wav2vec2-base-finetuned-gtzan
This model is a fine-tuned version of facebook/wav2vec2-base on the GTZAN dataset. It achieves the following results on the evaluation set:
- Loss: 0.5074
- Accuracy: 0.88
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.82 | 1.0 | 113 | 1.8198 | 0.4 |
| 1.3566 | 2.0 | 226 | 1.5226 | 0.51 |
| 1.181 | 3.0 | 339 | 1.1743 | 0.71 |
| 0.8429 | 4.0 | 452 | 1.0554 | 0.71 |
| 0.7025 | 5.0 | 565 | 1.2023 | 0.65 |
| 0.517 | 6.0 | 678 | 0.8908 | 0.76 |
| 0.3073 | 7.0 | 791 | 0.6248 | 0.85 |
| 0.2062 | 8.0 | 904 | 0.6668 | 0.85 |
| 0.1973 | 9.0 | 1017 | 0.5620 | 0.84 |
| 0.0877 | 10.0 | 1130 | 0.5074 | 0.88 |
Framework versions
- Transformers 4.57.3
- Pytorch 2.9.0+cu126
- Datasets 3.6.0
- Tokenizers 0.22.1