Ahmed107 commited on
Commit
433c3d5
1 Parent(s): 3404db7

End of training

Browse files
Files changed (2) hide show
  1. README.md +58 -0
  2. generation_config.json +320 -0
README.md ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: openai/whisper-medium
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - nadsoft/Jordan-Audio
8
+ model-index:
9
+ - name: hamsa-medium
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # hamsa-medium
17
+
18
+ This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the nadsoft/Jordan-Audio dataset.
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
+
32
+ ## Training procedure
33
+
34
+ ### Training hyperparameters
35
+
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 1e-05
38
+ - train_batch_size: 32
39
+ - eval_batch_size: 16
40
+ - seed: 42
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: 10
45
+
46
+ ### Training results
47
+
48
+ | Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
49
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|
50
+ | No log | 0.07 | 10 | 3.0803 | 0.4207 | 0.3471 |
51
+
52
+
53
+ ### Framework versions
54
+
55
+ - Transformers 4.34.1
56
+ - Pytorch 2.1.0+cu118
57
+ - Datasets 2.14.5
58
+ - Tokenizers 0.14.1
generation_config.json ADDED
@@ -0,0 +1,320 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ 3,
114
+ 50363
115
+ ]
116
+ ],
117
+ "is_multilingual": true,
118
+ "lang_to_id": {
119
+ "<|af|>": 50327,
120
+ "<|am|>": 50334,
121
+ "<|ar|>": 50272,
122
+ "<|as|>": 50350,
123
+ "<|az|>": 50304,
124
+ "<|ba|>": 50355,
125
+ "<|be|>": 50330,
126
+ "<|bg|>": 50292,
127
+ "<|bn|>": 50302,
128
+ "<|bo|>": 50347,
129
+ "<|br|>": 50309,
130
+ "<|bs|>": 50315,
131
+ "<|ca|>": 50270,
132
+ "<|cs|>": 50283,
133
+ "<|cy|>": 50297,
134
+ "<|da|>": 50285,
135
+ "<|de|>": 50261,
136
+ "<|el|>": 50281,
137
+ "<|en|>": 50259,
138
+ "<|es|>": 50262,
139
+ "<|et|>": 50307,
140
+ "<|eu|>": 50310,
141
+ "<|fa|>": 50300,
142
+ "<|fi|>": 50277,
143
+ "<|fo|>": 50338,
144
+ "<|fr|>": 50265,
145
+ "<|gl|>": 50319,
146
+ "<|gu|>": 50333,
147
+ "<|haw|>": 50352,
148
+ "<|ha|>": 50354,
149
+ "<|he|>": 50279,
150
+ "<|hi|>": 50276,
151
+ "<|hr|>": 50291,
152
+ "<|ht|>": 50339,
153
+ "<|hu|>": 50286,
154
+ "<|hy|>": 50312,
155
+ "<|id|>": 50275,
156
+ "<|is|>": 50311,
157
+ "<|it|>": 50274,
158
+ "<|ja|>": 50266,
159
+ "<|jw|>": 50356,
160
+ "<|ka|>": 50329,
161
+ "<|kk|>": 50316,
162
+ "<|km|>": 50323,
163
+ "<|kn|>": 50306,
164
+ "<|ko|>": 50264,
165
+ "<|la|>": 50294,
166
+ "<|lb|>": 50345,
167
+ "<|ln|>": 50353,
168
+ "<|lo|>": 50336,
169
+ "<|lt|>": 50293,
170
+ "<|lv|>": 50301,
171
+ "<|mg|>": 50349,
172
+ "<|mi|>": 50295,
173
+ "<|mk|>": 50308,
174
+ "<|ml|>": 50296,
175
+ "<|mn|>": 50314,
176
+ "<|mr|>": 50320,
177
+ "<|ms|>": 50282,
178
+ "<|mt|>": 50343,
179
+ "<|my|>": 50346,
180
+ "<|ne|>": 50313,
181
+ "<|nl|>": 50271,
182
+ "<|nn|>": 50342,
183
+ "<|no|>": 50288,
184
+ "<|oc|>": 50328,
185
+ "<|pa|>": 50321,
186
+ "<|pl|>": 50269,
187
+ "<|ps|>": 50340,
188
+ "<|pt|>": 50267,
189
+ "<|ro|>": 50284,
190
+ "<|ru|>": 50263,
191
+ "<|sa|>": 50344,
192
+ "<|sd|>": 50332,
193
+ "<|si|>": 50322,
194
+ "<|sk|>": 50298,
195
+ "<|sl|>": 50305,
196
+ "<|sn|>": 50324,
197
+ "<|so|>": 50326,
198
+ "<|sq|>": 50317,
199
+ "<|sr|>": 50303,
200
+ "<|su|>": 50357,
201
+ "<|sv|>": 50273,
202
+ "<|sw|>": 50318,
203
+ "<|ta|>": 50287,
204
+ "<|te|>": 50299,
205
+ "<|tg|>": 50331,
206
+ "<|th|>": 50289,
207
+ "<|tk|>": 50341,
208
+ "<|tl|>": 50348,
209
+ "<|tr|>": 50268,
210
+ "<|tt|>": 50351,
211
+ "<|uk|>": 50280,
212
+ "<|ur|>": 50290,
213
+ "<|uz|>": 50337,
214
+ "<|vi|>": 50278,
215
+ "<|yi|>": 50335,
216
+ "<|yo|>": 50325,
217
+ "<|zh|>": 50260
218
+ },
219
+ "max_initial_timestamp_index": 1,
220
+ "max_length": 448,
221
+ "no_timestamps_token_id": 50363,
222
+ "pad_token_id": 50257,
223
+ "return_timestamps": false,
224
+ "suppress_tokens": [
225
+ 1,
226
+ 2,
227
+ 7,
228
+ 8,
229
+ 9,
230
+ 10,
231
+ 14,
232
+ 25,
233
+ 26,
234
+ 27,
235
+ 28,
236
+ 29,
237
+ 31,
238
+ 58,
239
+ 59,
240
+ 60,
241
+ 61,
242
+ 62,
243
+ 63,
244
+ 90,
245
+ 91,
246
+ 92,
247
+ 93,
248
+ 359,
249
+ 503,
250
+ 522,
251
+ 542,
252
+ 873,
253
+ 893,
254
+ 902,
255
+ 918,
256
+ 922,
257
+ 931,
258
+ 1350,
259
+ 1853,
260
+ 1982,
261
+ 2460,
262
+ 2627,
263
+ 3246,
264
+ 3253,
265
+ 3268,
266
+ 3536,
267
+ 3846,
268
+ 3961,
269
+ 4183,
270
+ 4667,
271
+ 6585,
272
+ 6647,
273
+ 7273,
274
+ 9061,
275
+ 9383,
276
+ 10428,
277
+ 10929,
278
+ 11938,
279
+ 12033,
280
+ 12331,
281
+ 12562,
282
+ 13793,
283
+ 14157,
284
+ 14635,
285
+ 15265,
286
+ 15618,
287
+ 16553,
288
+ 16604,
289
+ 18362,
290
+ 18956,
291
+ 20075,
292
+ 21675,
293
+ 22520,
294
+ 26130,
295
+ 26161,
296
+ 26435,
297
+ 28279,
298
+ 29464,
299
+ 31650,
300
+ 32302,
301
+ 32470,
302
+ 36865,
303
+ 42863,
304
+ 47425,
305
+ 49870,
306
+ 50254,
307
+ 50258,
308
+ 50358,
309
+ 50359,
310
+ 50360,
311
+ 50361,
312
+ 50362
313
+ ],
314
+ "task": "transcribe",
315
+ "task_to_id": {
316
+ "transcribe": 50359,
317
+ "translate": 50358
318
+ },
319
+ "transformers_version": "4.34.1"
320
+ }