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: 0.5454
- Accuracy: 0.84
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: 12
- eval_batch_size: 12
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.0594 | 1.0 | 75 | 1.9411 | 0.59 |
| 1.4643 | 2.0 | 150 | 1.3450 | 0.72 |
| 1.1926 | 3.0 | 225 | 1.1038 | 0.7 |
| 0.9126 | 4.0 | 300 | 0.9084 | 0.71 |
| 0.6716 | 5.0 | 375 | 0.7864 | 0.77 |
| 0.5595 | 6.0 | 450 | 0.6647 | 0.8 |
| 0.4235 | 7.0 | 525 | 0.6587 | 0.8 |
| 0.3118 | 8.0 | 600 | 0.6317 | 0.81 |
| 0.2283 | 9.0 | 675 | 0.5696 | 0.84 |
| 0.264 | 10.0 | 750 | 0.5454 | 0.84 |
Framework versions
- Transformers 4.31.0.dev0
- Pytorch 1.13.1
- Datasets 2.13.1
- Tokenizers 0.13.3