Hugging Face's logo

flaneur-ml
/
distilhubert-finetuned-gtzan

Audio Classification
Transformers
TensorBoard
Safetensors
hubert
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
flaneur-ml commited on
Commit
6bc6ea6
·
verified ·
1 Parent(s): 585859b

End of training

Browse files
Files changed (2) hide show
  1. README.md +28 -28
  2. model.safetensors +1 -1
README.md CHANGED
@@ -33,7 +33,7 @@ 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: 0.8175
37
  - Accuracy: 0.86
38
 
39
  ## Model description
@@ -53,11 +53,11 @@ More information needed
53
  ### Training hyperparameters
54
 
55
  The following hyperparameters were used during training:
56
- - learning_rate: 5e-05
57
- - train_batch_size: 8
58
- - eval_batch_size: 8
59
  - seed: 42
60
- - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
  - lr_scheduler_type: linear
62
  - lr_scheduler_warmup_ratio: 0.1
63
  - num_epochs: 20
@@ -67,31 +67,31 @@ The following hyperparameters were used during training:
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
70
- | 2.1414 | 1.0 | 113 | 2.0686 | 0.52 |
71
- | 1.3917 | 2.0 | 226 | 1.4505 | 0.56 |
72
- | 1.109 | 3.0 | 339 | 1.0342 | 0.71 |
73
- | 0.6752 | 4.0 | 452 | 0.8531 | 0.74 |
74
- | 0.5346 | 5.0 | 565 | 0.7352 | 0.74 |
75
- | 0.3598 | 6.0 | 678 | 0.5552 | 0.82 |
76
- | 0.32 | 7.0 | 791 | 0.5660 | 0.84 |
77
- | 0.1663 | 8.0 | 904 | 0.5829 | 0.84 |
78
- | 0.0369 | 9.0 | 1017 | 0.7868 | 0.83 |
79
- | 0.0235 | 10.0 | 1130 | 0.8371 | 0.84 |
80
- | 0.0087 | 11.0 | 1243 | 0.7114 | 0.84 |
81
- | 0.0064 | 12.0 | 1356 | 0.7578 | 0.84 |
82
- | 0.0046 | 13.0 | 1469 | 0.7859 | 0.83 |
83
- | 0.0042 | 14.0 | 1582 | 0.8681 | 0.86 |
84
- | 0.0032 | 15.0 | 1695 | 0.8926 | 0.86 |
85
- | 0.0031 | 16.0 | 1808 | 0.8339 | 0.84 |
86
- | 0.0029 | 17.0 | 1921 | 0.7772 | 0.86 |
87
- | 0.0025 | 18.0 | 2034 | 0.8376 | 0.86 |
88
- | 0.0025 | 19.0 | 2147 | 0.8175 | 0.86 |
89
- | 0.0024 | 20.0 | 2260 | 0.8175 | 0.86 |
90
 
91
 
92
  ### Framework versions
93
 
94
- - Transformers 4.44.2
95
- - Pytorch 2.5.0+cu121
96
  - Datasets 3.1.0
97
- - Tokenizers 0.19.1
 
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: 0.6628
37
  - Accuracy: 0.86
38
 
39
  ## Model description
 
53
  ### Training hyperparameters
54
 
55
  The following hyperparameters were used during training:
56
+ - learning_rate: 8e-05
57
+ - train_batch_size: 12
58
+ - eval_batch_size: 12
59
  - seed: 42
60
+ - optimizer: Use 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: 20
 
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
70
+ | 2.1317 | 1.0 | 75 | 2.0386 | 0.33 |
71
+ | 1.36 | 2.0 | 150 | 1.4142 | 0.58 |
72
+ | 1.1456 | 3.0 | 225 | 1.1110 | 0.66 |
73
+ | 0.6417 | 4.0 | 300 | 1.0142 | 0.69 |
74
+ | 0.3324 | 5.0 | 375 | 0.5881 | 0.82 |
75
+ | 0.2208 | 6.0 | 450 | 0.5516 | 0.84 |
76
+ | 0.3346 | 7.0 | 525 | 0.5267 | 0.87 |
77
+ | 0.2309 | 8.0 | 600 | 0.7404 | 0.8 |
78
+ | 0.0267 | 9.0 | 675 | 0.6636 | 0.8 |
79
+ | 0.0309 | 10.0 | 750 | 0.6390 | 0.84 |
80
+ | 0.0076 | 11.0 | 825 | 0.6949 | 0.85 |
81
+ | 0.0053 | 12.0 | 900 | 0.6405 | 0.87 |
82
+ | 0.005 | 13.0 | 975 | 0.7065 | 0.84 |
83
+ | 0.004 | 14.0 | 1050 | 0.8570 | 0.84 |
84
+ | 0.0031 | 15.0 | 1125 | 0.6735 | 0.88 |
85
+ | 0.0028 | 16.0 | 1200 | 0.7023 | 0.85 |
86
+ | 0.0027 | 17.0 | 1275 | 0.6823 | 0.86 |
87
+ | 0.0369 | 18.0 | 1350 | 0.7320 | 0.85 |
88
+ | 0.0024 | 19.0 | 1425 | 0.6656 | 0.86 |
89
+ | 0.0023 | 20.0 | 1500 | 0.6628 | 0.86 |
90
 
91
 
92
  ### Framework versions
93
 
94
+ - Transformers 4.46.2
95
+ - Pytorch 2.5.1+cu124
96
  - Datasets 3.1.0
97
+ - Tokenizers 0.20.3
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:380c49abe3021417db86150214ae4c5b0db1149a19c99670db6fe12d685a1773
3
  size 94771728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8cfe47bca52a90244deba9a423a3dcb5f4ca89d2af7c462a23e29bf38ebc79c0
3
  size 94771728