bintz_model
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: 2.6698
- Accuracy: 0.0796
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: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- 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: 3
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 4 | 2.6419 | 0.0442 |
| No log | 2.0 | 8 | 2.6441 | 0.0531 |
| 10.0062 | 3.0 | 12 | 2.6446 | 0.0531 |
| 10.0062 | 4.0 | 16 | 2.6600 | 0.0708 |
| 9.7075 | 5.0 | 20 | 2.6613 | 0.0442 |
| 9.7075 | 6.0 | 24 | 2.6624 | 0.0531 |
| 9.7075 | 7.0 | 28 | 2.6674 | 0.0531 |
| 9.9472 | 8.0 | 32 | 2.6698 | 0.0796 |
| 9.9472 | 9.0 | 36 | 2.6672 | 0.0619 |
| 9.6374 | 10.0 | 40 | 2.6671 | 0.0619 |
Framework versions
- Transformers 5.16.1
- Pytorch 2.11.0+cu128
- Datasets 4.8.5
- Tokenizers 0.23.1