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: 0.4704
- 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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.5809 | 1.0 | 112 | 0.5176 | 0.86 |
0.349 | 2.0 | 225 | 0.5205 | 0.84 |
0.4557 | 3.0 | 337 | 0.6501 | 0.8 |
0.4334 | 4.0 | 450 | 0.4594 | 0.87 |
0.2882 | 5.0 | 562 | 0.4685 | 0.9 |
0.0033 | 6.0 | 675 | 0.5052 | 0.89 |
0.0013 | 7.0 | 787 | 0.5655 | 0.91 |
0.006 | 8.0 | 900 | 0.7625 | 0.86 |
0.2319 | 9.0 | 1012 | 0.5293 | 0.91 |
0.0002 | 10.0 | 1125 | 0.4704 | 0.9 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
- Downloads last month
- 5
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.