davidilag commited on
Commit
51e7b8c
1 Parent(s): 3a8946a

End of training

Browse files
README.md ADDED
@@ -0,0 +1,107 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: NbAiLab/nb-whisper-large
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - samromur_asr
8
+ metrics:
9
+ - wer
10
+ model-index:
11
+ - name: whisper-large-no-is-145h-30k-steps
12
+ results:
13
+ - task:
14
+ name: Automatic Speech Recognition
15
+ type: automatic-speech-recognition
16
+ dataset:
17
+ name: samromur_asr
18
+ type: samromur_asr
19
+ config: samromur_asr
20
+ split: test
21
+ args: samromur_asr
22
+ metrics:
23
+ - name: Wer
24
+ type: wer
25
+ value: 8.020304568527918
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/a8dup09b)
32
+ # whisper-large-no-is-145h-30k-steps
33
+
34
+ This model is a fine-tuned version of [NbAiLab/nb-whisper-large](https://huggingface.co/NbAiLab/nb-whisper-large) on the samromur_asr dataset.
35
+ It achieves the following results on the evaluation set:
36
+ - Loss: 0.1280
37
+ - Wer: 8.0203
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
+ | 0.3237 | 0.1778 | 1000 | 0.3331 | 24.9280 |
71
+ | 0.2334 | 0.3556 | 2000 | 0.2369 | 19.2033 |
72
+ | 0.1932 | 0.5333 | 3000 | 0.2121 | 16.7584 |
73
+ | 0.1681 | 0.7111 | 4000 | 0.1874 | 14.9669 |
74
+ | 0.1487 | 0.8889 | 5000 | 0.1720 | 13.8274 |
75
+ | 0.0911 | 1.0667 | 6000 | 0.1644 | 13.2541 |
76
+ | 0.0765 | 1.2444 | 7000 | 0.1577 | 12.3332 |
77
+ | 0.0843 | 1.4222 | 8000 | 0.1469 | 11.8722 |
78
+ | 0.0797 | 1.6 | 9000 | 0.1437 | 11.3156 |
79
+ | 0.0778 | 1.7778 | 10000 | 0.1365 | 10.9298 |
80
+ | 0.0789 | 1.9556 | 11000 | 0.1329 | 10.7554 |
81
+ | 0.039 | 2.1333 | 12000 | 0.1353 | 10.3625 |
82
+ | 0.0394 | 2.3111 | 13000 | 0.1399 | 10.2872 |
83
+ | 0.0438 | 2.4889 | 14000 | 0.1357 | 10.0926 |
84
+ | 0.0365 | 2.6667 | 15000 | 0.1287 | 9.8632 |
85
+ | 0.0406 | 2.8444 | 16000 | 0.1266 | 9.6435 |
86
+ | 0.0193 | 3.0222 | 17000 | 0.1284 | 9.3258 |
87
+ | 0.0181 | 3.2 | 18000 | 0.1309 | 9.3019 |
88
+ | 0.0215 | 3.3778 | 19000 | 0.1317 | 9.1299 |
89
+ | 0.0207 | 3.5556 | 20000 | 0.1287 | 8.8767 |
90
+ | 0.0185 | 3.7333 | 21000 | 0.1285 | 8.8731 |
91
+ | 0.0162 | 3.9111 | 22000 | 0.1272 | 8.6581 |
92
+ | 0.007 | 4.0889 | 23000 | 0.1285 | 8.6175 |
93
+ | 0.0076 | 4.2667 | 24000 | 0.1296 | 8.3655 |
94
+ | 0.0074 | 4.4444 | 25000 | 0.1277 | 8.4037 |
95
+ | 0.0076 | 4.6222 | 26000 | 0.1289 | 8.3464 |
96
+ | 0.006 | 4.8 | 27000 | 0.1257 | 8.1756 |
97
+ | 0.0067 | 4.9778 | 28000 | 0.1252 | 8.0979 |
98
+ | 0.0025 | 5.1556 | 29000 | 0.1269 | 8.0322 |
99
+ | 0.0032 | 5.3333 | 30000 | 0.1280 | 8.0203 |
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 +1,256 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 7,
5
+ 0
6
+ ],
7
+ [
8
+ 10,
9
+ 17
10
+ ],
11
+ [
12
+ 12,
13
+ 18
14
+ ],
15
+ [
16
+ 13,
17
+ 12
18
+ ],
19
+ [
20
+ 16,
21
+ 1
22
+ ],
23
+ [
24
+ 17,
25
+ 14
26
+ ],
27
+ [
28
+ 19,
29
+ 11
30
+ ],
31
+ [
32
+ 21,
33
+ 4
34
+ ],
35
+ [
36
+ 24,
37
+ 1
38
+ ],
39
+ [
40
+ 25,
41
+ 6
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
+ "is_multilingual": true,
52
+ "lang_to_id": {
53
+ "<|af|>": 50327,
54
+ "<|am|>": 50334,
55
+ "<|ar|>": 50272,
56
+ "<|as|>": 50350,
57
+ "<|az|>": 50304,
58
+ "<|ba|>": 50355,
59
+ "<|be|>": 50330,
60
+ "<|bg|>": 50292,
61
+ "<|bn|>": 50302,
62
+ "<|bo|>": 50347,
63
+ "<|br|>": 50309,
64
+ "<|bs|>": 50315,
65
+ "<|ca|>": 50270,
66
+ "<|cs|>": 50283,
67
+ "<|cy|>": 50297,
68
+ "<|da|>": 50285,
69
+ "<|de|>": 50261,
70
+ "<|el|>": 50281,
71
+ "<|en|>": 50259,
72
+ "<|es|>": 50262,
73
+ "<|et|>": 50307,
74
+ "<|eu|>": 50310,
75
+ "<|fa|>": 50300,
76
+ "<|fi|>": 50277,
77
+ "<|fo|>": 50338,
78
+ "<|fr|>": 50265,
79
+ "<|gl|>": 50319,
80
+ "<|gu|>": 50333,
81
+ "<|haw|>": 50352,
82
+ "<|ha|>": 50354,
83
+ "<|he|>": 50279,
84
+ "<|hi|>": 50276,
85
+ "<|hr|>": 50291,
86
+ "<|ht|>": 50339,
87
+ "<|hu|>": 50286,
88
+ "<|hy|>": 50312,
89
+ "<|id|>": 50275,
90
+ "<|is|>": 50311,
91
+ "<|it|>": 50274,
92
+ "<|ja|>": 50266,
93
+ "<|jw|>": 50356,
94
+ "<|ka|>": 50329,
95
+ "<|kk|>": 50316,
96
+ "<|km|>": 50323,
97
+ "<|kn|>": 50306,
98
+ "<|ko|>": 50264,
99
+ "<|la|>": 50294,
100
+ "<|lb|>": 50345,
101
+ "<|ln|>": 50353,
102
+ "<|lo|>": 50336,
103
+ "<|lt|>": 50293,
104
+ "<|lv|>": 50301,
105
+ "<|mg|>": 50349,
106
+ "<|mi|>": 50295,
107
+ "<|mk|>": 50308,
108
+ "<|ml|>": 50296,
109
+ "<|mn|>": 50314,
110
+ "<|mr|>": 50320,
111
+ "<|ms|>": 50282,
112
+ "<|mt|>": 50343,
113
+ "<|my|>": 50346,
114
+ "<|ne|>": 50313,
115
+ "<|nl|>": 50271,
116
+ "<|nn|>": 50342,
117
+ "<|no|>": 50288,
118
+ "<|oc|>": 50328,
119
+ "<|pa|>": 50321,
120
+ "<|pl|>": 50269,
121
+ "<|ps|>": 50340,
122
+ "<|pt|>": 50267,
123
+ "<|ro|>": 50284,
124
+ "<|ru|>": 50263,
125
+ "<|sa|>": 50344,
126
+ "<|sd|>": 50332,
127
+ "<|si|>": 50322,
128
+ "<|sk|>": 50298,
129
+ "<|sl|>": 50305,
130
+ "<|sn|>": 50324,
131
+ "<|so|>": 50326,
132
+ "<|sq|>": 50317,
133
+ "<|sr|>": 50303,
134
+ "<|su|>": 50357,
135
+ "<|sv|>": 50273,
136
+ "<|sw|>": 50318,
137
+ "<|ta|>": 50287,
138
+ "<|te|>": 50299,
139
+ "<|tg|>": 50331,
140
+ "<|th|>": 50289,
141
+ "<|tk|>": 50341,
142
+ "<|tl|>": 50348,
143
+ "<|tr|>": 50268,
144
+ "<|tt|>": 50351,
145
+ "<|uk|>": 50280,
146
+ "<|ur|>": 50290,
147
+ "<|uz|>": 50337,
148
+ "<|vi|>": 50278,
149
+ "<|yi|>": 50335,
150
+ "<|yo|>": 50325,
151
+ "<|yue|>": 50358,
152
+ "<|zh|>": 50260
153
+ },
154
+ "language": "<|is|>",
155
+ "max_initial_timestamp_index": 1,
156
+ "max_length": 448,
157
+ "no_timestamps_token_id": 50364,
158
+ "pad_token_id": 50257,
159
+ "return_timestamps": false,
160
+ "suppress_tokens": [
161
+ 1,
162
+ 2,
163
+ 7,
164
+ 8,
165
+ 9,
166
+ 10,
167
+ 14,
168
+ 25,
169
+ 26,
170
+ 27,
171
+ 28,
172
+ 29,
173
+ 31,
174
+ 58,
175
+ 59,
176
+ 60,
177
+ 61,
178
+ 62,
179
+ 63,
180
+ 90,
181
+ 91,
182
+ 92,
183
+ 93,
184
+ 359,
185
+ 503,
186
+ 522,
187
+ 542,
188
+ 873,
189
+ 893,
190
+ 902,
191
+ 918,
192
+ 922,
193
+ 931,
194
+ 1350,
195
+ 1853,
196
+ 1982,
197
+ 2460,
198
+ 2627,
199
+ 3246,
200
+ 3253,
201
+ 3268,
202
+ 3536,
203
+ 3846,
204
+ 3961,
205
+ 4183,
206
+ 4667,
207
+ 6585,
208
+ 6647,
209
+ 7273,
210
+ 9061,
211
+ 9383,
212
+ 10428,
213
+ 10929,
214
+ 11938,
215
+ 12033,
216
+ 12331,
217
+ 12562,
218
+ 13793,
219
+ 14157,
220
+ 14635,
221
+ 15265,
222
+ 15618,
223
+ 16553,
224
+ 16604,
225
+ 18362,
226
+ 18956,
227
+ 20075,
228
+ 21675,
229
+ 22520,
230
+ 26130,
231
+ 26161,
232
+ 26435,
233
+ 28279,
234
+ 29464,
235
+ 31650,
236
+ 32302,
237
+ 32470,
238
+ 36865,
239
+ 42863,
240
+ 47425,
241
+ 49870,
242
+ 50254,
243
+ 50258,
244
+ 50359,
245
+ 50360,
246
+ 50361,
247
+ 50362,
248
+ 50363
249
+ ],
250
+ "task": "transcribe",
251
+ "task_to_id": {
252
+ "transcribe": 50360,
253
+ "translate": 50359
254
+ },
255
+ "transformers_version": "4.42.4"
256
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c131bd99fe4bb3e889e712ecf01c1fb05a0d5c2e9b2ff95d9aef3d6e044ea500
3
+ size 4993448880
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:181e752e67860caa0b4c604373c6fb6e2b9d0af4707dcbc18b46bf00f8391b87
3
+ size 1180663192
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff