distil-ast-audioset-finetuned-cry
This model is a fine-tuned version of bookbot/distil-ast-audioset on the DonateACry dataset. It achieves the following results on the evaluation set:
- Loss: 1.8592
- Accuracy: 0.6364
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: 2e-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: 0.05
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.9595 | 1.0 | 11 | 1.6120 | 0.0909 |
| 1.3053 | 2.0 | 22 | 1.3677 | 0.2727 |
| 0.7604 | 3.0 | 33 | 1.9563 | 0.1818 |
| 0.4351 | 4.0 | 44 | 1.3875 | 0.5455 |
| 0.316 | 5.0 | 55 | 1.7235 | 0.5455 |
| 0.0949 | 6.0 | 66 | 1.5362 | 0.6364 |
| 0.0355 | 7.0 | 77 | 1.8020 | 0.5455 |
| 0.0156 | 8.0 | 88 | 1.8320 | 0.6364 |
| 0.0102 | 9.0 | 99 | 1.9028 | 0.6364 |
| 0.0061 | 10.0 | 110 | 1.8592 | 0.6364 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1