distilhubert-finetuned-gtzan
This model is a fine-tuned version of sm226/distilhubert-finetuned-gtzan on the GTZAN dataset. It achieves the following results on the evaluation set:
- Loss: 0.8792
- Accuracy: 0.87
Model description
Base model used for fine tuning: ntu-spml/distilhubert
Intended uses & limitations
More information needed
Training and evaluation data
marsyas/gtzan
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0005
- 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.5395 | 1.0 | 113 | 1.7428 | 0.39 |
| 1.317 | 2.0 | 226 | 1.3209 | 0.53 |
| 1.3358 | 3.0 | 339 | 1.0134 | 0.7 |
| 0.9406 | 4.0 | 452 | 1.3402 | 0.53 |
| 0.5655 | 5.0 | 565 | 0.8318 | 0.74 |
| 0.3066 | 6.0 | 678 | 0.8744 | 0.8 |
| 0.2673 | 7.0 | 791 | 1.0217 | 0.78 |
| 0.2582 | 8.0 | 904 | 1.0037 | 0.78 |
| 0.0319 | 9.0 | 1017 | 0.9541 | 0.85 |
| 0.0011 | 10.0 | 1130 | 0.8792 | 0.87 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3