fydhfzh commited on
Commit
be448dc
1 Parent(s): ae2e40a

Model save

Browse files
README.md ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: facebook/wav2vec2-base
4
+ tags:
5
+ - generated_from_trainer
6
+ model-index:
7
+ - name: wav2vec2-classifier
8
+ results: []
9
+ ---
10
+
11
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
+ should probably proofread and complete it, then remove this comment. -->
13
+
14
+ # wav2vec2-classifier
15
+
16
+ This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on an unknown dataset.
17
+
18
+ ## Model description
19
+
20
+ More information needed
21
+
22
+ ## Intended uses & limitations
23
+
24
+ More information needed
25
+
26
+ ## Training and evaluation data
27
+
28
+ More information needed
29
+
30
+ ## Training procedure
31
+
32
+ ### Training hyperparameters
33
+
34
+ The following hyperparameters were used during training:
35
+ - learning_rate: 3e-05
36
+ - train_batch_size: 32
37
+ - eval_batch_size: 32
38
+ - seed: 42
39
+ - gradient_accumulation_steps: 4
40
+ - total_train_batch_size: 128
41
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
+ - lr_scheduler_type: linear
43
+ - num_epochs: 10
44
+ - mixed_precision_training: Native AMP
45
+
46
+ ### Framework versions
47
+
48
+ - Transformers 4.38.2
49
+ - Pytorch 2.3.0
50
+ - Datasets 2.19.1
51
+ - Tokenizers 0.15.1
config.json ADDED
@@ -0,0 +1,290 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "LABEL_0",
61
+ "1": "LABEL_1",
62
+ "2": "LABEL_2",
63
+ "3": "LABEL_3",
64
+ "4": "LABEL_4",
65
+ "5": "LABEL_5",
66
+ "6": "LABEL_6",
67
+ "7": "LABEL_7",
68
+ "8": "LABEL_8",
69
+ "9": "LABEL_9",
70
+ "10": "LABEL_10",
71
+ "11": "LABEL_11",
72
+ "12": "LABEL_12",
73
+ "13": "LABEL_13",
74
+ "14": "LABEL_14",
75
+ "15": "LABEL_15",
76
+ "16": "LABEL_16",
77
+ "17": "LABEL_17",
78
+ "18": "LABEL_18",
79
+ "19": "LABEL_19",
80
+ "20": "LABEL_20",
81
+ "21": "LABEL_21",
82
+ "22": "LABEL_22",
83
+ "23": "LABEL_23",
84
+ "24": "LABEL_24",
85
+ "25": "LABEL_25",
86
+ "26": "LABEL_26",
87
+ "27": "LABEL_27",
88
+ "28": "LABEL_28",
89
+ "29": "LABEL_29",
90
+ "30": "LABEL_30",
91
+ "31": "LABEL_31",
92
+ "32": "LABEL_32",
93
+ "33": "LABEL_33",
94
+ "34": "LABEL_34",
95
+ "35": "LABEL_35",
96
+ "36": "LABEL_36",
97
+ "37": "LABEL_37",
98
+ "38": "LABEL_38",
99
+ "39": "LABEL_39",
100
+ "40": "LABEL_40",
101
+ "41": "LABEL_41",
102
+ "42": "LABEL_42",
103
+ "43": "LABEL_43",
104
+ "44": "LABEL_44",
105
+ "45": "LABEL_45",
106
+ "46": "LABEL_46",
107
+ "47": "LABEL_47",
108
+ "48": "LABEL_48",
109
+ "49": "LABEL_49",
110
+ "50": "LABEL_50",
111
+ "51": "LABEL_51",
112
+ "52": "LABEL_52",
113
+ "53": "LABEL_53",
114
+ "54": "LABEL_54",
115
+ "55": "LABEL_55",
116
+ "56": "LABEL_56",
117
+ "57": "LABEL_57",
118
+ "58": "LABEL_58",
119
+ "59": "LABEL_59",
120
+ "60": "LABEL_60",
121
+ "61": "LABEL_61",
122
+ "62": "LABEL_62",
123
+ "63": "LABEL_63",
124
+ "64": "LABEL_64",
125
+ "65": "LABEL_65",
126
+ "66": "LABEL_66",
127
+ "67": "LABEL_67",
128
+ "68": "LABEL_68",
129
+ "69": "LABEL_69",
130
+ "70": "LABEL_70",
131
+ "71": "LABEL_71",
132
+ "72": "LABEL_72",
133
+ "73": "LABEL_73",
134
+ "74": "LABEL_74",
135
+ "75": "LABEL_75",
136
+ "76": "LABEL_76",
137
+ "77": "LABEL_77",
138
+ "78": "LABEL_78",
139
+ "79": "LABEL_79",
140
+ "80": "LABEL_80",
141
+ "81": "LABEL_81",
142
+ "82": "LABEL_82",
143
+ "83": "LABEL_83"
144
+ },
145
+ "initializer_range": 0.02,
146
+ "intermediate_size": 3072,
147
+ "label2id": {
148
+ "LABEL_0": 0,
149
+ "LABEL_1": 1,
150
+ "LABEL_10": 10,
151
+ "LABEL_11": 11,
152
+ "LABEL_12": 12,
153
+ "LABEL_13": 13,
154
+ "LABEL_14": 14,
155
+ "LABEL_15": 15,
156
+ "LABEL_16": 16,
157
+ "LABEL_17": 17,
158
+ "LABEL_18": 18,
159
+ "LABEL_19": 19,
160
+ "LABEL_2": 2,
161
+ "LABEL_20": 20,
162
+ "LABEL_21": 21,
163
+ "LABEL_22": 22,
164
+ "LABEL_23": 23,
165
+ "LABEL_24": 24,
166
+ "LABEL_25": 25,
167
+ "LABEL_26": 26,
168
+ "LABEL_27": 27,
169
+ "LABEL_28": 28,
170
+ "LABEL_29": 29,
171
+ "LABEL_3": 3,
172
+ "LABEL_30": 30,
173
+ "LABEL_31": 31,
174
+ "LABEL_32": 32,
175
+ "LABEL_33": 33,
176
+ "LABEL_34": 34,
177
+ "LABEL_35": 35,
178
+ "LABEL_36": 36,
179
+ "LABEL_37": 37,
180
+ "LABEL_38": 38,
181
+ "LABEL_39": 39,
182
+ "LABEL_4": 4,
183
+ "LABEL_40": 40,
184
+ "LABEL_41": 41,
185
+ "LABEL_42": 42,
186
+ "LABEL_43": 43,
187
+ "LABEL_44": 44,
188
+ "LABEL_45": 45,
189
+ "LABEL_46": 46,
190
+ "LABEL_47": 47,
191
+ "LABEL_48": 48,
192
+ "LABEL_49": 49,
193
+ "LABEL_5": 5,
194
+ "LABEL_50": 50,
195
+ "LABEL_51": 51,
196
+ "LABEL_52": 52,
197
+ "LABEL_53": 53,
198
+ "LABEL_54": 54,
199
+ "LABEL_55": 55,
200
+ "LABEL_56": 56,
201
+ "LABEL_57": 57,
202
+ "LABEL_58": 58,
203
+ "LABEL_59": 59,
204
+ "LABEL_6": 6,
205
+ "LABEL_60": 60,
206
+ "LABEL_61": 61,
207
+ "LABEL_62": 62,
208
+ "LABEL_63": 63,
209
+ "LABEL_64": 64,
210
+ "LABEL_65": 65,
211
+ "LABEL_66": 66,
212
+ "LABEL_67": 67,
213
+ "LABEL_68": 68,
214
+ "LABEL_69": 69,
215
+ "LABEL_7": 7,
216
+ "LABEL_70": 70,
217
+ "LABEL_71": 71,
218
+ "LABEL_72": 72,
219
+ "LABEL_73": 73,
220
+ "LABEL_74": 74,
221
+ "LABEL_75": 75,
222
+ "LABEL_76": 76,
223
+ "LABEL_77": 77,
224
+ "LABEL_78": 78,
225
+ "LABEL_79": 79,
226
+ "LABEL_8": 8,
227
+ "LABEL_80": 80,
228
+ "LABEL_81": 81,
229
+ "LABEL_82": 82,
230
+ "LABEL_83": 83,
231
+ "LABEL_9": 9
232
+ },
233
+ "layer_norm_eps": 1e-05,
234
+ "layerdrop": 0.0,
235
+ "mask_channel_length": 10,
236
+ "mask_channel_min_space": 1,
237
+ "mask_channel_other": 0.0,
238
+ "mask_channel_prob": 0.0,
239
+ "mask_channel_selection": "static",
240
+ "mask_feature_length": 10,
241
+ "mask_feature_min_masks": 0,
242
+ "mask_feature_prob": 0.0,
243
+ "mask_time_length": 10,
244
+ "mask_time_min_masks": 2,
245
+ "mask_time_min_space": 1,
246
+ "mask_time_other": 0.0,
247
+ "mask_time_prob": 0.05,
248
+ "mask_time_selection": "static",
249
+ "model_type": "wav2vec2",
250
+ "no_mask_channel_overlap": false,
251
+ "no_mask_time_overlap": false,
252
+ "num_adapter_layers": 3,
253
+ "num_attention_heads": 12,
254
+ "num_codevector_groups": 2,
255
+ "num_codevectors_per_group": 320,
256
+ "num_conv_pos_embedding_groups": 16,
257
+ "num_conv_pos_embeddings": 128,
258
+ "num_feat_extract_layers": 7,
259
+ "num_hidden_layers": 12,
260
+ "num_negatives": 100,
261
+ "output_hidden_size": 768,
262
+ "pad_token_id": 0,
263
+ "proj_codevector_dim": 256,
264
+ "tdnn_dilation": [
265
+ 1,
266
+ 2,
267
+ 3,
268
+ 1,
269
+ 1
270
+ ],
271
+ "tdnn_dim": [
272
+ 512,
273
+ 512,
274
+ 512,
275
+ 512,
276
+ 1500
277
+ ],
278
+ "tdnn_kernel": [
279
+ 5,
280
+ 3,
281
+ 3,
282
+ 1,
283
+ 1
284
+ ],
285
+ "torch_dtype": "float32",
286
+ "transformers_version": "4.38.2",
287
+ "use_weighted_layer_sum": false,
288
+ "vocab_size": 32,
289
+ "xvector_output_dim": 512
290
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d63c79bf0db0f342abc9798ec25613af22903f49f53bb5c497b13ac1dc57dd54
3
+ size 378386672
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
+ }
runs/Jun13_00-04-21_LAPTOP-1GID9RGH/events.out.tfevents.1718211863.LAPTOP-1GID9RGH.20912.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3de9618ed6150df2afda2c32c3f4a6bf1eb20b27cce8fe938364d78f9aec3fc
3
+ size 42035
runs/Jun13_00-04-21_LAPTOP-1GID9RGH/events.out.tfevents.1718214542.LAPTOP-1GID9RGH.20912.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb8f581191c3c940222981d2881d7dee49c3ba0c42e95d3e3e32e8621ddae7f8
3
+ size 610
runs/Jun13_07-44-24_LAPTOP-1GID9RGH/events.out.tfevents.1718257465.LAPTOP-1GID9RGH.9096.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3a30b1a76dd8f00b1501d9c7b72d11f268b4b85aa463ea29acb6b881c11a3198
3
+ size 18015
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16307a07642fa67050a166782088d0e1fb7bff5b352f3fc4b6f1eed9c85adc52
3
+ size 4984