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: 0.5595
- Accuracy: 0.85
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 adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.8906 | 1.0 | 113 | 1.8154 | 0.43 |
| 1.1888 | 2.0 | 226 | 1.2004 | 0.67 |
| 0.9917 | 3.0 | 339 | 1.0252 | 0.72 |
| 0.5937 | 4.0 | 452 | 0.7423 | 0.81 |
| 0.4669 | 5.0 | 565 | 0.7231 | 0.77 |
| 0.3205 | 6.0 | 678 | 0.6279 | 0.82 |
| 0.3508 | 7.0 | 791 | 0.5861 | 0.83 |
| 0.1254 | 8.0 | 904 | 0.5335 | 0.84 |
| 0.2296 | 9.0 | 1017 | 0.5837 | 0.84 |
| 0.0852 | 10.0 | 1130 | 0.5595 | 0.85 |
Framework versions
- Transformers 5.13.0
- Pytorch 2.12.1+cu130
- Datasets 2.21.0
- Tokenizers 0.22.2