Hugging Face's logo

8688chris
/
Helldivers2ASR_V3

Safetensors
wav2vec2
Generated from Trainer
Model card Files
xet
Community
8688chris commited on
Commit
3fe066a
·
verified ·
1 Parent(s): ee83b33

Model save

Browse files
Files changed (4) hide show
  1. README.md +33 -33
  2. config.json +4 -4
  3. model.safetensors +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [facebook/wav2vec2-base-960h](https://huggingface.co/facebook/wav2vec2-base-960h) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 75.0289
21
- - Wer: 0.0338
22
 
23
  ## Model description
24
 
@@ -37,7 +37,7 @@ More information needed
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
- - learning_rate: 7e-05
41
  - train_batch_size: 32
42
  - eval_batch_size: 32
43
  - seed: 42
@@ -50,36 +50,36 @@ The following hyperparameters were used during training:
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Wer |
52
  |:-------------:|:-----:|:----:|:---------------:|:------:|
53
- | 1580.0143 | 1.0 | 165 | 894.1667 | 0.4445 |
54
- | 1064.4354 | 2.0 | 330 | 644.3724 | 0.3276 |
55
- | 869.033 | 3.0 | 495 | 618.4804 | 0.2878 |
56
- | 765.606 | 4.0 | 660 | 467.2128 | 0.2274 |
57
- | 648.5554 | 5.0 | 825 | 465.9539 | 0.2182 |
58
- | 590.8423 | 6.0 | 990 | 417.7243 | 0.2012 |
59
- | 500.8294 | 7.0 | 1155 | 360.0778 | 0.1614 |
60
- | 468.3419 | 8.0 | 1320 | 423.7473 | 0.1842 |
61
- | 438.408 | 9.0 | 1485 | 343.4666 | 0.1572 |
62
- | 405.389 | 10.0 | 1650 | 272.7067 | 0.1332 |
63
- | 338.7186 | 11.0 | 1815 | 239.3145 | 0.1133 |
64
- | 335.6381 | 12.0 | 1980 | 183.9102 | 0.0918 |
65
- | 327.5601 | 13.0 | 2145 | 175.3572 | 0.0827 |
66
- | 255.2483 | 14.0 | 2310 | 175.2945 | 0.0848 |
67
- | 257.8507 | 15.0 | 2475 | 149.6469 | 0.0733 |
68
- | 221.9156 | 16.0 | 2640 | 177.4545 | 0.0756 |
69
- | 232.816 | 17.0 | 2805 | 226.1758 | 0.0984 |
70
- | 213.0188 | 18.0 | 2970 | 178.6724 | 0.0756 |
71
- | 207.679 | 19.0 | 3135 | 131.9617 | 0.0534 |
72
- | 198.4934 | 20.0 | 3300 | 132.1202 | 0.0484 |
73
- | 177.8431 | 21.0 | 3465 | 86.7172 | 0.0447 |
74
- | 178.1199 | 22.0 | 3630 | 93.3159 | 0.0447 |
75
- | 178.4063 | 23.0 | 3795 | 103.9663 | 0.0463 |
76
- | 170.2629 | 24.0 | 3960 | 96.0694 | 0.0385 |
77
- | 147.0691 | 25.0 | 4125 | 90.9803 | 0.0361 |
78
- | 139.5575 | 26.0 | 4290 | 89.6164 | 0.0385 |
79
- | 141.0887 | 27.0 | 4455 | 64.7452 | 0.0311 |
80
- | 138.4675 | 28.0 | 4620 | 87.2011 | 0.0385 |
81
- | 137.3686 | 29.0 | 4785 | 60.8403 | 0.0277 |
82
- | 108.6982 | 30.0 | 4950 | 75.0289 | 0.0338 |
83
 
84
 
85
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [facebook/wav2vec2-base-960h](https://huggingface.co/facebook/wav2vec2-base-960h) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 135.5781
21
+ - Wer: 0.0427
22
 
23
  ## Model description
24
 
 
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
+ - learning_rate: 8e-05
41
  - train_batch_size: 32
42
  - eval_batch_size: 32
43
  - seed: 42
 
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Wer |
52
  |:-------------:|:-----:|:----:|:---------------:|:------:|
53
+ | 1518.9794 | 1.0 | 168 | 1000.8097 | 0.4552 |
54
+ | 1028.341 | 2.0 | 336 | 779.6321 | 0.3719 |
55
+ | 820.663 | 3.0 | 504 | 659.3882 | 0.2988 |
56
+ | 718.5459 | 4.0 | 672 | 516.9663 | 0.2303 |
57
+ | 606.0076 | 5.0 | 840 | 421.7630 | 0.1998 |
58
+ | 535.3864 | 6.0 | 1008 | 478.5203 | 0.2051 |
59
+ | 466.5894 | 7.0 | 1176 | 440.6602 | 0.1739 |
60
+ | 432.4227 | 8.0 | 1344 | 294.9408 | 0.1323 |
61
+ | 373.2972 | 9.0 | 1512 | 261.7947 | 0.1122 |
62
+ | 355.762 | 10.0 | 1680 | 315.7706 | 0.1300 |
63
+ | 314.1884 | 11.0 | 1848 | 350.8355 | 0.1181 |
64
+ | 288.5761 | 12.0 | 2016 | 310.4185 | 0.1201 |
65
+ | 291.121 | 13.0 | 2184 | 255.1681 | 0.0876 |
66
+ | 236.6435 | 14.0 | 2352 | 240.4549 | 0.0873 |
67
+ | 219.9664 | 15.0 | 2520 | 237.7248 | 0.0922 |
68
+ | 201.4798 | 16.0 | 2688 | 162.6640 | 0.0619 |
69
+ | 199.2839 | 17.0 | 2856 | 232.3585 | 0.0800 |
70
+ | 194.1537 | 18.0 | 3024 | 215.3707 | 0.0772 |
71
+ | 177.814 | 19.0 | 3192 | 171.7732 | 0.0589 |
72
+ | 166.7409 | 20.0 | 3360 | 166.2487 | 0.0597 |
73
+ | 169.7996 | 21.0 | 3528 | 135.5038 | 0.0546 |
74
+ | 153.8049 | 22.0 | 3696 | 150.6883 | 0.0518 |
75
+ | 143.4673 | 23.0 | 3864 | 179.0132 | 0.0541 |
76
+ | 147.514 | 24.0 | 4032 | 131.5579 | 0.0419 |
77
+ | 138.0108 | 25.0 | 4200 | 154.8247 | 0.0493 |
78
+ | 142.9634 | 26.0 | 4368 | 165.7421 | 0.0586 |
79
+ | 123.1378 | 27.0 | 4536 | 160.7584 | 0.0472 |
80
+ | 129.9836 | 28.0 | 4704 | 104.9703 | 0.0366 |
81
+ | 113.1207 | 29.0 | 4872 | 172.9598 | 0.0490 |
82
+ | 110.3937 | 30.0 | 5040 | 135.5781 | 0.0427 |
83
 
84
 
85
  ### Framework versions
config.json CHANGED
@@ -9,7 +9,7 @@
9
  "architectures": [
10
  "Wav2Vec2ForCTC"
11
  ],
12
- "attention_dropout": 0.2,
13
  "bos_token_id": 1,
14
  "classifier_proj_size": 256,
15
  "codevector_dim": 256,
@@ -50,12 +50,12 @@
50
  "feat_extract_activation": "gelu",
51
  "feat_extract_dropout": 0.0,
52
  "feat_extract_norm": "group",
53
- "feat_proj_dropout": 0.2,
54
  "feat_quantizer_dropout": 0.0,
55
- "final_dropout": 0.2,
56
  "gradient_checkpointing": false,
57
  "hidden_act": "gelu",
58
- "hidden_dropout": 0.2,
59
  "hidden_dropout_prob": 0.1,
60
  "hidden_size": 768,
61
  "initializer_range": 0.02,
 
9
  "architectures": [
10
  "Wav2Vec2ForCTC"
11
  ],
12
+ "attention_dropout": 0.15,
13
  "bos_token_id": 1,
14
  "classifier_proj_size": 256,
15
  "codevector_dim": 256,
 
50
  "feat_extract_activation": "gelu",
51
  "feat_extract_dropout": 0.0,
52
  "feat_extract_norm": "group",
53
+ "feat_proj_dropout": 0.15,
54
  "feat_quantizer_dropout": 0.0,
55
+ "final_dropout": 0.15,
56
  "gradient_checkpointing": false,
57
  "hidden_act": "gelu",
58
+ "hidden_dropout": 0.15,
59
  "hidden_dropout_prob": 0.1,
60
  "hidden_size": 768,
61
  "initializer_range": 0.02,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b4ec641f9b257ef7c8c09b29da1143c80b592ba9a8a28c40fa31b9596c1919e3
3
  size 377611120
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b8fe0a83b85190e9c5df256386b63f109df3c6d5bc2d19218bec38e8582c02ab
3
  size 377611120
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e6b2cba4971a81e5703a54274bd33925e85207e462105f2ec00de6d59969abe3
3
  size 5176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cb190c3f93a398db49cf968b1143a0a2d6ea9ba39aea61f61608d6d31dfaa5a
3
  size 5176