apv53-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: 0.5644
- 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 adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.9682 | 1.0 | 113 | 1.9097 | 0.42 |
| 1.2063 | 2.0 | 226 | 1.2490 | 0.7 |
| 1.0341 | 3.0 | 339 | 0.8957 | 0.78 |
| 0.5967 | 4.0 | 452 | 0.8012 | 0.73 |
| 0.6163 | 5.0 | 565 | 0.6948 | 0.79 |
| 0.2877 | 6.0 | 678 | 0.6130 | 0.82 |
| 0.2392 | 7.0 | 791 | 0.5434 | 0.84 |
| 0.1176 | 8.0 | 904 | 0.5847 | 0.86 |
| 0.1138 | 9.0 | 1017 | 0.5545 | 0.86 |
| 0.0885 | 10.0 | 1130 | 0.5644 | 0.87 |
Framework versions
- Transformers 4.48.0.dev0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0