distilhubert-finetuned-gtzan-hlv
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: 0.6077
- 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 OptimizerNames.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 |
|---|---|---|---|---|
| 1.8606 | 1.0 | 113 | 1.8668 | 0.49 |
| 1.2705 | 2.0 | 226 | 1.2509 | 0.68 |
| 1.0827 | 3.0 | 339 | 1.0456 | 0.73 |
| 0.7267 | 4.0 | 452 | 0.8592 | 0.76 |
| 0.6331 | 5.0 | 565 | 0.7087 | 0.79 |
| 0.4329 | 6.0 | 678 | 0.6839 | 0.79 |
| 0.2621 | 7.0 | 791 | 0.5815 | 0.84 |
| 0.1294 | 8.0 | 904 | 0.6285 | 0.83 |
| 0.1263 | 9.0 | 1017 | 0.6109 | 0.85 |
| 0.0958 | 10.0 | 1130 | 0.6077 | 0.85 |
Framework versions
- Transformers 5.20.2
- Pytorch 2.11.0+cu130
- Datasets 4.8.5
- Tokenizers 0.23.3