distilhubert-gtzan-exp04-augmentation
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: 0.5938
- Accuracy: 0.8
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- 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 |
|---|---|---|---|---|
| 3.9386 | 1.0 | 113 | 1.8870 | 0.48 |
| 2.5958 | 2.0 | 226 | 1.2741 | 0.69 |
| 2.0548 | 3.0 | 339 | 1.0959 | 0.68 |
| 1.4009 | 4.0 | 452 | 0.8958 | 0.72 |
| 1.2379 | 5.0 | 565 | 0.7504 | 0.82 |
| 1.0527 | 6.0 | 678 | 0.6666 | 0.78 |
| 0.7448 | 7.0 | 791 | 0.6178 | 0.81 |
| 0.4440 | 8.0 | 904 | 0.6107 | 0.81 |
| 0.5523 | 9.0 | 1017 | 0.5715 | 0.83 |
| 0.2745 | 10.0 | 1130 | 0.5938 | 0.8 |
Framework versions
- Transformers 5.15.0
- Pytorch 2.11.0+cu128
- Datasets 2.21.0
- Tokenizers 0.22.2