zongxiao commited on
Commit
c56c1e8
1 Parent(s): 7ac9004

Training in progress, step 100

Browse files
Files changed (5) hide show
  1. README.md +80 -0
  2. config.json +1 -1
  3. generation_config.json +269 -0
  4. pytorch_model.bin +1 -1
  5. training_args.bin +1 -1
README.md ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: openai/whisper-tiny
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - PolyAI/minds14
8
+ metrics:
9
+ - wer
10
+ model-index:
11
+ - name: whisper-small-US-2
12
+ results:
13
+ - task:
14
+ name: Automatic Speech Recognition
15
+ type: automatic-speech-recognition
16
+ dataset:
17
+ name: PolyAI/minds14
18
+ type: PolyAI/minds14
19
+ config: en-US
20
+ split: train
21
+ args: en-US
22
+ metrics:
23
+ - name: Wer
24
+ type: wer
25
+ value: 0.04095185390149419
26
+ ---
27
+
28
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
29
+ should probably proofread and complete it, then remove this comment. -->
30
+
31
+ # whisper-small-US-2
32
+
33
+ This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
34
+ It achieves the following results on the evaluation set:
35
+ - Loss: 0.1118
36
+ - Wer Ortho: 0.0387
37
+ - Wer: 0.0410
38
+
39
+ ## Model description
40
+
41
+ More information needed
42
+
43
+ ## Intended uses & limitations
44
+
45
+ More information needed
46
+
47
+ ## Training and evaluation data
48
+
49
+ More information needed
50
+
51
+ ## Training procedure
52
+
53
+ ### Training hyperparameters
54
+
55
+ The following hyperparameters were used during training:
56
+ - learning_rate: 5e-06
57
+ - train_batch_size: 4
58
+ - eval_batch_size: 8
59
+ - seed: 42
60
+ - gradient_accumulation_steps: 4
61
+ - total_train_batch_size: 16
62
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
63
+ - lr_scheduler_type: constant_with_warmup
64
+ - lr_scheduler_warmup_steps: 50
65
+ - training_steps: 200
66
+
67
+ ### Training results
68
+
69
+ | Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
70
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|
71
+ | 0.0183 | 3.57 | 100 | 0.1015 | 0.0398 | 0.0426 |
72
+ | 0.0003 | 7.14 | 200 | 0.1118 | 0.0387 | 0.0410 |
73
+
74
+
75
+ ### Framework versions
76
+
77
+ - Transformers 4.33.3
78
+ - Pytorch 2.0.1
79
+ - Datasets 2.14.5
80
+ - Tokenizers 0.13.3
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "/home/danfouer/Videos/HFAudio/whisper-small-US/checkpoint-500",
3
  "activation_dropout": 0.0,
4
  "activation_function": "gelu",
5
  "apply_spec_augment": false,
 
1
  {
2
+ "_name_or_path": "/home/danfouer/Videos/HFAudio/whisper-small-US-2/checkpoint-200",
3
  "activation_dropout": 0.0,
4
  "activation_function": "gelu",
5
  "apply_spec_augment": false,
generation_config.json ADDED
@@ -0,0 +1,269 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 5,
5
+ 3
6
+ ],
7
+ [
8
+ 5,
9
+ 9
10
+ ],
11
+ [
12
+ 8,
13
+ 0
14
+ ],
15
+ [
16
+ 8,
17
+ 4
18
+ ],
19
+ [
20
+ 8,
21
+ 7
22
+ ],
23
+ [
24
+ 8,
25
+ 8
26
+ ],
27
+ [
28
+ 9,
29
+ 0
30
+ ],
31
+ [
32
+ 9,
33
+ 7
34
+ ],
35
+ [
36
+ 9,
37
+ 9
38
+ ],
39
+ [
40
+ 10,
41
+ 5
42
+ ]
43
+ ],
44
+ "begin_suppress_tokens": [
45
+ 220,
46
+ 50257
47
+ ],
48
+ "bos_token_id": 50257,
49
+ "decoder_start_token_id": 50258,
50
+ "eos_token_id": 50257,
51
+ "forced_decoder_ids": [
52
+ [
53
+ 1,
54
+ 50259
55
+ ],
56
+ [
57
+ 2,
58
+ 50359
59
+ ],
60
+ [
61
+ 3,
62
+ 50363
63
+ ]
64
+ ],
65
+ "is_multilingual": true,
66
+ "lang_to_id": {
67
+ "<|af|>": 50327,
68
+ "<|am|>": 50334,
69
+ "<|ar|>": 50272,
70
+ "<|as|>": 50350,
71
+ "<|az|>": 50304,
72
+ "<|ba|>": 50355,
73
+ "<|be|>": 50330,
74
+ "<|bg|>": 50292,
75
+ "<|bn|>": 50302,
76
+ "<|bo|>": 50347,
77
+ "<|br|>": 50309,
78
+ "<|bs|>": 50315,
79
+ "<|ca|>": 50270,
80
+ "<|cs|>": 50283,
81
+ "<|cy|>": 50297,
82
+ "<|da|>": 50285,
83
+ "<|de|>": 50261,
84
+ "<|el|>": 50281,
85
+ "<|en|>": 50259,
86
+ "<|es|>": 50262,
87
+ "<|et|>": 50307,
88
+ "<|eu|>": 50310,
89
+ "<|fa|>": 50300,
90
+ "<|fi|>": 50277,
91
+ "<|fo|>": 50338,
92
+ "<|fr|>": 50265,
93
+ "<|gl|>": 50319,
94
+ "<|gu|>": 50333,
95
+ "<|haw|>": 50352,
96
+ "<|ha|>": 50354,
97
+ "<|he|>": 50279,
98
+ "<|hi|>": 50276,
99
+ "<|hr|>": 50291,
100
+ "<|ht|>": 50339,
101
+ "<|hu|>": 50286,
102
+ "<|hy|>": 50312,
103
+ "<|id|>": 50275,
104
+ "<|is|>": 50311,
105
+ "<|it|>": 50274,
106
+ "<|ja|>": 50266,
107
+ "<|jw|>": 50356,
108
+ "<|ka|>": 50329,
109
+ "<|kk|>": 50316,
110
+ "<|km|>": 50323,
111
+ "<|kn|>": 50306,
112
+ "<|ko|>": 50264,
113
+ "<|la|>": 50294,
114
+ "<|lb|>": 50345,
115
+ "<|ln|>": 50353,
116
+ "<|lo|>": 50336,
117
+ "<|lt|>": 50293,
118
+ "<|lv|>": 50301,
119
+ "<|mg|>": 50349,
120
+ "<|mi|>": 50295,
121
+ "<|mk|>": 50308,
122
+ "<|ml|>": 50296,
123
+ "<|mn|>": 50314,
124
+ "<|mr|>": 50320,
125
+ "<|ms|>": 50282,
126
+ "<|mt|>": 50343,
127
+ "<|my|>": 50346,
128
+ "<|ne|>": 50313,
129
+ "<|nl|>": 50271,
130
+ "<|nn|>": 50342,
131
+ "<|no|>": 50288,
132
+ "<|oc|>": 50328,
133
+ "<|pa|>": 50321,
134
+ "<|pl|>": 50269,
135
+ "<|ps|>": 50340,
136
+ "<|pt|>": 50267,
137
+ "<|ro|>": 50284,
138
+ "<|ru|>": 50263,
139
+ "<|sa|>": 50344,
140
+ "<|sd|>": 50332,
141
+ "<|si|>": 50322,
142
+ "<|sk|>": 50298,
143
+ "<|sl|>": 50305,
144
+ "<|sn|>": 50324,
145
+ "<|so|>": 50326,
146
+ "<|sq|>": 50317,
147
+ "<|sr|>": 50303,
148
+ "<|su|>": 50357,
149
+ "<|sv|>": 50273,
150
+ "<|sw|>": 50318,
151
+ "<|ta|>": 50287,
152
+ "<|te|>": 50299,
153
+ "<|tg|>": 50331,
154
+ "<|th|>": 50289,
155
+ "<|tk|>": 50341,
156
+ "<|tl|>": 50348,
157
+ "<|tr|>": 50268,
158
+ "<|tt|>": 50351,
159
+ "<|uk|>": 50280,
160
+ "<|ur|>": 50290,
161
+ "<|uz|>": 50337,
162
+ "<|vi|>": 50278,
163
+ "<|yi|>": 50335,
164
+ "<|yo|>": 50325,
165
+ "<|zh|>": 50260
166
+ },
167
+ "language": "english",
168
+ "max_initial_timestamp_index": 1,
169
+ "max_length": 448,
170
+ "no_timestamps_token_id": 50363,
171
+ "pad_token_id": 50257,
172
+ "return_timestamps": false,
173
+ "suppress_tokens": [
174
+ 1,
175
+ 2,
176
+ 7,
177
+ 8,
178
+ 9,
179
+ 10,
180
+ 14,
181
+ 25,
182
+ 26,
183
+ 27,
184
+ 28,
185
+ 29,
186
+ 31,
187
+ 58,
188
+ 59,
189
+ 60,
190
+ 61,
191
+ 62,
192
+ 63,
193
+ 90,
194
+ 91,
195
+ 92,
196
+ 93,
197
+ 359,
198
+ 503,
199
+ 522,
200
+ 542,
201
+ 873,
202
+ 893,
203
+ 902,
204
+ 918,
205
+ 922,
206
+ 931,
207
+ 1350,
208
+ 1853,
209
+ 1982,
210
+ 2460,
211
+ 2627,
212
+ 3246,
213
+ 3253,
214
+ 3268,
215
+ 3536,
216
+ 3846,
217
+ 3961,
218
+ 4183,
219
+ 4667,
220
+ 6585,
221
+ 6647,
222
+ 7273,
223
+ 9061,
224
+ 9383,
225
+ 10428,
226
+ 10929,
227
+ 11938,
228
+ 12033,
229
+ 12331,
230
+ 12562,
231
+ 13793,
232
+ 14157,
233
+ 14635,
234
+ 15265,
235
+ 15618,
236
+ 16553,
237
+ 16604,
238
+ 18362,
239
+ 18956,
240
+ 20075,
241
+ 21675,
242
+ 22520,
243
+ 26130,
244
+ 26161,
245
+ 26435,
246
+ 28279,
247
+ 29464,
248
+ 31650,
249
+ 32302,
250
+ 32470,
251
+ 36865,
252
+ 42863,
253
+ 47425,
254
+ 49870,
255
+ 50254,
256
+ 50258,
257
+ 50358,
258
+ 50359,
259
+ 50360,
260
+ 50361,
261
+ 50362
262
+ ],
263
+ "task": "transcribe",
264
+ "task_to_id": {
265
+ "transcribe": 50359,
266
+ "translate": 50358
267
+ },
268
+ "transformers_version": "4.33.3"
269
+ }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e97edb0a6121bc7f10e6d96186c09308523c1ee6fa8723d2ee081391f69c49ea
3
  size 967102729
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:606a954a7374b10bdc271d4c0680e3f4346aa26554f546ea142e913d3e898519
3
  size 967102729
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:85e9711b4194fa7080a1552246cbc6cbaf61c9acebeea5735a9fe526de700ca6
3
  size 4219
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a64a73edb443046fd8e2411c3c81bbad2b33ba111950fc5049fa1fb41c08801
3
  size 4219