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: 1.0188
- Accuracy: 0.79
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.1062 | 0.96 | 12 | 1.1947 | 0.76 |
| 1.0606 | 2.0 | 25 | 1.1383 | 0.77 |
| 0.956 | 2.96 | 37 | 1.0684 | 0.785 |
| 0.8017 | 4.0 | 50 | 1.0306 | 0.785 |
| 0.7576 | 4.8 | 60 | 1.0188 | 0.79 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2