Hugging Face's logo

Gwenn-LR
/
distilhubert-finetuned-gtzan

Audio Classification
Transformers
Safetensors
hubert
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
Gwenn-LR commited on
Commit
0f8dc28
·
verified ·
1 Parent(s): ff0b9ad

End of training

Browse files
Files changed (3) hide show
  1. README.md +25 -16
  2. model.safetensors +1 -1
  3. training_args.bin +2 -2
README.md CHANGED
@@ -23,7 +23,7 @@ model-index:
23
  metrics:
24
  - name: Accuracy
25
  type: accuracy
26
- value: 0.86
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
33
 
34
  This model is a fine-tuned version of [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on the GTZAN dataset.
35
  It achieves the following results on the evaluation set:
36
- - Loss: 0.5504
37
- - Accuracy: 0.86
38
 
39
  ## Model description
40
 
@@ -58,25 +58,34 @@ The following hyperparameters were used during training:
58
  - eval_batch_size: 8
59
  - seed: 42
60
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
61
- - lr_scheduler_type: linear
62
  - lr_scheduler_warmup_ratio: 0.1
63
- - num_epochs: 10
64
- - mixed_precision_training: Native AMP
65
 
66
  ### Training results
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
70
- | 1.9285 | 1.0 | 113 | 1.8711 | 0.5 |
71
- | 1.1558 | 2.0 | 226 | 1.2355 | 0.65 |
72
- | 1.0674 | 3.0 | 339 | 0.9868 | 0.74 |
73
- | 0.5664 | 4.0 | 452 | 0.8502 | 0.69 |
74
- | 0.5117 | 5.0 | 565 | 0.6532 | 0.84 |
75
- | 0.3208 | 6.0 | 678 | 0.6253 | 0.8 |
76
- | 0.2667 | 7.0 | 791 | 0.5786 | 0.84 |
77
- | 0.1143 | 8.0 | 904 | 0.5564 | 0.82 |
78
- | 0.1529 | 9.0 | 1017 | 0.5973 | 0.82 |
79
- | 0.0841 | 10.0 | 1130 | 0.5504 | 0.86 |
 
 
 
 
 
 
 
 
 
 
80
 
81
 
82
  ### Framework versions
 
23
  metrics:
24
  - name: Accuracy
25
  type: accuracy
26
+ value: 0.83
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
33
 
34
  This model is a fine-tuned version of [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on the GTZAN dataset.
35
  It achieves the following results on the evaluation set:
36
+ - Loss: 1.0006
37
+ - Accuracy: 0.83
38
 
39
  ## Model description
40
 
 
58
  - eval_batch_size: 8
59
  - seed: 42
60
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
61
+ - lr_scheduler_type: cosine_with_restarts
62
  - lr_scheduler_warmup_ratio: 0.1
63
+ - num_epochs: 20
 
64
 
65
  ### Training results
66
 
67
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
68
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
69
+ | 2.1596 | 1.0 | 113 | 2.0536 | 0.57 |
70
+ | 1.3991 | 2.0 | 226 | 1.4396 | 0.62 |
71
+ | 1.1572 | 3.0 | 339 | 1.1858 | 0.72 |
72
+ | 0.7911 | 4.0 | 452 | 0.9571 | 0.74 |
73
+ | 0.6416 | 5.0 | 565 | 0.8319 | 0.76 |
74
+ | 0.5359 | 6.0 | 678 | 0.7642 | 0.77 |
75
+ | 0.4587 | 7.0 | 791 | 0.5897 | 0.82 |
76
+ | 0.1906 | 8.0 | 904 | 0.7896 | 0.76 |
77
+ | 0.3607 | 9.0 | 1017 | 0.6474 | 0.78 |
78
+ | 0.1783 | 10.0 | 1130 | 0.6477 | 0.82 |
79
+ | 0.065 | 11.0 | 1243 | 0.5573 | 0.84 |
80
+ | 0.1105 | 12.0 | 1356 | 0.7545 | 0.85 |
81
+ | 0.0193 | 13.0 | 1469 | 0.7586 | 0.85 |
82
+ | 0.0114 | 14.0 | 1582 | 0.7817 | 0.81 |
83
+ | 0.0204 | 15.0 | 1695 | 0.8912 | 0.84 |
84
+ | 0.0059 | 16.0 | 1808 | 0.9070 | 0.84 |
85
+ | 0.004 | 17.0 | 1921 | 1.1039 | 0.81 |
86
+ | 0.0031 | 18.0 | 2034 | 0.6472 | 0.88 |
87
+ | 0.0023 | 19.0 | 2147 | 1.1127 | 0.82 |
88
+ | 0.0023 | 20.0 | 2260 | 1.0006 | 0.83 |
89
 
90
 
91
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ae52bb9cc9192df3de047a5c65f4c38f44b0e5a3c47b0490c418a8acb508dd60
3
  size 94771728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5015d5a0b138d7006404a5e398d1e5ebab6ced29b1ddd155f602eb326597bd44
3
  size 94771728
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c94fa1204bd81fb1e052fd283a36a6a937ec3b01c18e6046f659b7fce79c19a7
3
- size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5782d47bb3449e50ec18a55eaefb6c164a8ad975bea9e3234799ab058a2ce916
3
+ size 5368