mirodil commited on
Commit
a1e206c
1 Parent(s): e22b495

End of training

Browse files
README.md CHANGED
@@ -16,6 +16,15 @@ should probably proofread and complete it, then remove this comment. -->
16
  # whisper-tiny-en-us-minds14
17
 
18
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
 
 
 
 
 
 
 
 
 
19
 
20
  ## Model description
21
 
@@ -41,7 +50,7 @@ The following hyperparameters were used during training:
41
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
  - lr_scheduler_type: constant_with_warmup
43
  - lr_scheduler_warmup_steps: 50
44
- - training_steps: 500
45
 
46
  ### Framework versions
47
 
 
16
  # whisper-tiny-en-us-minds14
17
 
18
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - eval_loss: 0.3247
21
+ - eval_wer_ortho: 36.7057
22
+ - eval_wer: 36.6588
23
+ - eval_runtime: 20.5021
24
+ - eval_samples_per_second: 5.512
25
+ - eval_steps_per_second: 0.39
26
+ - epoch: 107.14
27
+ - step: 3000
28
 
29
  ## Model description
30
 
 
50
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
  - lr_scheduler_type: constant_with_warmup
52
  - lr_scheduler_warmup_steps: 50
53
+ - training_steps: 4000
54
 
55
  ### Framework versions
56
 
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "openai/whisper-tiny.en",
3
  "activation_dropout": 0.0,
4
  "activation_function": "gelu",
5
  "apply_spec_augment": false,
@@ -9,7 +9,7 @@
9
  "attention_dropout": 0.0,
10
  "begin_suppress_tokens": [
11
  220,
12
- 50256
13
  ],
14
  "bos_token_id": 50257,
15
  "classifier_proj_size": 256,
@@ -18,17 +18,25 @@
18
  "decoder_ffn_dim": 1536,
19
  "decoder_layerdrop": 0.0,
20
  "decoder_layers": 4,
21
- "decoder_start_token_id": 50257,
22
  "dropout": 0.0,
23
  "encoder_attention_heads": 6,
24
  "encoder_ffn_dim": 1536,
25
  "encoder_layerdrop": 0.0,
26
  "encoder_layers": 4,
27
- "eos_token_id": 50256,
28
  "forced_decoder_ids": [
29
  [
30
  1,
31
- 50362
 
 
 
 
 
 
 
 
32
  ]
33
  ],
34
  "init_std": 0.02,
@@ -46,7 +54,7 @@
46
  "model_type": "whisper",
47
  "num_hidden_layers": 4,
48
  "num_mel_bins": 80,
49
- "pad_token_id": 50256,
50
  "scale_embedding": false,
51
  "suppress_tokens": [
52
  1,
@@ -72,77 +80,75 @@
72
  91,
73
  92,
74
  93,
75
- 357,
76
- 366,
77
- 438,
78
- 532,
79
- 685,
80
- 705,
81
- 796,
82
- 930,
83
- 1058,
84
- 1220,
85
- 1267,
86
- 1279,
87
- 1303,
88
- 1343,
89
- 1377,
90
- 1391,
91
- 1635,
92
- 1782,
93
- 1875,
94
- 2162,
95
- 2361,
96
- 2488,
97
- 3467,
98
- 4008,
99
- 4211,
100
- 4600,
101
- 4808,
102
- 5299,
103
- 5855,
104
- 6329,
105
- 7203,
106
- 9609,
107
- 9959,
108
- 10563,
109
- 10786,
110
- 11420,
111
- 11709,
112
- 11907,
113
- 13163,
114
- 13697,
115
- 13700,
116
- 14808,
117
- 15306,
118
- 16410,
119
- 16791,
120
- 17992,
121
- 19203,
122
- 19510,
123
- 20724,
124
- 22305,
125
- 22935,
126
- 27007,
127
- 30109,
128
- 30420,
129
- 33409,
130
- 34949,
131
- 40283,
132
- 40493,
133
- 40549,
134
- 47282,
135
- 49146,
136
- 50257,
137
- 50357,
138
  50358,
139
  50359,
140
  50360,
141
- 50361
 
142
  ],
143
  "torch_dtype": "float32",
144
  "transformers_version": "4.35.1",
145
  "use_cache": false,
146
  "use_weighted_layer_sum": false,
147
- "vocab_size": 51864
148
  }
 
1
  {
2
+ "_name_or_path": "openai/whisper-tiny",
3
  "activation_dropout": 0.0,
4
  "activation_function": "gelu",
5
  "apply_spec_augment": false,
 
9
  "attention_dropout": 0.0,
10
  "begin_suppress_tokens": [
11
  220,
12
+ 50257
13
  ],
14
  "bos_token_id": 50257,
15
  "classifier_proj_size": 256,
 
18
  "decoder_ffn_dim": 1536,
19
  "decoder_layerdrop": 0.0,
20
  "decoder_layers": 4,
21
+ "decoder_start_token_id": 50258,
22
  "dropout": 0.0,
23
  "encoder_attention_heads": 6,
24
  "encoder_ffn_dim": 1536,
25
  "encoder_layerdrop": 0.0,
26
  "encoder_layers": 4,
27
+ "eos_token_id": 50257,
28
  "forced_decoder_ids": [
29
  [
30
  1,
31
+ 50259
32
+ ],
33
+ [
34
+ 2,
35
+ 50359
36
+ ],
37
+ [
38
+ 3,
39
+ 50363
40
  ]
41
  ],
42
  "init_std": 0.02,
 
54
  "model_type": "whisper",
55
  "num_hidden_layers": 4,
56
  "num_mel_bins": 80,
57
+ "pad_token_id": 50257,
58
  "scale_embedding": false,
59
  "suppress_tokens": [
60
  1,
 
80
  91,
81
  92,
82
  93,
83
+ 359,
84
+ 503,
85
+ 522,
86
+ 542,
87
+ 873,
88
+ 893,
89
+ 902,
90
+ 918,
91
+ 922,
92
+ 931,
93
+ 1350,
94
+ 1853,
95
+ 1982,
96
+ 2460,
97
+ 2627,
98
+ 3246,
99
+ 3253,
100
+ 3268,
101
+ 3536,
102
+ 3846,
103
+ 3961,
104
+ 4183,
105
+ 4667,
106
+ 6585,
107
+ 6647,
108
+ 7273,
109
+ 9061,
110
+ 9383,
111
+ 10428,
112
+ 10929,
113
+ 11938,
114
+ 12033,
115
+ 12331,
116
+ 12562,
117
+ 13793,
118
+ 14157,
119
+ 14635,
120
+ 15265,
121
+ 15618,
122
+ 16553,
123
+ 16604,
124
+ 18362,
125
+ 18956,
126
+ 20075,
127
+ 21675,
128
+ 22520,
129
+ 26130,
130
+ 26161,
131
+ 26435,
132
+ 28279,
133
+ 29464,
134
+ 31650,
135
+ 32302,
136
+ 32470,
137
+ 36865,
138
+ 42863,
139
+ 47425,
140
+ 49870,
141
+ 50254,
142
+ 50258,
 
 
 
143
  50358,
144
  50359,
145
  50360,
146
+ 50361,
147
+ 50362
148
  ],
149
  "torch_dtype": "float32",
150
  "transformers_version": "4.35.1",
151
  "use_cache": false,
152
  "use_weighted_layer_sum": false,
153
+ "vocab_size": 51865
154
  }
generation_config.json CHANGED
@@ -1,25 +1,13 @@
1
  {
2
  "alignment_heads": [
3
- [
4
- 1,
5
- 0
6
- ],
7
- [
8
- 2,
9
- 0
10
- ],
11
  [
12
  2,
13
- 5
14
  ],
15
  [
16
  3,
17
  0
18
  ],
19
- [
20
- 3,
21
- 1
22
- ],
23
  [
24
  3,
25
  2
@@ -31,26 +19,140 @@
31
  [
32
  3,
33
  4
 
 
 
 
34
  ]
35
  ],
36
  "begin_suppress_tokens": [
37
  220,
38
- 50256
39
  ],
40
  "bos_token_id": 50257,
41
- "decoder_start_token_id": 50257,
42
- "eos_token_id": 50256,
43
  "forced_decoder_ids": [
44
  [
45
  1,
46
- 50362
 
 
 
 
 
 
 
 
47
  ]
48
  ],
49
- "is_multilingual": false,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
50
  "max_initial_timestamp_index": 1,
51
  "max_length": 448,
52
- "no_timestamps_token_id": 50362,
53
- "pad_token_id": 50256,
54
  "return_timestamps": false,
55
  "suppress_tokens": [
56
  1,
@@ -76,73 +178,76 @@
76
  91,
77
  92,
78
  93,
79
- 357,
80
- 366,
81
- 438,
82
- 532,
83
- 685,
84
- 705,
85
- 796,
86
- 930,
87
- 1058,
88
- 1220,
89
- 1267,
90
- 1279,
91
- 1303,
92
- 1343,
93
- 1377,
94
- 1391,
95
- 1635,
96
- 1782,
97
- 1875,
98
- 2162,
99
- 2361,
100
- 2488,
101
- 3467,
102
- 4008,
103
- 4211,
104
- 4600,
105
- 4808,
106
- 5299,
107
- 5855,
108
- 6329,
109
- 7203,
110
- 9609,
111
- 9959,
112
- 10563,
113
- 10786,
114
- 11420,
115
- 11709,
116
- 11907,
117
- 13163,
118
- 13697,
119
- 13700,
120
- 14808,
121
- 15306,
122
- 16410,
123
- 16791,
124
- 17992,
125
- 19203,
126
- 19510,
127
- 20724,
128
- 22305,
129
- 22935,
130
- 27007,
131
- 30109,
132
- 30420,
133
- 33409,
134
- 34949,
135
- 40283,
136
- 40493,
137
- 40549,
138
- 47282,
139
- 49146,
140
- 50257,
141
- 50357,
142
  50358,
143
  50359,
144
  50360,
145
- 50361
 
146
  ],
 
 
 
 
 
147
  "transformers_version": "4.35.1"
148
  }
 
1
  {
2
  "alignment_heads": [
 
 
 
 
 
 
 
 
3
  [
4
  2,
5
+ 2
6
  ],
7
  [
8
  3,
9
  0
10
  ],
 
 
 
 
11
  [
12
  3,
13
  2
 
19
  [
20
  3,
21
  4
22
+ ],
23
+ [
24
+ 3,
25
+ 5
26
  ]
27
  ],
28
  "begin_suppress_tokens": [
29
  220,
30
+ 50257
31
  ],
32
  "bos_token_id": 50257,
33
+ "decoder_start_token_id": 50258,
34
+ "eos_token_id": 50257,
35
  "forced_decoder_ids": [
36
  [
37
  1,
38
+ 50259
39
+ ],
40
+ [
41
+ 2,
42
+ 50359
43
+ ],
44
+ [
45
+ 3,
46
+ 50363
47
  ]
48
  ],
49
+ "is_multilingual": true,
50
+ "lang_to_id": {
51
+ "<|af|>": 50327,
52
+ "<|am|>": 50334,
53
+ "<|ar|>": 50272,
54
+ "<|as|>": 50350,
55
+ "<|az|>": 50304,
56
+ "<|ba|>": 50355,
57
+ "<|be|>": 50330,
58
+ "<|bg|>": 50292,
59
+ "<|bn|>": 50302,
60
+ "<|bo|>": 50347,
61
+ "<|br|>": 50309,
62
+ "<|bs|>": 50315,
63
+ "<|ca|>": 50270,
64
+ "<|cs|>": 50283,
65
+ "<|cy|>": 50297,
66
+ "<|da|>": 50285,
67
+ "<|de|>": 50261,
68
+ "<|el|>": 50281,
69
+ "<|en|>": 50259,
70
+ "<|es|>": 50262,
71
+ "<|et|>": 50307,
72
+ "<|eu|>": 50310,
73
+ "<|fa|>": 50300,
74
+ "<|fi|>": 50277,
75
+ "<|fo|>": 50338,
76
+ "<|fr|>": 50265,
77
+ "<|gl|>": 50319,
78
+ "<|gu|>": 50333,
79
+ "<|haw|>": 50352,
80
+ "<|ha|>": 50354,
81
+ "<|he|>": 50279,
82
+ "<|hi|>": 50276,
83
+ "<|hr|>": 50291,
84
+ "<|ht|>": 50339,
85
+ "<|hu|>": 50286,
86
+ "<|hy|>": 50312,
87
+ "<|id|>": 50275,
88
+ "<|is|>": 50311,
89
+ "<|it|>": 50274,
90
+ "<|ja|>": 50266,
91
+ "<|jw|>": 50356,
92
+ "<|ka|>": 50329,
93
+ "<|kk|>": 50316,
94
+ "<|km|>": 50323,
95
+ "<|kn|>": 50306,
96
+ "<|ko|>": 50264,
97
+ "<|la|>": 50294,
98
+ "<|lb|>": 50345,
99
+ "<|ln|>": 50353,
100
+ "<|lo|>": 50336,
101
+ "<|lt|>": 50293,
102
+ "<|lv|>": 50301,
103
+ "<|mg|>": 50349,
104
+ "<|mi|>": 50295,
105
+ "<|mk|>": 50308,
106
+ "<|ml|>": 50296,
107
+ "<|mn|>": 50314,
108
+ "<|mr|>": 50320,
109
+ "<|ms|>": 50282,
110
+ "<|mt|>": 50343,
111
+ "<|my|>": 50346,
112
+ "<|ne|>": 50313,
113
+ "<|nl|>": 50271,
114
+ "<|nn|>": 50342,
115
+ "<|no|>": 50288,
116
+ "<|oc|>": 50328,
117
+ "<|pa|>": 50321,
118
+ "<|pl|>": 50269,
119
+ "<|ps|>": 50340,
120
+ "<|pt|>": 50267,
121
+ "<|ro|>": 50284,
122
+ "<|ru|>": 50263,
123
+ "<|sa|>": 50344,
124
+ "<|sd|>": 50332,
125
+ "<|si|>": 50322,
126
+ "<|sk|>": 50298,
127
+ "<|sl|>": 50305,
128
+ "<|sn|>": 50324,
129
+ "<|so|>": 50326,
130
+ "<|sq|>": 50317,
131
+ "<|sr|>": 50303,
132
+ "<|su|>": 50357,
133
+ "<|sv|>": 50273,
134
+ "<|sw|>": 50318,
135
+ "<|ta|>": 50287,
136
+ "<|te|>": 50299,
137
+ "<|tg|>": 50331,
138
+ "<|th|>": 50289,
139
+ "<|tk|>": 50341,
140
+ "<|tl|>": 50348,
141
+ "<|tr|>": 50268,
142
+ "<|tt|>": 50351,
143
+ "<|uk|>": 50280,
144
+ "<|ur|>": 50290,
145
+ "<|uz|>": 50337,
146
+ "<|vi|>": 50278,
147
+ "<|yi|>": 50335,
148
+ "<|yo|>": 50325,
149
+ "<|zh|>": 50260
150
+ },
151
+ "language": "english",
152
  "max_initial_timestamp_index": 1,
153
  "max_length": 448,
154
+ "no_timestamps_token_id": 50363,
155
+ "pad_token_id": 50257,
156
  "return_timestamps": false,
157
  "suppress_tokens": [
158
  1,
 
178
  91,
179
  92,
180
  93,
181
+ 359,
182
+ 503,
183
+ 522,
184
+ 542,
185
+ 873,
186
+ 893,
187
+ 902,
188
+ 918,
189
+ 922,
190
+ 931,
191
+ 1350,
192
+ 1853,
193
+ 1982,
194
+ 2460,
195
+ 2627,
196
+ 3246,
197
+ 3253,
198
+ 3268,
199
+ 3536,
200
+ 3846,
201
+ 3961,
202
+ 4183,
203
+ 4667,
204
+ 6585,
205
+ 6647,
206
+ 7273,
207
+ 9061,
208
+ 9383,
209
+ 10428,
210
+ 10929,
211
+ 11938,
212
+ 12033,
213
+ 12331,
214
+ 12562,
215
+ 13793,
216
+ 14157,
217
+ 14635,
218
+ 15265,
219
+ 15618,
220
+ 16553,
221
+ 16604,
222
+ 18362,
223
+ 18956,
224
+ 20075,
225
+ 21675,
226
+ 22520,
227
+ 26130,
228
+ 26161,
229
+ 26435,
230
+ 28279,
231
+ 29464,
232
+ 31650,
233
+ 32302,
234
+ 32470,
235
+ 36865,
236
+ 42863,
237
+ 47425,
238
+ 49870,
239
+ 50254,
240
+ 50258,
 
 
 
241
  50358,
242
  50359,
243
  50360,
244
+ 50361,
245
+ 50362
246
  ],
247
+ "task": "transcribe",
248
+ "task_to_id": {
249
+ "transcribe": 50359,
250
+ "translate": 50358
251
+ },
252
  "transformers_version": "4.35.1"
253
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:831f62443de237150ec851f27a0192427a8bab634857d9cd2b454d9db7b35541
3
- size 151060136
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:027b3d66f24afd94d50350ebcaf9a989ceb6fd48ec52f95284c4c7fa87e79eb3
3
+ size 151061672
runs/Nov22_22-06-34_L67DDV9G7R/events.out.tfevents.1700719599.L67DDV9G7R.91939.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d777a4dbca6586fcdcbff18e1cf52dda3a67437f74d1572e6be7ee5622edc8f
3
+ size 29293
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cfeb795886834490b18828520c7d8f0b81dd70b83ca6caa73a161a47e7902531
3
  size 4728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7568170c03c6d5eb0ce47f667fa952d9d0ccd6264480ee48c53a163c07951d58
3
  size 4728