results
This model is a fine-tuned version of r-f/wav2vec-english-speech-emotion-recognition on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1011
- Accuracy: 0.9724
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: 0.001
- train_batch_size: 10
- eval_batch_size: 5
- 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: 50
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.4918 | 1.0 | 232 | 1.3591 | 0.3672 |
| 1.0899 | 2.0 | 464 | 0.9012 | 0.5672 |
| 0.9523 | 3.0 | 696 | 1.2430 | 0.4862 |
| 0.8062 | 4.0 | 928 | 0.6423 | 0.7759 |
| 0.5591 | 5.0 | 1160 | 0.5161 | 0.8276 |
| 0.4538 | 6.0 | 1392 | 0.6369 | 0.8069 |
| 0.3527 | 7.0 | 1624 | 0.2526 | 0.9207 |
| 0.3833 | 8.0 | 1856 | 0.2226 | 0.9328 |
| 0.2532 | 9.0 | 2088 | 0.1955 | 0.9466 |
| 0.1296 | 10.0 | 2320 | 0.1860 | 0.9483 |
| 0.144 | 11.0 | 2552 | 0.1885 | 0.9552 |
| 0.1976 | 12.0 | 2784 | 0.1243 | 0.9655 |
| 0.0147 | 13.0 | 3016 | 0.1375 | 0.9655 |
| 0.0149 | 14.0 | 3248 | 0.1061 | 0.9776 |
| 0.0199 | 15.0 | 3480 | 0.1011 | 0.9724 |
Framework versions
- Transformers 4.53.0
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.2
- Downloads last month
- 3