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: 0.9340
- 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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 15
- label_smoothing_factor: 0.1
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.4359 | 1.0 | 225 | 1.4200 | 0.6 |
| 1.2373 | 2.0 | 450 | 1.0609 | 0.76 |
| 0.9500 | 3.0 | 675 | 0.9684 | 0.8 |
| 0.6715 | 4.0 | 900 | 0.9773 | 0.81 |
| 0.6988 | 5.0 | 1125 | 0.9237 | 0.82 |
| 0.5831 | 6.0 | 1350 | 1.0907 | 0.79 |
| 0.5171 | 7.0 | 1575 | 0.9349 | 0.85 |
| 0.5029 | 8.0 | 1800 | 0.9940 | 0.85 |
| 0.5031 | 9.0 | 2025 | 0.9774 | 0.86 |
| 0.5040 | 10.0 | 2250 | 0.9287 | 0.86 |
| 0.5008 | 11.0 | 2475 | 0.9158 | 0.87 |
| 0.5081 | 12.0 | 2700 | 0.9471 | 0.86 |
| 0.5011 | 13.0 | 2925 | 0.9285 | 0.87 |
| 0.5008 | 14.0 | 3150 | 0.9309 | 0.87 |
| 0.5009 | 15.0 | 3375 | 0.9340 | 0.87 |
Framework versions
- Transformers 5.15.1
- Pytorch 2.11.0
- Datasets 5.0.1
- Tokenizers 0.22.2