Jungwonchang commited on
Commit
c28a540
1 Parent(s): c0821f0

End of training

Browse files
README.md ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ license: apache-2.0
5
+ base_model: openai/whisper-large-v2
6
+ tags:
7
+ - generated_from_trainer
8
+ datasets:
9
+ - Jungwonchang/spgispeech_xs
10
+ model-index:
11
+ - name: openai/whisper-large-v2, all the parameters updated for 5 epochs
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # openai/whisper-large-v2, all the parameters updated for 5 epochs
19
+
20
+ This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the 2 hour dataset of SPGIspeech(custom dataset) dataset.
21
+
22
+ ## Model description
23
+
24
+ More information needed
25
+
26
+ ## Intended uses & limitations
27
+
28
+ More information needed
29
+
30
+ ## Training and evaluation data
31
+
32
+ More information needed
33
+
34
+ ## Training procedure
35
+
36
+ ### Training hyperparameters
37
+
38
+ The following hyperparameters were used during training:
39
+ - learning_rate: 1e-05
40
+ - train_batch_size: 16
41
+ - eval_batch_size: 16
42
+ - seed: 42
43
+ - gradient_accumulation_steps: 2
44
+ - total_train_batch_size: 32
45
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
+ - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_steps: 50
48
+ - training_steps: 120
49
+ - mixed_precision_training: Native AMP
50
+
51
+ ### Training results
52
+
53
+
54
+
55
+ ### Framework versions
56
+
57
+ - Transformers 4.36.0.dev0
58
+ - Pytorch 1.12.1+cu116
59
+ - Datasets 2.4.0
60
+ - Tokenizers 0.15.0
config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai/whisper-large-v2",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "apply_spec_augment": false,
6
+ "architectures": [
7
+ "WhisperForConditionalGeneration"
8
+ ],
9
+ "attention_dropout": 0.0,
10
+ "begin_suppress_tokens": [
11
+ 220,
12
+ 50257
13
+ ],
14
+ "bos_token_id": 50257,
15
+ "classifier_proj_size": 256,
16
+ "d_model": 1280,
17
+ "decoder_attention_heads": 20,
18
+ "decoder_ffn_dim": 5120,
19
+ "decoder_layerdrop": 0.0,
20
+ "decoder_layers": 32,
21
+ "decoder_start_token_id": 50258,
22
+ "dropout": 0.0,
23
+ "encoder_attention_heads": 20,
24
+ "encoder_ffn_dim": 5120,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 32,
27
+ "eos_token_id": 50257,
28
+ "forced_decoder_ids": null,
29
+ "init_std": 0.02,
30
+ "is_encoder_decoder": true,
31
+ "mask_feature_length": 10,
32
+ "mask_feature_min_masks": 0,
33
+ "mask_feature_prob": 0.0,
34
+ "mask_time_length": 10,
35
+ "mask_time_min_masks": 2,
36
+ "mask_time_prob": 0.05,
37
+ "max_length": 448,
38
+ "max_source_positions": 1500,
39
+ "max_target_positions": 448,
40
+ "median_filter_width": 7,
41
+ "model_type": "whisper",
42
+ "num_hidden_layers": 32,
43
+ "num_mel_bins": 80,
44
+ "pad_token_id": 50257,
45
+ "scale_embedding": false,
46
+ "suppress_tokens": [],
47
+ "torch_dtype": "float32",
48
+ "transformers_version": "4.36.0.dev0",
49
+ "use_cache": false,
50
+ "use_weighted_layer_sum": false,
51
+ "vocab_size": 51865
52
+ }
generation_config.json ADDED
@@ -0,0 +1,315 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 10,
5
+ 12
6
+ ],
7
+ [
8
+ 13,
9
+ 17
10
+ ],
11
+ [
12
+ 16,
13
+ 11
14
+ ],
15
+ [
16
+ 16,
17
+ 12
18
+ ],
19
+ [
20
+ 16,
21
+ 13
22
+ ],
23
+ [
24
+ 17,
25
+ 15
26
+ ],
27
+ [
28
+ 17,
29
+ 16
30
+ ],
31
+ [
32
+ 18,
33
+ 4
34
+ ],
35
+ [
36
+ 18,
37
+ 11
38
+ ],
39
+ [
40
+ 18,
41
+ 19
42
+ ],
43
+ [
44
+ 19,
45
+ 11
46
+ ],
47
+ [
48
+ 21,
49
+ 2
50
+ ],
51
+ [
52
+ 21,
53
+ 3
54
+ ],
55
+ [
56
+ 22,
57
+ 3
58
+ ],
59
+ [
60
+ 22,
61
+ 9
62
+ ],
63
+ [
64
+ 22,
65
+ 12
66
+ ],
67
+ [
68
+ 23,
69
+ 5
70
+ ],
71
+ [
72
+ 23,
73
+ 7
74
+ ],
75
+ [
76
+ 23,
77
+ 13
78
+ ],
79
+ [
80
+ 25,
81
+ 5
82
+ ],
83
+ [
84
+ 26,
85
+ 1
86
+ ],
87
+ [
88
+ 26,
89
+ 12
90
+ ],
91
+ [
92
+ 27,
93
+ 15
94
+ ]
95
+ ],
96
+ "begin_suppress_tokens": [
97
+ 220,
98
+ 50257
99
+ ],
100
+ "bos_token_id": 50257,
101
+ "decoder_start_token_id": 50258,
102
+ "eos_token_id": 50257,
103
+ "forced_decoder_ids": [
104
+ [
105
+ 1,
106
+ null
107
+ ],
108
+ [
109
+ 2,
110
+ 50359
111
+ ]
112
+ ],
113
+ "is_multilingual": true,
114
+ "lang_to_id": {
115
+ "<|af|>": 50327,
116
+ "<|am|>": 50334,
117
+ "<|ar|>": 50272,
118
+ "<|as|>": 50350,
119
+ "<|az|>": 50304,
120
+ "<|ba|>": 50355,
121
+ "<|be|>": 50330,
122
+ "<|bg|>": 50292,
123
+ "<|bn|>": 50302,
124
+ "<|bo|>": 50347,
125
+ "<|br|>": 50309,
126
+ "<|bs|>": 50315,
127
+ "<|ca|>": 50270,
128
+ "<|cs|>": 50283,
129
+ "<|cy|>": 50297,
130
+ "<|da|>": 50285,
131
+ "<|de|>": 50261,
132
+ "<|el|>": 50281,
133
+ "<|en|>": 50259,
134
+ "<|es|>": 50262,
135
+ "<|et|>": 50307,
136
+ "<|eu|>": 50310,
137
+ "<|fa|>": 50300,
138
+ "<|fi|>": 50277,
139
+ "<|fo|>": 50338,
140
+ "<|fr|>": 50265,
141
+ "<|gl|>": 50319,
142
+ "<|gu|>": 50333,
143
+ "<|haw|>": 50352,
144
+ "<|ha|>": 50354,
145
+ "<|he|>": 50279,
146
+ "<|hi|>": 50276,
147
+ "<|hr|>": 50291,
148
+ "<|ht|>": 50339,
149
+ "<|hu|>": 50286,
150
+ "<|hy|>": 50312,
151
+ "<|id|>": 50275,
152
+ "<|is|>": 50311,
153
+ "<|it|>": 50274,
154
+ "<|ja|>": 50266,
155
+ "<|jw|>": 50356,
156
+ "<|ka|>": 50329,
157
+ "<|kk|>": 50316,
158
+ "<|km|>": 50323,
159
+ "<|kn|>": 50306,
160
+ "<|ko|>": 50264,
161
+ "<|la|>": 50294,
162
+ "<|lb|>": 50345,
163
+ "<|ln|>": 50353,
164
+ "<|lo|>": 50336,
165
+ "<|lt|>": 50293,
166
+ "<|lv|>": 50301,
167
+ "<|mg|>": 50349,
168
+ "<|mi|>": 50295,
169
+ "<|mk|>": 50308,
170
+ "<|ml|>": 50296,
171
+ "<|mn|>": 50314,
172
+ "<|mr|>": 50320,
173
+ "<|ms|>": 50282,
174
+ "<|mt|>": 50343,
175
+ "<|my|>": 50346,
176
+ "<|ne|>": 50313,
177
+ "<|nl|>": 50271,
178
+ "<|nn|>": 50342,
179
+ "<|no|>": 50288,
180
+ "<|oc|>": 50328,
181
+ "<|pa|>": 50321,
182
+ "<|pl|>": 50269,
183
+ "<|ps|>": 50340,
184
+ "<|pt|>": 50267,
185
+ "<|ro|>": 50284,
186
+ "<|ru|>": 50263,
187
+ "<|sa|>": 50344,
188
+ "<|sd|>": 50332,
189
+ "<|si|>": 50322,
190
+ "<|sk|>": 50298,
191
+ "<|sl|>": 50305,
192
+ "<|sn|>": 50324,
193
+ "<|so|>": 50326,
194
+ "<|sq|>": 50317,
195
+ "<|sr|>": 50303,
196
+ "<|su|>": 50357,
197
+ "<|sv|>": 50273,
198
+ "<|sw|>": 50318,
199
+ "<|ta|>": 50287,
200
+ "<|te|>": 50299,
201
+ "<|tg|>": 50331,
202
+ "<|th|>": 50289,
203
+ "<|tk|>": 50341,
204
+ "<|tl|>": 50348,
205
+ "<|tr|>": 50268,
206
+ "<|tt|>": 50351,
207
+ "<|uk|>": 50280,
208
+ "<|ur|>": 50290,
209
+ "<|uz|>": 50337,
210
+ "<|vi|>": 50278,
211
+ "<|yi|>": 50335,
212
+ "<|yo|>": 50325,
213
+ "<|zh|>": 50260
214
+ },
215
+ "max_initial_timestamp_index": 1,
216
+ "max_length": 448,
217
+ "no_timestamps_token_id": 50363,
218
+ "pad_token_id": 50257,
219
+ "return_timestamps": false,
220
+ "suppress_tokens": [
221
+ 1,
222
+ 2,
223
+ 7,
224
+ 8,
225
+ 9,
226
+ 10,
227
+ 14,
228
+ 25,
229
+ 26,
230
+ 27,
231
+ 28,
232
+ 29,
233
+ 31,
234
+ 58,
235
+ 59,
236
+ 60,
237
+ 61,
238
+ 62,
239
+ 63,
240
+ 90,
241
+ 91,
242
+ 92,
243
+ 93,
244
+ 359,
245
+ 503,
246
+ 522,
247
+ 542,
248
+ 873,
249
+ 893,
250
+ 902,
251
+ 918,
252
+ 922,
253
+ 931,
254
+ 1350,
255
+ 1853,
256
+ 1982,
257
+ 2460,
258
+ 2627,
259
+ 3246,
260
+ 3253,
261
+ 3268,
262
+ 3536,
263
+ 3846,
264
+ 3961,
265
+ 4183,
266
+ 4667,
267
+ 6585,
268
+ 6647,
269
+ 7273,
270
+ 9061,
271
+ 9383,
272
+ 10428,
273
+ 10929,
274
+ 11938,
275
+ 12033,
276
+ 12331,
277
+ 12562,
278
+ 13793,
279
+ 14157,
280
+ 14635,
281
+ 15265,
282
+ 15618,
283
+ 16553,
284
+ 16604,
285
+ 18362,
286
+ 18956,
287
+ 20075,
288
+ 21675,
289
+ 22520,
290
+ 26130,
291
+ 26161,
292
+ 26435,
293
+ 28279,
294
+ 29464,
295
+ 31650,
296
+ 32302,
297
+ 32470,
298
+ 36865,
299
+ 42863,
300
+ 47425,
301
+ 49870,
302
+ 50254,
303
+ 50258,
304
+ 50358,
305
+ 50359,
306
+ 50360,
307
+ 50361,
308
+ 50362
309
+ ],
310
+ "task_to_id": {
311
+ "transcribe": 50359,
312
+ "translate": 50358
313
+ },
314
+ "transformers_version": "4.36.0.dev0"
315
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0fd407735ebe1606921fb354806c1dbaf780f5baa3f35282d821f93eeb0af5d6
3
+ size 4992706480
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74e78e422dd2ae9b2171b061e61f1d75680e436edb961e495af22081fb458a7c
3
+ size 1180663192
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
preprocessor_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length": 30,
3
+ "feature_extractor_type": "WhisperFeatureExtractor",
4
+ "feature_size": 80,
5
+ "hop_length": 160,
6
+ "n_fft": 400,
7
+ "n_samples": 480000,
8
+ "nb_max_frames": 3000,
9
+ "padding_side": "right",
10
+ "padding_value": 0.0,
11
+ "processor_class": "WhisperProcessor",
12
+ "return_attention_mask": false,
13
+ "sampling_rate": 16000
14
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fec3db125f621cbf7f13b1749d3385c0033cd9b7b3e5ef878e6481cf9b4e044
3
+ size 4399