davidilag commited on
Commit
3ec5503
1 Parent(s): 3f3daa9

End of training

Browse files
README.md ADDED
@@ -0,0 +1,107 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: openai/whisper-large
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - ravnursson_asr
8
+ metrics:
9
+ - wer
10
+ model-index:
11
+ - name: whisper-large-fo-100h-30k-steps
12
+ results:
13
+ - task:
14
+ name: Automatic Speech Recognition
15
+ type: automatic-speech-recognition
16
+ dataset:
17
+ name: ravnursson_asr
18
+ type: ravnursson_asr
19
+ config: ravnursson_asr
20
+ split: test
21
+ args: ravnursson_asr
22
+ metrics:
23
+ - name: Wer
24
+ type: wer
25
+ value: 4.957720958324945
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
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/setur/huggingface/runs/woejhwzd)
32
+ # whisper-large-fo-100h-30k-steps
33
+
34
+ This model is a fine-tuned version of [openai/whisper-large](https://huggingface.co/openai/whisper-large) on the ravnursson_asr dataset.
35
+ It achieves the following results on the evaluation set:
36
+ - Loss: 0.0872
37
+ - Wer: 4.9577
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: 1e-05
57
+ - train_batch_size: 16
58
+ - eval_batch_size: 8
59
+ - seed: 42
60
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
+ - lr_scheduler_type: linear
62
+ - lr_scheduler_warmup_steps: 500
63
+ - training_steps: 30000
64
+ - mixed_precision_training: Native AMP
65
+
66
+ ### Training results
67
+
68
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
69
+ |:-------------:|:------:|:-----:|:---------------:|:-------:|
70
+ | 0.2261 | 0.2320 | 1000 | 0.2668 | 20.1379 |
71
+ | 0.1577 | 0.4640 | 2000 | 0.1840 | 15.0997 |
72
+ | 0.1205 | 0.6961 | 3000 | 0.1456 | 11.9489 |
73
+ | 0.1151 | 0.9281 | 4000 | 0.1300 | 10.6906 |
74
+ | 0.0457 | 1.1601 | 5000 | 0.1241 | 9.7745 |
75
+ | 0.0423 | 1.3921 | 6000 | 0.1221 | 9.4876 |
76
+ | 0.0428 | 1.6241 | 7000 | 0.1080 | 8.4709 |
77
+ | 0.0486 | 1.8561 | 8000 | 0.1053 | 8.5011 |
78
+ | 0.0205 | 2.0882 | 9000 | 0.1014 | 7.4643 |
79
+ | 0.0184 | 2.3202 | 10000 | 0.1003 | 8.1387 |
80
+ | 0.0165 | 2.5522 | 11000 | 0.0969 | 7.1472 |
81
+ | 0.025 | 2.7842 | 12000 | 0.0907 | 6.8804 |
82
+ | 0.0048 | 3.0162 | 13000 | 0.0936 | 6.9005 |
83
+ | 0.0092 | 3.2483 | 14000 | 0.0923 | 6.7244 |
84
+ | 0.006 | 3.4803 | 15000 | 0.0921 | 6.3519 |
85
+ | 0.0095 | 3.7123 | 16000 | 0.0922 | 6.3821 |
86
+ | 0.0089 | 3.9443 | 17000 | 0.0929 | 6.3771 |
87
+ | 0.0023 | 4.1763 | 18000 | 0.0915 | 6.0650 |
88
+ | 0.0033 | 4.4084 | 19000 | 0.0924 | 5.9543 |
89
+ | 0.0028 | 4.6404 | 20000 | 0.0909 | 5.9040 |
90
+ | 0.0021 | 4.8724 | 21000 | 0.0884 | 5.7328 |
91
+ | 0.002 | 5.1044 | 22000 | 0.0874 | 5.4057 |
92
+ | 0.0008 | 5.3364 | 23000 | 0.0890 | 5.3654 |
93
+ | 0.0005 | 5.5684 | 24000 | 0.0857 | 5.2597 |
94
+ | 0.002 | 5.8005 | 25000 | 0.0860 | 5.2144 |
95
+ | 0.0007 | 6.0325 | 26000 | 0.0873 | 5.1842 |
96
+ | 0.0002 | 6.2645 | 27000 | 0.0850 | 4.9879 |
97
+ | 0.001 | 6.4965 | 28000 | 0.0889 | 4.9376 |
98
+ | 0.0001 | 6.7285 | 29000 | 0.0878 | 5.0081 |
99
+ | 0.0003 | 6.9606 | 30000 | 0.0872 | 4.9577 |
100
+
101
+
102
+ ### Framework versions
103
+
104
+ - Transformers 4.42.4
105
+ - Pytorch 2.3.1+cu121
106
+ - Datasets 2.20.0
107
+ - Tokenizers 0.19.1
generation_config.json ADDED
@@ -0,0 +1,252 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 9,
5
+ 19
6
+ ],
7
+ [
8
+ 11,
9
+ 2
10
+ ],
11
+ [
12
+ 11,
13
+ 4
14
+ ],
15
+ [
16
+ 11,
17
+ 17
18
+ ],
19
+ [
20
+ 22,
21
+ 7
22
+ ],
23
+ [
24
+ 22,
25
+ 11
26
+ ],
27
+ [
28
+ 22,
29
+ 17
30
+ ],
31
+ [
32
+ 23,
33
+ 2
34
+ ],
35
+ [
36
+ 23,
37
+ 15
38
+ ]
39
+ ],
40
+ "begin_suppress_tokens": [
41
+ 220,
42
+ 50257
43
+ ],
44
+ "bos_token_id": 50257,
45
+ "decoder_start_token_id": 50258,
46
+ "eos_token_id": 50257,
47
+ "is_multilingual": true,
48
+ "lang_to_id": {
49
+ "<|af|>": 50327,
50
+ "<|am|>": 50334,
51
+ "<|ar|>": 50272,
52
+ "<|as|>": 50350,
53
+ "<|az|>": 50304,
54
+ "<|ba|>": 50355,
55
+ "<|be|>": 50330,
56
+ "<|bg|>": 50292,
57
+ "<|bn|>": 50302,
58
+ "<|bo|>": 50347,
59
+ "<|br|>": 50309,
60
+ "<|bs|>": 50315,
61
+ "<|ca|>": 50270,
62
+ "<|cs|>": 50283,
63
+ "<|cy|>": 50297,
64
+ "<|da|>": 50285,
65
+ "<|de|>": 50261,
66
+ "<|el|>": 50281,
67
+ "<|en|>": 50259,
68
+ "<|es|>": 50262,
69
+ "<|et|>": 50307,
70
+ "<|eu|>": 50310,
71
+ "<|fa|>": 50300,
72
+ "<|fi|>": 50277,
73
+ "<|fo|>": 50338,
74
+ "<|fr|>": 50265,
75
+ "<|gl|>": 50319,
76
+ "<|gu|>": 50333,
77
+ "<|haw|>": 50352,
78
+ "<|ha|>": 50354,
79
+ "<|he|>": 50279,
80
+ "<|hi|>": 50276,
81
+ "<|hr|>": 50291,
82
+ "<|ht|>": 50339,
83
+ "<|hu|>": 50286,
84
+ "<|hy|>": 50312,
85
+ "<|id|>": 50275,
86
+ "<|is|>": 50311,
87
+ "<|it|>": 50274,
88
+ "<|ja|>": 50266,
89
+ "<|jw|>": 50356,
90
+ "<|ka|>": 50329,
91
+ "<|kk|>": 50316,
92
+ "<|km|>": 50323,
93
+ "<|kn|>": 50306,
94
+ "<|ko|>": 50264,
95
+ "<|la|>": 50294,
96
+ "<|lb|>": 50345,
97
+ "<|ln|>": 50353,
98
+ "<|lo|>": 50336,
99
+ "<|lt|>": 50293,
100
+ "<|lv|>": 50301,
101
+ "<|mg|>": 50349,
102
+ "<|mi|>": 50295,
103
+ "<|mk|>": 50308,
104
+ "<|ml|>": 50296,
105
+ "<|mn|>": 50314,
106
+ "<|mr|>": 50320,
107
+ "<|ms|>": 50282,
108
+ "<|mt|>": 50343,
109
+ "<|my|>": 50346,
110
+ "<|ne|>": 50313,
111
+ "<|nl|>": 50271,
112
+ "<|nn|>": 50342,
113
+ "<|no|>": 50288,
114
+ "<|oc|>": 50328,
115
+ "<|pa|>": 50321,
116
+ "<|pl|>": 50269,
117
+ "<|ps|>": 50340,
118
+ "<|pt|>": 50267,
119
+ "<|ro|>": 50284,
120
+ "<|ru|>": 50263,
121
+ "<|sa|>": 50344,
122
+ "<|sd|>": 50332,
123
+ "<|si|>": 50322,
124
+ "<|sk|>": 50298,
125
+ "<|sl|>": 50305,
126
+ "<|sn|>": 50324,
127
+ "<|so|>": 50326,
128
+ "<|sq|>": 50317,
129
+ "<|sr|>": 50303,
130
+ "<|su|>": 50357,
131
+ "<|sv|>": 50273,
132
+ "<|sw|>": 50318,
133
+ "<|ta|>": 50287,
134
+ "<|te|>": 50299,
135
+ "<|tg|>": 50331,
136
+ "<|th|>": 50289,
137
+ "<|tk|>": 50341,
138
+ "<|tl|>": 50348,
139
+ "<|tr|>": 50268,
140
+ "<|tt|>": 50351,
141
+ "<|uk|>": 50280,
142
+ "<|ur|>": 50290,
143
+ "<|uz|>": 50337,
144
+ "<|vi|>": 50278,
145
+ "<|yi|>": 50335,
146
+ "<|yo|>": 50325,
147
+ "<|zh|>": 50260
148
+ },
149
+ "language": "<|is|>",
150
+ "max_initial_timestamp_index": 50,
151
+ "max_length": 448,
152
+ "no_timestamps_token_id": 50363,
153
+ "pad_token_id": 50257,
154
+ "prev_sot_token_id": 50361,
155
+ "return_timestamps": false,
156
+ "suppress_tokens": [
157
+ 1,
158
+ 2,
159
+ 7,
160
+ 8,
161
+ 9,
162
+ 10,
163
+ 14,
164
+ 25,
165
+ 26,
166
+ 27,
167
+ 28,
168
+ 29,
169
+ 31,
170
+ 58,
171
+ 59,
172
+ 60,
173
+ 61,
174
+ 62,
175
+ 63,
176
+ 90,
177
+ 91,
178
+ 92,
179
+ 93,
180
+ 359,
181
+ 503,
182
+ 522,
183
+ 542,
184
+ 873,
185
+ 893,
186
+ 902,
187
+ 918,
188
+ 922,
189
+ 931,
190
+ 1350,
191
+ 1853,
192
+ 1982,
193
+ 2460,
194
+ 2627,
195
+ 3246,
196
+ 3253,
197
+ 3268,
198
+ 3536,
199
+ 3846,
200
+ 3961,
201
+ 4183,
202
+ 4667,
203
+ 6585,
204
+ 6647,
205
+ 7273,
206
+ 9061,
207
+ 9383,
208
+ 10428,
209
+ 10929,
210
+ 11938,
211
+ 12033,
212
+ 12331,
213
+ 12562,
214
+ 13793,
215
+ 14157,
216
+ 14635,
217
+ 15265,
218
+ 15618,
219
+ 16553,
220
+ 16604,
221
+ 18362,
222
+ 18956,
223
+ 20075,
224
+ 21675,
225
+ 22520,
226
+ 26130,
227
+ 26161,
228
+ 26435,
229
+ 28279,
230
+ 29464,
231
+ 31650,
232
+ 32302,
233
+ 32470,
234
+ 36865,
235
+ 42863,
236
+ 47425,
237
+ 49870,
238
+ 50254,
239
+ 50258,
240
+ 50358,
241
+ 50359,
242
+ 50360,
243
+ 50361,
244
+ 50362
245
+ ],
246
+ "task": "transcribe",
247
+ "task_to_id": {
248
+ "transcribe": 50359,
249
+ "translate": 50358
250
+ },
251
+ "transformers_version": "4.42.4"
252
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:36a6175ad5c5fdf3d74a375b9ab30fa64e940540b6c6ef1cd6775c85ff89c55e
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:bc343015783a5419759467f87ba51c779566e5713af44977e88b97bb72144951
3
+ size 1180663192
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff