Hugging Face's logo

alkiskoudounas
/
wav2vec2-base-fsc-gold

Audio Classification
Transformers
Safetensors
English
wav2vec2
intent-classification
slu
Model card Files Community
alkiskoudounas commited on
Commit
b96fc0f
·
verified ·
1 Parent(s): 7b761f0

Upload with huggingface_hub

Browse files
Files changed (7) hide show
  1. config.json +184 -0
  2. model.safetensors +3 -0
  3. optimizer.pt +3 -0
  4. rng_state.pth +3 -0
  5. scheduler.pt +3 -0
  6. trainer_state.json +223 -0
  7. training_args.bin +3 -0
config.json ADDED
@@ -0,0 +1,184 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "change languagenonenone",
61
+ "1": "activatemusicnone",
62
+ "10": "deactivatelampnone",
63
+ "11": "activatelightskitchen",
64
+ "12": "activatelightsbedroom",
65
+ "13": "activatelightswashroom",
66
+ "14": "deactivatelightskitchen",
67
+ "15": "deactivatelightsbedroom",
68
+ "16": "deactivatelightswashroom",
69
+ "17": "increaseheatkitchen",
70
+ "18": "increaseheatbedroom",
71
+ "19": "increaseheatwashroom",
72
+ "2": "activatelightsnone",
73
+ "20": "decreaseheatkitchen",
74
+ "21": "decreaseheatbedroom",
75
+ "22": "decreaseheatwashroom",
76
+ "23": "bringnewspapernone",
77
+ "24": "bringjuicenone",
78
+ "25": "bringsocksnone",
79
+ "26": "change languageChinesenone",
80
+ "27": "change languageKoreannone",
81
+ "28": "change languageEnglishnone",
82
+ "29": "change languageGermannone",
83
+ "3": "deactivatelightsnone",
84
+ "30": "bringshoesnone",
85
+ "4": "increasevolumenone",
86
+ "5": "decreasevolumenone",
87
+ "6": "increaseheatnone",
88
+ "7": "decreaseheatnone",
89
+ "8": "deactivatemusicnone",
90
+ "9": "activatelampnone"
91
+ },
92
+ "initializer_range": 0.02,
93
+ "intermediate_size": 3072,
94
+ "label2id": {
95
+ "activatelampnone": "9",
96
+ "activatelightsbedroom": "12",
97
+ "activatelightskitchen": "11",
98
+ "activatelightsnone": "2",
99
+ "activatelightswashroom": "13",
100
+ "activatemusicnone": "1",
101
+ "bringjuicenone": "24",
102
+ "bringnewspapernone": "23",
103
+ "bringshoesnone": "30",
104
+ "bringsocksnone": "25",
105
+ "change languageChinesenone": "26",
106
+ "change languageEnglishnone": "28",
107
+ "change languageGermannone": "29",
108
+ "change languageKoreannone": "27",
109
+ "change languagenonenone": "0",
110
+ "deactivatelampnone": "10",
111
+ "deactivatelightsbedroom": "15",
112
+ "deactivatelightskitchen": "14",
113
+ "deactivatelightsnone": "3",
114
+ "deactivatelightswashroom": "16",
115
+ "deactivatemusicnone": "8",
116
+ "decreaseheatbedroom": "21",
117
+ "decreaseheatkitchen": "20",
118
+ "decreaseheatnone": "7",
119
+ "decreaseheatwashroom": "22",
120
+ "decreasevolumenone": "5",
121
+ "increaseheatbedroom": "18",
122
+ "increaseheatkitchen": "17",
123
+ "increaseheatnone": "6",
124
+ "increaseheatwashroom": "19",
125
+ "increasevolumenone": "4"
126
+ },
127
+ "layer_norm_eps": 1e-05,
128
+ "layerdrop": 0.0,
129
+ "mask_channel_length": 10,
130
+ "mask_channel_min_space": 1,
131
+ "mask_channel_other": 0.0,
132
+ "mask_channel_prob": 0.0,
133
+ "mask_channel_selection": "static",
134
+ "mask_feature_length": 10,
135
+ "mask_feature_min_masks": 0,
136
+ "mask_feature_prob": 0.0,
137
+ "mask_time_length": 10,
138
+ "mask_time_min_masks": 2,
139
+ "mask_time_min_space": 1,
140
+ "mask_time_other": 0.0,
141
+ "mask_time_prob": 0.05,
142
+ "mask_time_selection": "static",
143
+ "model_type": "wav2vec2",
144
+ "no_mask_channel_overlap": false,
145
+ "no_mask_time_overlap": false,
146
+ "num_adapter_layers": 3,
147
+ "num_attention_heads": 12,
148
+ "num_codevector_groups": 2,
149
+ "num_codevectors_per_group": 320,
150
+ "num_conv_pos_embedding_groups": 16,
151
+ "num_conv_pos_embeddings": 128,
152
+ "num_feat_extract_layers": 7,
153
+ "num_hidden_layers": 12,
154
+ "num_negatives": 100,
155
+ "output_hidden_size": 768,
156
+ "pad_token_id": 0,
157
+ "proj_codevector_dim": 256,
158
+ "tdnn_dilation": [
159
+ 1,
160
+ 2,
161
+ 3,
162
+ 1,
163
+ 1
164
+ ],
165
+ "tdnn_dim": [
166
+ 512,
167
+ 512,
168
+ 512,
169
+ 512,
170
+ 1500
171
+ ],
172
+ "tdnn_kernel": [
173
+ 5,
174
+ 3,
175
+ 3,
176
+ 1,
177
+ 1
178
+ ],
179
+ "torch_dtype": "float32",
180
+ "transformers_version": "4.46.1",
181
+ "use_weighted_layer_sum": false,
182
+ "vocab_size": 32,
183
+ "xvector_output_dim": 512
184
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5a31acaf3226ae35bc4f8b4c7cf6d44115d9f0e45a65ebb3b2887874f5fa48d
3
+ size 378332188
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f92dee02caf3c3931041711437d8326916040f17d405c51b3559fb8d506c61af
3
+ size 756792122
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7910b10d57e574f940f61fcfc055b2ff417a40c4a2c3007688161a4cf22aa627
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6a1e3232c9dd0558d3b582345b4a537ca2d63d9ae2df7d7189cbdc92c4e384f
3
+ size 1064
trainer_state.json ADDED
@@ -0,0 +1,223 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9746632456703015,
3
+ "best_model_checkpoint": "/data3/akoudounas/speech_unlearning/models/fsc/facebook/wav2vec2-base/4/_retain/checkpoint-2000",
4
+ "epoch": 2.869440459110473,
5
+ "eval_steps": 400,
6
+ "global_step": 2000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.14347202295552366,
13
+ "grad_norm": 10.184388160705566,
14
+ "learning_rate": 5e-05,
15
+ "loss": 13.3354,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 0.28694404591104733,
20
+ "grad_norm": 22.545684814453125,
21
+ "learning_rate": 0.0001,
22
+ "loss": 10.7429,
23
+ "step": 200
24
+ },
25
+ {
26
+ "epoch": 0.430416068866571,
27
+ "grad_norm": 56.65978240966797,
28
+ "learning_rate": 9.444444444444444e-05,
29
+ "loss": 6.6371,
30
+ "step": 300
31
+ },
32
+ {
33
+ "epoch": 0.5738880918220947,
34
+ "grad_norm": 58.855796813964844,
35
+ "learning_rate": 8.888888888888889e-05,
36
+ "loss": 4.8056,
37
+ "step": 400
38
+ },
39
+ {
40
+ "epoch": 0.5738880918220947,
41
+ "eval_accuracy": 0.6023091725465042,
42
+ "eval_f1_macro": 0.5295300525459568,
43
+ "eval_loss": 1.0795540809631348,
44
+ "eval_runtime": 13.9266,
45
+ "eval_samples_per_second": 223.887,
46
+ "eval_steps_per_second": 28.004,
47
+ "step": 400
48
+ },
49
+ {
50
+ "epoch": 0.7173601147776184,
51
+ "grad_norm": 19.51607322692871,
52
+ "learning_rate": 8.333333333333334e-05,
53
+ "loss": 3.8297,
54
+ "step": 500
55
+ },
56
+ {
57
+ "epoch": 0.860832137733142,
58
+ "grad_norm": 290.1480712890625,
59
+ "learning_rate": 7.777777777777778e-05,
60
+ "loss": 2.8656,
61
+ "step": 600
62
+ },
63
+ {
64
+ "epoch": 1.0043041606886658,
65
+ "grad_norm": 20.89418601989746,
66
+ "learning_rate": 7.222222222222222e-05,
67
+ "loss": 2.0125,
68
+ "step": 700
69
+ },
70
+ {
71
+ "epoch": 1.1477761836441893,
72
+ "grad_norm": 12.759949684143066,
73
+ "learning_rate": 6.666666666666667e-05,
74
+ "loss": 1.3945,
75
+ "step": 800
76
+ },
77
+ {
78
+ "epoch": 1.1477761836441893,
79
+ "eval_accuracy": 0.9207825529185375,
80
+ "eval_f1_macro": 0.9143966950672161,
81
+ "eval_loss": 0.3709124028682709,
82
+ "eval_runtime": 19.3466,
83
+ "eval_samples_per_second": 161.165,
84
+ "eval_steps_per_second": 20.159,
85
+ "step": 800
86
+ },
87
+ {
88
+ "epoch": 1.291248206599713,
89
+ "grad_norm": 4.4212541580200195,
90
+ "learning_rate": 6.111111111111112e-05,
91
+ "loss": 1.0552,
92
+ "step": 900
93
+ },
94
+ {
95
+ "epoch": 1.4347202295552366,
96
+ "grad_norm": 3.2309963703155518,
97
+ "learning_rate": 5.555555555555556e-05,
98
+ "loss": 0.9127,
99
+ "step": 1000
100
+ },
101
+ {
102
+ "epoch": 1.5781922525107603,
103
+ "grad_norm": 50.245872497558594,
104
+ "learning_rate": 5e-05,
105
+ "loss": 0.8042,
106
+ "step": 1100
107
+ },
108
+ {
109
+ "epoch": 1.721664275466284,
110
+ "grad_norm": 190.7835235595703,
111
+ "learning_rate": 4.4444444444444447e-05,
112
+ "loss": 0.6361,
113
+ "step": 1200
114
+ },
115
+ {
116
+ "epoch": 1.721664275466284,
117
+ "eval_accuracy": 0.9528543938422065,
118
+ "eval_f1_macro": 0.9475574056240644,
119
+ "eval_loss": 0.2340688854455948,
120
+ "eval_runtime": 13.9095,
121
+ "eval_samples_per_second": 224.163,
122
+ "eval_steps_per_second": 28.038,
123
+ "step": 1200
124
+ },
125
+ {
126
+ "epoch": 1.8651362984218078,
127
+ "grad_norm": 0.5275015234947205,
128
+ "learning_rate": 3.888888888888889e-05,
129
+ "loss": 0.6868,
130
+ "step": 1300
131
+ },
132
+ {
133
+ "epoch": 2.0086083213773316,
134
+ "grad_norm": 0.28355035185813904,
135
+ "learning_rate": 3.3333333333333335e-05,
136
+ "loss": 0.6716,
137
+ "step": 1400
138
+ },
139
+ {
140
+ "epoch": 2.152080344332855,
141
+ "grad_norm": 0.17906032502651215,
142
+ "learning_rate": 2.777777777777778e-05,
143
+ "loss": 0.4126,
144
+ "step": 1500
145
+ },
146
+ {
147
+ "epoch": 2.2955523672883786,
148
+ "grad_norm": 0.20960959792137146,
149
+ "learning_rate": 2.2222222222222223e-05,
150
+ "loss": 0.3714,
151
+ "step": 1600
152
+ },
153
+ {
154
+ "epoch": 2.2955523672883786,
155
+ "eval_accuracy": 0.967928159076331,
156
+ "eval_f1_macro": 0.9641675805500564,
157
+ "eval_loss": 0.16309456527233124,
158
+ "eval_runtime": 13.748,
159
+ "eval_samples_per_second": 226.797,
160
+ "eval_steps_per_second": 28.368,
161
+ "step": 1600
162
+ },
163
+ {
164
+ "epoch": 2.4390243902439024,
165
+ "grad_norm": 5.835017204284668,
166
+ "learning_rate": 1.6666666666666667e-05,
167
+ "loss": 0.4885,
168
+ "step": 1700
169
+ },
170
+ {
171
+ "epoch": 2.582496413199426,
172
+ "grad_norm": 0.15750645101070404,
173
+ "learning_rate": 1.1111111111111112e-05,
174
+ "loss": 0.3515,
175
+ "step": 1800
176
+ },
177
+ {
178
+ "epoch": 2.72596843615495,
179
+ "grad_norm": 29.3380126953125,
180
+ "learning_rate": 5.555555555555556e-06,
181
+ "loss": 0.3854,
182
+ "step": 1900
183
+ },
184
+ {
185
+ "epoch": 2.869440459110473,
186
+ "grad_norm": 12.397686004638672,
187
+ "learning_rate": 0.0,
188
+ "loss": 0.3996,
189
+ "step": 2000
190
+ },
191
+ {
192
+ "epoch": 2.869440459110473,
193
+ "eval_accuracy": 0.9746632456703015,
194
+ "eval_f1_macro": 0.9700002035002927,
195
+ "eval_loss": 0.13398364186286926,
196
+ "eval_runtime": 13.9617,
197
+ "eval_samples_per_second": 223.325,
198
+ "eval_steps_per_second": 27.933,
199
+ "step": 2000
200
+ }
201
+ ],
202
+ "logging_steps": 100,
203
+ "max_steps": 2000,
204
+ "num_input_tokens_seen": 0,
205
+ "num_train_epochs": 3,
206
+ "save_steps": 2000,
207
+ "stateful_callbacks": {
208
+ "TrainerControl": {
209
+ "args": {
210
+ "should_epoch_stop": false,
211
+ "should_evaluate": false,
212
+ "should_log": false,
213
+ "should_save": true,
214
+ "should_training_stop": true
215
+ },
216
+ "attributes": {}
217
+ }
218
+ },
219
+ "total_flos": 2.324313120768e+18,
220
+ "train_batch_size": 8,
221
+ "trial_name": null,
222
+ "trial_params": null
223
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55a30dd8df422f6f86c864f1aaf0fd05e1c290dfdc0ea4c226ffd77e4f365fbe
3
+ size 5368