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: 0.4717
- 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: 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: 20
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.7581 | 1.0 | 56 | 0.7029 | 0.78 |
0.3942 | 1.99 | 112 | 0.4646 | 0.86 |
0.3298 | 2.99 | 168 | 0.3861 | 0.88 |
0.1227 | 4.0 | 225 | 0.4702 | 0.86 |
0.0774 | 5.0 | 281 | 0.4492 | 0.9 |
0.0039 | 5.99 | 337 | 0.4607 | 0.9 |
0.0014 | 6.99 | 393 | 0.5022 | 0.9 |
0.0022 | 8.0 | 450 | 0.4711 | 0.9 |
0.0193 | 9.0 | 506 | 0.5226 | 0.86 |
0.0004 | 9.99 | 562 | 0.6055 | 0.82 |
0.0003 | 10.99 | 618 | 0.4793 | 0.89 |
0.0002 | 12.0 | 675 | 0.5052 | 0.9 |
0.0002 | 13.0 | 731 | 0.4652 | 0.89 |
0.0001 | 13.99 | 787 | 0.4617 | 0.9 |
0.0001 | 14.99 | 843 | 0.4653 | 0.9 |
0.0001 | 16.0 | 900 | 0.4635 | 0.91 |
0.0001 | 17.0 | 956 | 0.4693 | 0.9 |
0.0001 | 17.99 | 1012 | 0.4697 | 0.9 |
0.0001 | 18.99 | 1068 | 0.4715 | 0.9 |
0.0025 | 19.91 | 1120 | 0.4717 | 0.9 |
Framework versions
- Transformers 4.31.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
- Downloads last month
- 56
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.