<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
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: 0.9928
- Accuracy: 0.83
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.1503 | 1.0 | 113 | 2.0510 | 0.39 |
| 1.5109 | 2.0 | 226 | 1.4380 | 0.52 |
| 1.1899 | 3.0 | 339 | 1.0524 | 0.75 |
| 1.0 | 4.0 | 452 | 0.8747 | 0.74 |
| 0.6723 | 5.0 | 565 | 0.7782 | 0.75 |
| 0.3818 | 6.0 | 678 | 0.6725 | 0.77 |
| 0.4716 | 7.0 | 791 | 0.5217 | 0.86 |
| 0.1227 | 8.0 | 904 | 0.5807 | 0.85 |
| 0.13 | 9.0 | 1017 | 0.6363 | 0.86 |
| 0.0273 | 10.0 | 1130 | 0.9141 | 0.79 |
| 0.0102 | 11.0 | 1243 | 0.8920 | 0.82 |
| 0.1046 | 12.0 | 1356 | 1.0016 | 0.83 |
| 0.0053 | 13.0 | 1469 | 0.8576 | 0.83 |
| 0.0045 | 14.0 | 1582 | 0.9312 | 0.84 |
| 0.0038 | 15.0 | 1695 | 0.8768 | 0.83 |
| 0.0033 | 16.0 | 1808 | 0.9150 | 0.83 |
| 0.0301 | 17.0 | 1921 | 0.9653 | 0.84 |
| 0.0027 | 18.0 | 2034 | 0.9828 | 0.84 |
| 0.0025 | 19.0 | 2147 | 0.9913 | 0.83 |
| 0.0029 | 20.0 | 2260 | 0.9928 | 0.83 |
Framework versions
- Transformers 4.35.0.dev0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
