Hugging Face's logo

coolbirdv
/
wav2vec2-base-finetuned-ks

TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Eval Results
Model card Files Metrics Community
coolbirdv commited on
Commit
30dd174
•
1 Parent(s): 9cc7150

Training in progress, epoch 0

Browse files
Files changed (5) hide show
  1. config.json +214 -124
  2. model.safetensors +2 -2
  3. preprocessor_config.json +9 -9
  4. runs/Aug10_22-37-11_9c2d26d2d62c/events.out.tfevents.1723329434.9c2d26d2d62c.23.0 +3 -0
  5. training_args.bin +1 -1
config.json CHANGED
@@ -1,124 +1,214 @@
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": "VIVOSSPK27"
61
- },
62
- "initializer_range": 0.02,
63
- "intermediate_size": 3072,
64
- "label2id": {
65
- "VIVOSSPK27": "0"
66
- },
67
- "layer_norm_eps": 1e-05,
68
- "layerdrop": 0.0,
69
- "mask_channel_length": 10,
70
- "mask_channel_min_space": 1,
71
- "mask_channel_other": 0.0,
72
- "mask_channel_prob": 0.0,
73
- "mask_channel_selection": "static",
74
- "mask_feature_length": 10,
75
- "mask_feature_min_masks": 0,
76
- "mask_feature_prob": 0.0,
77
- "mask_time_length": 10,
78
- "mask_time_min_masks": 2,
79
- "mask_time_min_space": 1,
80
- "mask_time_other": 0.0,
81
- "mask_time_prob": 0.05,
82
- "mask_time_selection": "static",
83
- "model_type": "wav2vec2",
84
- "no_mask_channel_overlap": false,
85
- "no_mask_time_overlap": false,
86
- "num_adapter_layers": 3,
87
- "num_attention_heads": 12,
88
- "num_codevector_groups": 2,
89
- "num_codevectors_per_group": 320,
90
- "num_conv_pos_embedding_groups": 16,
91
- "num_conv_pos_embeddings": 128,
92
- "num_feat_extract_layers": 7,
93
- "num_hidden_layers": 12,
94
- "num_negatives": 100,
95
- "output_hidden_size": 768,
96
- "pad_token_id": 0,
97
- "proj_codevector_dim": 256,
98
- "tdnn_dilation": [
99
- 1,
100
- 2,
101
- 3,
102
- 1,
103
- 1
104
- ],
105
- "tdnn_dim": [
106
- 512,
107
- 512,
108
- 512,
109
- 512,
110
- 1500
111
- ],
112
- "tdnn_kernel": [
113
- 5,
114
- 3,
115
- 3,
116
- 1,
117
- 1
118
- ],
119
- "torch_dtype": "float32",
120
- "transformers_version": "4.42.4",
121
- "use_weighted_layer_sum": false,
122
- "vocab_size": 32,
123
- "xvector_output_dim": 512
124
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "VIVOSSPK01",
61
+ "1": "VIVOSSPK02",
62
+ "10": "VIVOSSPK11",
63
+ "11": "VIVOSSPK12",
64
+ "12": "VIVOSSPK13",
65
+ "13": "VIVOSSPK14",
66
+ "14": "VIVOSSPK15",
67
+ "15": "VIVOSSPK16",
68
+ "16": "VIVOSSPK17",
69
+ "17": "VIVOSSPK18",
70
+ "18": "VIVOSSPK19",
71
+ "19": "VIVOSSPK20",
72
+ "2": "VIVOSSPK03",
73
+ "20": "VIVOSSPK21",
74
+ "21": "VIVOSSPK22",
75
+ "22": "VIVOSSPK23",
76
+ "23": "VIVOSSPK24",
77
+ "24": "VIVOSSPK25",
78
+ "25": "VIVOSSPK26",
79
+ "26": "VIVOSSPK27",
80
+ "27": "VIVOSSPK28",
81
+ "28": "VIVOSSPK29",
82
+ "29": "VIVOSSPK30",
83
+ "3": "VIVOSSPK04",
84
+ "30": "VIVOSSPK31",
85
+ "31": "VIVOSSPK32",
86
+ "32": "VIVOSSPK33",
87
+ "33": "VIVOSSPK34",
88
+ "34": "VIVOSSPK35",
89
+ "35": "VIVOSSPK36",
90
+ "36": "VIVOSSPK37",
91
+ "37": "VIVOSSPK38",
92
+ "38": "VIVOSSPK39",
93
+ "39": "VIVOSSPK40",
94
+ "4": "VIVOSSPK05",
95
+ "40": "VIVOSSPK41",
96
+ "41": "VIVOSSPK42",
97
+ "42": "VIVOSSPK43",
98
+ "43": "VIVOSSPK44",
99
+ "44": "VIVOSSPK45",
100
+ "45": "VIVOSSPK46",
101
+ "5": "VIVOSSPK06",
102
+ "6": "VIVOSSPK07",
103
+ "7": "VIVOSSPK08",
104
+ "8": "VIVOSSPK09",
105
+ "9": "VIVOSSPK10"
106
+ },
107
+ "initializer_range": 0.02,
108
+ "intermediate_size": 3072,
109
+ "label2id": {
110
+ "VIVOSSPK01": "0",
111
+ "VIVOSSPK02": "1",
112
+ "VIVOSSPK03": "2",
113
+ "VIVOSSPK04": "3",
114
+ "VIVOSSPK05": "4",
115
+ "VIVOSSPK06": "5",
116
+ "VIVOSSPK07": "6",
117
+ "VIVOSSPK08": "7",
118
+ "VIVOSSPK09": "8",
119
+ "VIVOSSPK10": "9",
120
+ "VIVOSSPK11": "10",
121
+ "VIVOSSPK12": "11",
122
+ "VIVOSSPK13": "12",
123
+ "VIVOSSPK14": "13",
124
+ "VIVOSSPK15": "14",
125
+ "VIVOSSPK16": "15",
126
+ "VIVOSSPK17": "16",
127
+ "VIVOSSPK18": "17",
128
+ "VIVOSSPK19": "18",
129
+ "VIVOSSPK20": "19",
130
+ "VIVOSSPK21": "20",
131
+ "VIVOSSPK22": "21",
132
+ "VIVOSSPK23": "22",
133
+ "VIVOSSPK24": "23",
134
+ "VIVOSSPK25": "24",
135
+ "VIVOSSPK26": "25",
136
+ "VIVOSSPK27": "26",
137
+ "VIVOSSPK28": "27",
138
+ "VIVOSSPK29": "28",
139
+ "VIVOSSPK30": "29",
140
+ "VIVOSSPK31": "30",
141
+ "VIVOSSPK32": "31",
142
+ "VIVOSSPK33": "32",
143
+ "VIVOSSPK34": "33",
144
+ "VIVOSSPK35": "34",
145
+ "VIVOSSPK36": "35",
146
+ "VIVOSSPK37": "36",
147
+ "VIVOSSPK38": "37",
148
+ "VIVOSSPK39": "38",
149
+ "VIVOSSPK40": "39",
150
+ "VIVOSSPK41": "40",
151
+ "VIVOSSPK42": "41",
152
+ "VIVOSSPK43": "42",
153
+ "VIVOSSPK44": "43",
154
+ "VIVOSSPK45": "44",
155
+ "VIVOSSPK46": "45"
156
+ },
157
+ "layer_norm_eps": 1e-05,
158
+ "layerdrop": 0.0,
159
+ "mask_channel_length": 10,
160
+ "mask_channel_min_space": 1,
161
+ "mask_channel_other": 0.0,
162
+ "mask_channel_prob": 0.0,
163
+ "mask_channel_selection": "static",
164
+ "mask_feature_length": 10,
165
+ "mask_feature_min_masks": 0,
166
+ "mask_feature_prob": 0.0,
167
+ "mask_time_length": 10,
168
+ "mask_time_min_masks": 2,
169
+ "mask_time_min_space": 1,
170
+ "mask_time_other": 0.0,
171
+ "mask_time_prob": 0.05,
172
+ "mask_time_selection": "static",
173
+ "model_type": "wav2vec2",
174
+ "no_mask_channel_overlap": false,
175
+ "no_mask_time_overlap": false,
176
+ "num_adapter_layers": 3,
177
+ "num_attention_heads": 12,
178
+ "num_codevector_groups": 2,
179
+ "num_codevectors_per_group": 320,
180
+ "num_conv_pos_embedding_groups": 16,
181
+ "num_conv_pos_embeddings": 128,
182
+ "num_feat_extract_layers": 7,
183
+ "num_hidden_layers": 12,
184
+ "num_negatives": 100,
185
+ "output_hidden_size": 768,
186
+ "pad_token_id": 0,
187
+ "proj_codevector_dim": 256,
188
+ "tdnn_dilation": [
189
+ 1,
190
+ 2,
191
+ 3,
192
+ 1,
193
+ 1
194
+ ],
195
+ "tdnn_dim": [
196
+ 512,
197
+ 512,
198
+ 512,
199
+ 512,
200
+ 1500
201
+ ],
202
+ "tdnn_kernel": [
203
+ 5,
204
+ 3,
205
+ 3,
206
+ 1,
207
+ 1
208
+ ],
209
+ "torch_dtype": "float32",
210
+ "transformers_version": "4.42.3",
211
+ "use_weighted_layer_sum": false,
212
+ "vocab_size": 32,
213
+ "xvector_output_dim": 512
214
+ }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2dc31ee48f921902a7a24f66f1cdc6940d02c489fbfa8ee7e37aae0895fb0e60
3
- size 378301332
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:878c082a3831e6cae89ada9873b777e5dbadc26f8172cd73a441680311d645ca
3
+ size 378347608
preprocessor_config.json CHANGED
@@ -1,9 +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
- }
 
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/Aug10_22-37-11_9c2d26d2d62c/events.out.tfevents.1723329434.9c2d26d2d62c.23.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0336011eb4bc3fd9a59a06cb22fa512c85d6372c8b6217fe597e1d996e66fd76
3
+ size 39681
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bb7fd671aaa524683102b1e8ab74decbec26405e581c5dcfabbcaecad095ebfc
3
  size 5176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1867f9a6fec1990b8b6a5ac693c7b15ecfd511f43aad6bb16fcb0098d4e38321
3
  size 5176