bqtsio commited on
Commit
817d836
1 Parent(s): 2db6d43

End of training

Browse files
Files changed (2) hide show
  1. README.md +78 -0
  2. generation_config.json +189 -0
README.md ADDED
@@ -0,0 +1,78 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ language:
4
+ - en
5
+ license: apache-2.0
6
+ base_model: openai/whisper-medium.en
7
+ tags:
8
+ - generated_from_trainer
9
+ datasets:
10
+ - Dev372/Medical_STT_Dataset_1.1
11
+ metrics:
12
+ - wer
13
+ model-index:
14
+ - name: Whisper Medium
15
+ results:
16
+ - task:
17
+ name: Automatic Speech Recognition
18
+ type: automatic-speech-recognition
19
+ dataset:
20
+ name: Medical STT
21
+ type: Dev372/Medical_STT_Dataset_1.1
22
+ metrics:
23
+ - name: Wer
24
+ type: wer
25
+ value: 3.2511210762331837
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 Medium
32
+
33
+ This model is a fine-tuned version of [openai/whisper-medium.en](https://huggingface.co/openai/whisper-medium.en) on the Medical STT dataset.
34
+ It achieves the following results on the evaluation set:
35
+ - Loss: 0.0977
36
+ - Wer Ortho: 5.4215
37
+ - Wer: 3.2511
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: 16
59
+ - seed: 42
60
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
+ - lr_scheduler_type: constant_with_warmup
62
+ - lr_scheduler_warmup_steps: 50
63
+ - training_steps: 500
64
+ - mixed_precision_training: Native AMP
65
+
66
+ ### Training results
67
+
68
+ | Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
69
+ |:-------------:|:------:|:----:|:---------------:|:---------:|:------:|
70
+ | 0.0481 | 1.2563 | 500 | 0.0977 | 5.4215 | 3.2511 |
71
+
72
+
73
+ ### Framework versions
74
+
75
+ - Transformers 4.45.1
76
+ - Pytorch 2.4.0
77
+ - Datasets 3.0.1
78
+ - Tokenizers 0.20.0
generation_config.json ADDED
@@ -0,0 +1,189 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 11,
5
+ 4
6
+ ],
7
+ [
8
+ 14,
9
+ 1
10
+ ],
11
+ [
12
+ 14,
13
+ 12
14
+ ],
15
+ [
16
+ 14,
17
+ 14
18
+ ],
19
+ [
20
+ 15,
21
+ 4
22
+ ],
23
+ [
24
+ 16,
25
+ 0
26
+ ],
27
+ [
28
+ 16,
29
+ 4
30
+ ],
31
+ [
32
+ 16,
33
+ 9
34
+ ],
35
+ [
36
+ 17,
37
+ 12
38
+ ],
39
+ [
40
+ 17,
41
+ 14
42
+ ],
43
+ [
44
+ 18,
45
+ 7
46
+ ],
47
+ [
48
+ 18,
49
+ 10
50
+ ],
51
+ [
52
+ 18,
53
+ 15
54
+ ],
55
+ [
56
+ 20,
57
+ 0
58
+ ],
59
+ [
60
+ 20,
61
+ 3
62
+ ],
63
+ [
64
+ 20,
65
+ 9
66
+ ],
67
+ [
68
+ 20,
69
+ 14
70
+ ],
71
+ [
72
+ 21,
73
+ 12
74
+ ]
75
+ ],
76
+ "begin_suppress_tokens": [
77
+ 220,
78
+ 50256
79
+ ],
80
+ "bos_token_id": 50257,
81
+ "decoder_start_token_id": 50257,
82
+ "eos_token_id": 50256,
83
+ "forced_decoder_ids": [
84
+ [
85
+ 1,
86
+ 50362
87
+ ]
88
+ ],
89
+ "is_multilingual": false,
90
+ "max_initial_timestamp_index": 50,
91
+ "max_length": 448,
92
+ "no_timestamps_token_id": 50362,
93
+ "pad_token_id": 50256,
94
+ "prev_sot_token_id": 50360,
95
+ "return_timestamps": false,
96
+ "suppress_tokens": [
97
+ 1,
98
+ 2,
99
+ 7,
100
+ 8,
101
+ 9,
102
+ 10,
103
+ 14,
104
+ 25,
105
+ 26,
106
+ 27,
107
+ 28,
108
+ 29,
109
+ 31,
110
+ 58,
111
+ 59,
112
+ 60,
113
+ 61,
114
+ 62,
115
+ 63,
116
+ 90,
117
+ 91,
118
+ 92,
119
+ 93,
120
+ 357,
121
+ 366,
122
+ 438,
123
+ 532,
124
+ 685,
125
+ 705,
126
+ 796,
127
+ 930,
128
+ 1058,
129
+ 1220,
130
+ 1267,
131
+ 1279,
132
+ 1303,
133
+ 1343,
134
+ 1377,
135
+ 1391,
136
+ 1635,
137
+ 1782,
138
+ 1875,
139
+ 2162,
140
+ 2361,
141
+ 2488,
142
+ 3467,
143
+ 4008,
144
+ 4211,
145
+ 4600,
146
+ 4808,
147
+ 5299,
148
+ 5855,
149
+ 6329,
150
+ 7203,
151
+ 9609,
152
+ 9959,
153
+ 10563,
154
+ 10786,
155
+ 11420,
156
+ 11709,
157
+ 11907,
158
+ 13163,
159
+ 13697,
160
+ 13700,
161
+ 14808,
162
+ 15306,
163
+ 16410,
164
+ 16791,
165
+ 17992,
166
+ 19203,
167
+ 19510,
168
+ 20724,
169
+ 22305,
170
+ 22935,
171
+ 27007,
172
+ 30109,
173
+ 30420,
174
+ 33409,
175
+ 34949,
176
+ 40283,
177
+ 40493,
178
+ 40549,
179
+ 47282,
180
+ 49146,
181
+ 50257,
182
+ 50357,
183
+ 50358,
184
+ 50359,
185
+ 50360,
186
+ 50361
187
+ ],
188
+ "transformers_version": "4.45.1"
189
+ }