distilhubert-finetuned-ccm-falsetto
This model is a fine-tuned version of ntu-spml/distilhubert on the Chest Voice and Falsetto Dataset dataset. It achieves the following results on the evaluation set:
- Loss: 0.4966
- Accuracy: 0.9066
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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.4816 | 1.0 | 96 | 0.5584 | 0.7160 |
| 0.4206 | 2.0 | 192 | 0.5590 | 0.8132 |
| 0.0713 | 3.0 | 288 | 0.4010 | 0.8755 |
| 0.0116 | 4.0 | 384 | 0.4312 | 0.8949 |
| 0.0264 | 5.0 | 480 | 0.4611 | 0.9105 |
| 0.0355 | 6.0 | 576 | 0.4798 | 0.9027 |
| 0.0013 | 7.0 | 672 | 0.4728 | 0.9222 |
| 0.001 | 8.0 | 768 | 0.4898 | 0.9027 |
| 0.0008 | 9.0 | 864 | 0.5173 | 0.8949 |
| 0.0007 | 10.0 | 960 | 0.4966 | 0.9066 |
Framework versions
- Transformers 4.49.0
- Pytorch 2.8.0
- Datasets 4.4.1
- Tokenizers 0.21.0