wav2vec2-4datasets-maletraining
This model is a fine-tuned version of facebook/wav2vec2-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9597
- Accuracy: 0.7089
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: 3e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- 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_ratio: 0.1
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.5806 | 1.0 | 505 | 1.6373 | 0.3655 |
| 0.9735 | 2.0 | 1010 | 1.0200 | 0.6399 |
| 0.7712 | 3.0 | 1515 | 0.8708 | 0.7018 |
| 0.5792 | 4.0 | 2020 | 0.8921 | 0.7047 |
| 0.5345 | 5.0 | 2525 | 0.9597 | 0.7089 |
Framework versions
- Transformers 4.53.3
- Pytorch 2.6.0+cu124
- Datasets 4.4.1
- Tokenizers 0.21.2