ast-finetuned-audioset-10-10-0.4593-finetuned-gtzan
This model is a fine-tuned version of MIT/ast-finetuned-audioset-10-10-0.4593 on the GTZAN dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7292
- Accuracy: 0.9
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: 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
- mixed_precision_training: Native AMP
Training results
| Training Loss |
Epoch |
Step |
Validation Loss |
Accuracy |
| 1.0564 |
1.0 |
113 |
0.6135 |
0.83 |
| 0.3513 |
2.0 |
226 |
0.5031 |
0.87 |
| 0.3781 |
3.0 |
339 |
0.4387 |
0.89 |
| 0.0142 |
4.0 |
452 |
0.6148 |
0.89 |
| 0.188 |
5.0 |
565 |
0.8578 |
0.88 |
| 0.0 |
6.0 |
678 |
1.0513 |
0.89 |
| 0.4527 |
7.0 |
791 |
0.9157 |
0.88 |
| 0.0 |
8.0 |
904 |
1.0898 |
0.85 |
| 0.0 |
9.0 |
1017 |
0.7875 |
0.89 |
| 0.0 |
10.0 |
1130 |
0.7292 |
0.9 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.2
- Datasets 2.17.0
- Tokenizers 0.15.1
