<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
test_new_model-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: 0.3299
- Accuracy: 0.89
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: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.7206 | 1.0 | 56 | 0.7277 | 0.76 |
| 0.3661 | 1.99 | 112 | 0.5018 | 0.82 |
| 0.3592 | 2.99 | 168 | 0.6271 | 0.82 |
| 0.0302 | 4.0 | 225 | 0.4343 | 0.86 |
| 0.1296 | 5.0 | 281 | 0.3727 | 0.87 |
| 0.0016 | 5.99 | 337 | 0.5113 | 0.88 |
| 0.0013 | 6.99 | 393 | 0.3726 | 0.89 |
| 0.0007 | 8.0 | 450 | 0.4790 | 0.89 |
| 0.0115 | 9.0 | 506 | 0.5684 | 0.85 |
| 0.0002 | 9.99 | 562 | 0.3963 | 0.88 |
| 0.0005 | 10.99 | 618 | 0.3753 | 0.88 |
| 0.0002 | 12.0 | 675 | 0.3909 | 0.89 |
| 0.0001 | 13.0 | 731 | 0.3340 | 0.89 |
| 0.0001 | 13.99 | 787 | 0.3281 | 0.89 |
| 0.0001 | 14.93 | 840 | 0.3299 | 0.89 |
Framework versions
- Transformers 4.31.0
- Pytorch 1.12.1+cu116
- Datasets 2.4.0
- Tokenizers 0.12.1
