Hugging Face's logo

jason1i
/
distilhubert-finetuned-gtzan

Audio Classification
Transformers
PyTorch
hubert
Generated from Trainer
Inference Endpoints
Model card Files Community
jason1i commited on
Commit
c2aae92
•
1 Parent(s): e22f1c7

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +34 -14
README.md CHANGED
@@ -3,7 +3,7 @@ license: apache-2.0
3
  tags:
4
  - generated_from_trainer
5
  datasets:
6
- - jason1i/gtzan
7
  metrics:
8
  - accuracy
9
  model-index:
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on the GTZAN dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.5943
22
- - Accuracy: 0.85
23
 
24
  ## Model description
25
 
@@ -45,23 +45,43 @@ The following hyperparameters were used during training:
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
  - lr_scheduler_warmup_ratio: 0.1
48
- - num_epochs: 10
49
  - mixed_precision_training: Native AMP
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
54
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
55
- | 1.8065 | 1.0 | 113 | 1.8499 | 0.42 |
56
- | 1.2026 | 2.0 | 226 | 1.2188 | 0.68 |
57
- | 0.9404 | 3.0 | 339 | 0.9297 | 0.75 |
58
- | 0.7641 | 4.0 | 452 | 0.8105 | 0.78 |
59
- | 0.4286 | 5.0 | 565 | 0.6862 | 0.78 |
60
- | 0.3356 | 6.0 | 678 | 0.6253 | 0.82 |
61
- | 0.2241 | 7.0 | 791 | 0.7296 | 0.79 |
62
- | 0.2979 | 8.0 | 904 | 0.5978 | 0.83 |
63
- | 0.2393 | 9.0 | 1017 | 0.5917 | 0.83 |
64
- | 0.3032 | 10.0 | 1130 | 0.5943 | 0.85 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
65
 
66
 
67
  ### Framework versions
 
3
  tags:
4
  - generated_from_trainer
5
  datasets:
6
+ - marsyas/gtzan
7
  metrics:
8
  - accuracy
9
  model-index:
 
18
 
19
  This model is a fine-tuned version of [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on the GTZAN dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 1.0908
22
+ - Accuracy: 0.84
23
 
24
  ## Model description
25
 
 
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
  - lr_scheduler_warmup_ratio: 0.1
48
+ - num_epochs: 30
49
  - mixed_precision_training: Native AMP
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
54
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
55
+ | 2.1531 | 1.0 | 113 | 2.1667 | 0.41 |
56
+ | 1.6622 | 2.0 | 226 | 1.6138 | 0.63 |
57
+ | 1.3112 | 3.0 | 339 | 1.2047 | 0.7 |
58
+ | 0.9374 | 4.0 | 452 | 0.9595 | 0.73 |
59
+ | 0.5475 | 5.0 | 565 | 0.7239 | 0.82 |
60
+ | 0.4845 | 6.0 | 678 | 0.7406 | 0.75 |
61
+ | 0.2489 | 7.0 | 791 | 0.6838 | 0.78 |
62
+ | 0.3272 | 8.0 | 904 | 0.8447 | 0.79 |
63
+ | 0.2244 | 9.0 | 1017 | 0.7184 | 0.81 |
64
+ | 0.0353 | 10.0 | 1130 | 0.8800 | 0.79 |
65
+ | 0.0201 | 11.0 | 1243 | 0.8800 | 0.83 |
66
+ | 0.0079 | 12.0 | 1356 | 0.8207 | 0.83 |
67
+ | 0.004 | 13.0 | 1469 | 0.9218 | 0.82 |
68
+ | 0.003 | 14.0 | 1582 | 1.0004 | 0.83 |
69
+ | 0.0024 | 15.0 | 1695 | 0.9446 | 0.84 |
70
+ | 0.0021 | 16.0 | 1808 | 0.9802 | 0.85 |
71
+ | 0.0018 | 17.0 | 1921 | 0.9766 | 0.84 |
72
+ | 0.0017 | 18.0 | 2034 | 1.0597 | 0.84 |
73
+ | 0.0014 | 19.0 | 2147 | 0.9541 | 0.84 |
74
+ | 0.0012 | 20.0 | 2260 | 1.0408 | 0.84 |
75
+ | 0.0011 | 21.0 | 2373 | 1.0364 | 0.84 |
76
+ | 0.001 | 22.0 | 2486 | 1.0993 | 0.84 |
77
+ | 0.001 | 23.0 | 2599 | 1.0620 | 0.84 |
78
+ | 0.0009 | 24.0 | 2712 | 1.0193 | 0.83 |
79
+ | 0.0009 | 25.0 | 2825 | 1.0164 | 0.83 |
80
+ | 0.0009 | 26.0 | 2938 | 1.0293 | 0.84 |
81
+ | 0.0008 | 27.0 | 3051 | 1.0478 | 0.84 |
82
+ | 0.0008 | 28.0 | 3164 | 1.0727 | 0.84 |
83
+ | 0.0008 | 29.0 | 3277 | 1.0773 | 0.84 |
84
+ | 0.0008 | 30.0 | 3390 | 1.0908 | 0.84 |
85
 
86
 
87
  ### Framework versions