whisper-base-finetuned-gtzan
This model is a fine-tuned version of openai/whisper-base on the GTZAN dataset. It achieves the following results on the evaluation set:
- Loss: 0.6867
- Accuracy: 0.87
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: 16
- eval_batch_size: 16
- 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 |
|---|---|---|---|---|
| 0.9075 | 1.0 | 57 | 1.0000 | 0.58 |
| 0.4569 | 2.0 | 114 | 0.6073 | 0.83 |
| 0.3761 | 3.0 | 171 | 0.6410 | 0.8 |
| 0.3049 | 4.0 | 228 | 0.4536 | 0.86 |
| 0.0284 | 5.0 | 285 | 0.5120 | 0.85 |
| 0.0165 | 6.0 | 342 | 0.4856 | 0.89 |
| 0.0087 | 7.0 | 399 | 0.6814 | 0.87 |
| 0.0038 | 8.0 | 456 | 0.7059 | 0.85 |
| 0.0032 | 9.0 | 513 | 0.6831 | 0.87 |
| 0.0034 | 10.0 | 570 | 0.6867 | 0.87 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3