wav2vec2-base-finetuned-ks-finetuned-gtzan
This model is a fine-tuned version of addy88/wav2vec2-base-finetuned-ks on the GTZAN dataset. It achieves the following results on the evaluation set:
- Loss: 0.6898
- Accuracy: 0.84
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 adamw_torch 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 |
|---|---|---|---|---|
| 3.7432 | 1.0 | 113 | 1.8692 | 0.32 |
| 2.5892 | 2.0 | 226 | 1.3959 | 0.55 |
| 2.2446 | 3.0 | 339 | 1.1914 | 0.66 |
| 1.4697 | 4.0 | 452 | 1.1078 | 0.65 |
| 1.1533 | 5.0 | 565 | 0.9681 | 0.72 |
| 1.8984 | 6.0 | 678 | 0.8457 | 0.78 |
| 1.3302 | 7.0 | 791 | 0.9367 | 0.74 |
| 0.5157 | 8.0 | 904 | 0.7358 | 0.83 |
| 0.4856 | 9.0 | 1017 | 0.6366 | 0.83 |
| 0.859 | 9.9156 | 1120 | 0.6898 | 0.84 |
Framework versions
- Transformers 4.48.0.dev0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0