rossevine commited on
Commit
0764333
·
1 Parent(s): 9338fb6

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +43 -42
README.md CHANGED
@@ -1,6 +1,8 @@
1
  ---
2
  tags:
3
  - generated_from_trainer
 
 
4
  model-index:
5
  - name: Model_G_S_D_Wav2Vec2
6
  results: []
@@ -13,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
13
 
14
  This model was trained from scratch on the None dataset.
15
  It achieves the following results on the evaluation set:
16
- - Loss: 0.0423
17
- - Wer: 0.0325
18
- - Cer: 0.0102
19
 
20
  ## Model description
21
 
@@ -44,52 +46,51 @@ The following hyperparameters were used during training:
44
  - lr_scheduler_type: linear
45
  - lr_scheduler_warmup_steps: 500
46
  - num_epochs: 30
47
- - mixed_precision_training: Native AMP
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
52
  |:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|
53
- | 0.1858 | 0.85 | 400 | 0.0676 | 0.0958 | 0.0230 |
54
- | 0.1418 | 1.71 | 800 | 0.0474 | 0.0578 | 0.0148 |
55
- | 0.1096 | 2.56 | 1200 | 0.0535 | 0.0671 | 0.0180 |
56
- | 0.1074 | 3.41 | 1600 | 0.0483 | 0.0558 | 0.0147 |
57
- | 0.0955 | 4.26 | 2000 | 0.0456 | 0.0571 | 0.0154 |
58
- | 0.0854 | 5.12 | 2400 | 0.0410 | 0.0548 | 0.0140 |
59
- | 0.0705 | 5.97 | 2800 | 0.0484 | 0.0577 | 0.0155 |
60
- | 0.0679 | 6.82 | 3200 | 0.0479 | 0.0542 | 0.0152 |
61
- | 0.0618 | 7.68 | 3600 | 0.0451 | 0.0472 | 0.0124 |
62
- | 0.0561 | 8.53 | 4000 | 0.0604 | 0.0677 | 0.0185 |
63
- | 0.0583 | 9.38 | 4400 | 0.0544 | 0.1183 | 0.0276 |
64
- | 0.0521 | 10.23 | 4800 | 0.0487 | 0.0469 | 0.0130 |
65
- | 0.0482 | 11.09 | 5200 | 0.0476 | 0.0484 | 0.0140 |
66
- | 0.0416 | 11.94 | 5600 | 0.0427 | 0.0452 | 0.0123 |
67
- | 0.0417 | 12.79 | 6000 | 0.0402 | 0.0431 | 0.0120 |
68
- | 0.034 | 13.65 | 6400 | 0.0427 | 0.0448 | 0.0131 |
69
- | 0.034 | 14.5 | 6800 | 0.0442 | 0.0431 | 0.0125 |
70
- | 0.0313 | 15.35 | 7200 | 0.0383 | 0.0407 | 0.0111 |
71
- | 0.0287 | 16.2 | 7600 | 0.0380 | 0.0394 | 0.0112 |
72
- | 0.0305 | 17.06 | 8000 | 0.0433 | 0.0386 | 0.0117 |
73
- | 0.0242 | 17.91 | 8400 | 0.0448 | 0.0426 | 0.0126 |
74
- | 0.0207 | 18.76 | 8800 | 0.0423 | 0.0368 | 0.0110 |
75
- | 0.0226 | 19.62 | 9200 | 0.0411 | 0.0375 | 0.0112 |
76
- | 0.0199 | 20.47 | 9600 | 0.0382 | 0.0373 | 0.0111 |
77
- | 0.018 | 21.32 | 10000 | 0.0407 | 0.0371 | 0.0109 |
78
- | 0.0146 | 22.17 | 10400 | 0.0441 | 0.0374 | 0.0117 |
79
- | 0.017 | 23.03 | 10800 | 0.0396 | 0.0341 | 0.0106 |
80
- | 0.0119 | 23.88 | 11200 | 0.0424 | 0.0334 | 0.0106 |
81
- | 0.0117 | 24.73 | 11600 | 0.0399 | 0.0320 | 0.0099 |
82
- | 0.0107 | 25.59 | 12000 | 0.0417 | 0.0325 | 0.0104 |
83
- | 0.0103 | 26.44 | 12400 | 0.0404 | 0.0323 | 0.0101 |
84
- | 0.0096 | 27.29 | 12800 | 0.0423 | 0.0319 | 0.0104 |
85
- | 0.0097 | 28.14 | 13200 | 0.0428 | 0.0326 | 0.0105 |
86
- | 0.0095 | 29.0 | 13600 | 0.0424 | 0.0323 | 0.0102 |
87
- | 0.0079 | 29.85 | 14000 | 0.0423 | 0.0325 | 0.0102 |
88
 
89
 
90
  ### Framework versions
91
 
92
- - Transformers 4.11.3
93
- - Pytorch 1.10.0+cu113
94
  - Datasets 1.18.3
95
- - Tokenizers 0.10.3
 
1
  ---
2
  tags:
3
  - generated_from_trainer
4
+ metrics:
5
+ - wer
6
  model-index:
7
  - name: Model_G_S_D_Wav2Vec2
8
  results: []
 
15
 
16
  This model was trained from scratch on the None dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.0425
19
+ - Wer: 0.0310
20
+ - Cer: 0.0095
21
 
22
  ## Model description
23
 
 
46
  - lr_scheduler_type: linear
47
  - lr_scheduler_warmup_steps: 500
48
  - num_epochs: 30
 
49
 
50
  ### Training results
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
53
  |:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|
54
+ | 0.4434 | 0.85 | 400 | 0.0763 | 0.0984 | 0.0254 |
55
+ | 0.1737 | 1.71 | 800 | 0.0639 | 0.0781 | 0.0199 |
56
+ | 0.1293 | 2.56 | 1200 | 0.0522 | 0.0653 | 0.0167 |
57
+ | 0.0965 | 3.41 | 1600 | 0.0471 | 0.0659 | 0.0163 |
58
+ | 0.0874 | 4.26 | 2000 | 0.0464 | 0.0535 | 0.0139 |
59
+ | 0.0679 | 5.12 | 2400 | 0.0395 | 0.0490 | 0.0132 |
60
+ | 0.0618 | 5.97 | 2800 | 0.0424 | 0.0533 | 0.0143 |
61
+ | 0.056 | 6.82 | 3200 | 0.0471 | 0.0511 | 0.0132 |
62
+ | 0.0568 | 7.68 | 3600 | 0.0432 | 0.0468 | 0.0123 |
63
+ | 0.046 | 8.53 | 4000 | 0.0425 | 0.0472 | 0.0130 |
64
+ | 0.0459 | 9.38 | 4400 | 0.0502 | 0.0499 | 0.0134 |
65
+ | 0.0408 | 10.23 | 4800 | 0.0450 | 0.0488 | 0.0131 |
66
+ | 0.0436 | 11.09 | 5200 | 0.0431 | 0.0420 | 0.0119 |
67
+ | 0.0375 | 11.94 | 5600 | 0.0463 | 0.0484 | 0.0132 |
68
+ | 0.0327 | 12.79 | 6000 | 0.0412 | 0.0424 | 0.0116 |
69
+ | 0.0322 | 13.65 | 6400 | 0.0381 | 0.0382 | 0.0111 |
70
+ | 0.0316 | 14.5 | 6800 | 0.0441 | 0.0460 | 0.0128 |
71
+ | 0.0296 | 15.35 | 7200 | 0.0426 | 0.0415 | 0.0119 |
72
+ | 0.0274 | 16.2 | 7600 | 0.0421 | 0.0383 | 0.0106 |
73
+ | 0.0247 | 17.06 | 8000 | 0.0442 | 0.0391 | 0.0120 |
74
+ | 0.0235 | 17.91 | 8400 | 0.0449 | 0.0409 | 0.0116 |
75
+ | 0.0219 | 18.76 | 8800 | 0.0394 | 0.0353 | 0.0106 |
76
+ | 0.0174 | 19.62 | 9200 | 0.0489 | 0.0393 | 0.0117 |
77
+ | 0.0161 | 20.47 | 9600 | 0.0421 | 0.0347 | 0.0099 |
78
+ | 0.0158 | 21.32 | 10000 | 0.0425 | 0.0349 | 0.0108 |
79
+ | 0.0141 | 22.17 | 10400 | 0.0436 | 0.0397 | 0.0116 |
80
+ | 0.0156 | 23.03 | 10800 | 0.0432 | 0.0375 | 0.0114 |
81
+ | 0.0138 | 23.88 | 11200 | 0.0438 | 0.0364 | 0.0110 |
82
+ | 0.0116 | 24.73 | 11600 | 0.0420 | 0.0368 | 0.0108 |
83
+ | 0.0108 | 25.59 | 12000 | 0.0407 | 0.0341 | 0.0103 |
84
+ | 0.0073 | 26.44 | 12400 | 0.0428 | 0.0336 | 0.0101 |
85
+ | 0.0085 | 27.29 | 12800 | 0.0432 | 0.0328 | 0.0101 |
86
+ | 0.0078 | 28.14 | 13200 | 0.0416 | 0.0318 | 0.0096 |
87
+ | 0.0065 | 29.0 | 13600 | 0.0423 | 0.0310 | 0.0097 |
88
+ | 0.0062 | 29.85 | 14000 | 0.0425 | 0.0310 | 0.0095 |
89
 
90
 
91
  ### Framework versions
92
 
93
+ - Transformers 4.31.0
94
+ - Pytorch 2.0.1+cu117
95
  - Datasets 1.18.3
96
+ - Tokenizers 0.13.3