distilhubert-finetuned-gtzan
This model is a fine-tuned version of ntu-spml/distilhubert on the GTZAN dataset. It achieves the following results on the evaluation set:
- Loss: 0.5647
- Accuracy: 0.87
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.2278 | 1.0 | 57 | 2.1709 | 0.44 |
| 1.7173 | 2.0 | 114 | 1.6084 | 0.57 |
| 1.1979 | 3.0 | 171 | 1.1897 | 0.67 |
| 1.1177 | 4.0 | 228 | 1.0003 | 0.72 |
| 0.8526 | 5.0 | 285 | 0.8854 | 0.73 |
| 0.6463 | 6.0 | 342 | 0.7791 | 0.79 |
| 0.5461 | 7.0 | 399 | 0.7468 | 0.78 |
| 0.3953 | 8.0 | 456 | 0.7352 | 0.75 |
| 0.3054 | 9.0 | 513 | 0.6757 | 0.79 |
| 0.18 | 10.0 | 570 | 0.5711 | 0.76 |
| 0.1526 | 11.0 | 627 | 0.6026 | 0.85 |
| 0.0812 | 12.0 | 684 | 0.5876 | 0.82 |
| 0.0578 | 13.0 | 741 | 0.5815 | 0.85 |
| 0.0318 | 14.0 | 798 | 0.5828 | 0.85 |
| 0.0283 | 15.0 | 855 | 0.5960 | 0.85 |
| 0.0393 | 16.0 | 912 | 0.5674 | 0.85 |
| 0.018 | 17.0 | 969 | 0.5647 | 0.87 |
Framework versions
- Transformers 4.31.0.dev0
- Pytorch 1.13.0
- Datasets 2.1.0
- Tokenizers 0.13.3