<!-- 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: 0.5748
- Accuracy: 0.89
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: 10
- eval_batch_size: 10
- seed: 40
- gradient_accumulation_steps: 6
- total_train_batch_size: 60
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.2685 | 1.0 | 15 | 1.2199 | 0.71 |
| 1.1248 | 2.0 | 30 | 1.0805 | 0.75 |
| 1.0651 | 3.0 | 45 | 0.9617 | 0.8 |
| 0.9201 | 4.0 | 60 | 0.9439 | 0.76 |
| 0.805 | 5.0 | 75 | 0.8118 | 0.84 |
| 0.6815 | 6.0 | 90 | 0.7881 | 0.84 |
| 0.6421 | 7.0 | 105 | 0.7476 | 0.81 |
| 0.5956 | 8.0 | 120 | 0.6870 | 0.84 |
| 0.4791 | 9.0 | 135 | 0.6403 | 0.88 |
| 0.4411 | 10.0 | 150 | 0.6420 | 0.82 |
| 0.3855 | 11.0 | 165 | 0.5990 | 0.89 |
| 0.3592 | 12.0 | 180 | 0.5927 | 0.87 |
| 0.3254 | 13.0 | 195 | 0.5891 | 0.87 |
| 0.3478 | 14.0 | 210 | 0.5887 | 0.85 |
| 0.2985 | 15.0 | 225 | 0.5748 | 0.89 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3
