distilhubert-finetuned-ravdessplus
This model is a fine-tuned version of ntu-spml/distilhubert on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.7436
- Accuracy: 0.8403
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_FUSED 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: 0.1
- num_epochs: 12
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.3510 | 1.0 | 1816 | 1.4881 | 0.4375 |
| 1.2669 | 2.0 | 3632 | 1.1122 | 0.6042 |
| 1.0401 | 3.0 | 5448 | 1.0237 | 0.6389 |
| 0.8741 | 4.0 | 7264 | 0.9408 | 0.75 |
| 0.8392 | 5.0 | 9080 | 0.9516 | 0.7708 |
| 1.0055 | 6.0 | 10896 | 0.8242 | 0.8125 |
| 0.9629 | 7.0 | 12712 | 0.8470 | 0.8125 |
| 0.9058 | 8.0 | 14528 | 0.7782 | 0.8194 |
| 1.0546 | 9.0 | 16344 | 0.7809 | 0.8056 |
| 0.8562 | 10.0 | 18160 | 0.7863 | 0.7986 |
| 1.1847 | 11.0 | 19976 | 0.7395 | 0.8264 |
| 0.8621 | 12.0 | 21792 | 0.7436 | 0.8403 |
Framework versions
- Transformers 5.16.1
- Pytorch 2.11.0+cu128
- Datasets 4.0.0
- Tokenizers 0.23.1