "81": 81,
142
+ "82": 82,
143
+ "83": 83,
144
+ "84": 84,
145
+ "85": 85,
146
+ "86": 86,
147
+ "87": 87,
148
+ "88": 88,
149
+ "89": 89,
150
+ "90": 90,
151
+ "91": 91,
152
+ "92": 92,
153
+ "93": 93,
154
+ "94": 94,
155
+ "95": 95,
156
+ "96": 96,
157
+ "97": 97,
158
+ "98": 98,
159
+ "99": 99,
160
+ "100": 100
161
+ },
162
+ "initializer_range": 0.02,
163
+ "intermediate_size": 3072,
164
+ "label2id": {
165
+ "0": 0,
166
+ "1": 1,
167
+ "2": 2,
168
+ "3": 3,
169
+ "4": 4,
170
+ "5": 5,
171
+ "6": 6,
172
+ "7": 7,
173
+ "8": 8,
174
+ "9": 9,
175
+ "10": 10,
176
+ "11": 11,
177
+ "12": 12,
178
+ "13": 13,
179
+ "14": 14,
180
+ "15": 15,
181
+ "16": 16,
182
+ "17": 17,
183
+ "18": 18,
184
+ "19": 19,
185
+ "20": 20,
186
+ "21": 21,
187
+ "22": 22,
188
+ "23": 23,
189
+ "24": 24,
190
+ "25": 25,
191
+ "26": 26,
192
+ "27": 27,
193
+ "28": 28,
194
+ "29": 29,
195
+ "30": 30,
196
+ "31": 31,
197
+ "32": 32,
198
+ "33": 33,
199
+ "34": 34,
200
+ "35": 35,
201
+ "36": 36,
202
+ "37": 37,
203
+ "38": 38,
204
+ "39": 39,
205
+ "40": 40,
206
+ "41": 41,
207
+ "42": 42,
208
+ "43": 43,
209
+ "44": 44,
210
+ "45": 45,
211
+ "46": 46,
212
+ "47": 47,
213
+ "48": 48,
214
+ "49": 49,
215
+ "50": 50,
216
+ "51": 51,
217
+ "52": 52,
218
+ "53": 53,
219
+ "54": 54,
220
+ "55": 55,
221
+ "56": 56,
222
+ "57": 57,
223
+ "58": 58,
224
+ "59": 59,
225
+ "60": 60,
226
+ "61": 61,
227
+ "62": 62,
228
+ "63": 63,
229
+ "64": 64,
230
+ "65": 65,
231
+ "66": 66,
232
+ "67": 67,
233
+ "68": 68,
234
+ "69": 69,
235
+ "70": 70,
236
+ "71": 71,
237
+ "72": 72,
238
+ "73": 73,
239
+ "74": 74,
240
+ "75": 75,
241
+ "76": 76,
242
+ "77": 77,
243
+ "78": 78,
244
+ "79": 79,
245
+ "80": 80,
246
+ "81": 81,
247
+ "82": 82,
248
+ "83": 83,
249
+ "84": 84,
250
+ "85": 85,
251
+ "86": 86,
252
+ "87": 87,
253
+ "88": 88,
254
+ "89": 89,
255
+ "90": 90,
256
+ "91": 91,
257
+ "92": 92,
258
+ "93": 93,
259
+ "94": 94,
260
+ "95": 95,
261
+ "96": 96,
262
+ "97": 97,
263
+ "98": 98,
264
+ "99": 99,
265
+ "100": 100
266
+ },
267
+ "layer_norm_eps": 1e-05,
268
+ "layerdrop": 0.0,
269
+ "mask_channel_length": 10,
270
+ "mask_channel_min_space": 1,
271
+ "mask_channel_other": 0.0,
272
+ "mask_channel_prob": 0.0,
273
+ "mask_channel_selection": "static",
274
+ "mask_feature_length": 10,
275
+ "mask_feature_min_masks": 0,
276
+ "mask_feature_prob": 0.0,
277
+ "mask_time_length": 10,
278
+ "mask_time_min_masks": 2,
279
+ "mask_time_min_space": 1,
280
+ "mask_time_other": 0.0,
281
+ "mask_time_prob": 0.05,
282
+ "mask_time_selection": "static",
283
+ "model_type": "wav2vec2",
284
+ "no_mask_channel_overlap": false,
285
+ "no_mask_time_overlap": false,
286
+ "num_adapter_layers": 3,
287
+ "num_attention_heads": 12,
288
+ "num_codevector_groups": 2,
289
+ "num_codevectors_per_group": 320,
290
+ "num_conv_pos_embedding_groups": 16,
291
+ "num_conv_pos_embeddings": 128,
292
+ "num_feat_extract_layers": 7,
293
+ "num_hidden_layers": 12,
294
+ "num_negatives": 100,
295
+ "output_hidden_size": 768,
296
+ "pad_token_id": 0,
297
+ "proj_codevector_dim": 256,
298
+ "tdnn_dilation": [
299
+ 1,
300
+ 2,
301
+ 3,
302
+ 1,
303
+ 1
304
+ ],
305
+ "tdnn_dim": [
306
+ 512,
307
+ 512,
308
+ 512,
309
+ 512,
310
+ 1500
311
+ ],
312
+ "tdnn_kernel": [
313
+ 5,
314
+ 3,
315
+ 3,
316
+ 1,
317
+ 1
318
+ ],
319
+ "torch_dtype": "float32",
320
+ "transformers_version": "4.42.0.dev0",
321
+ "use_weighted_layer_sum": false,
322
+ "vocab_size": 32,
323
+ "xvector_output_dim": 512
324
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5bafbb8e27b48bbcc08c65593304a1a2c7f03520ccd30c36e848e93d6c8cc785
3
+ size 378404148
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
+ }
runs/May20_11-43-51_7cb4956f5d3d/events.out.tfevents.1716205447.7cb4956f5d3d.34.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0482930161cb09ce3ef226cc7a490f656254d7dfad4cd9ec87302594043f9c06
3
+ size 30090
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e69ac8af8cd1ae7c9188fbed36684338a48702a1238bfb5920f01197efc2ec9f
3
+ size 5176