wav2vec2-large-F4-New
This model is a fine-tuned version of facebook/wav2vec2-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9070
- Accuracy: 0.8271
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_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 9
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.1837 | 0.5714 | 400 | 1.0977 | 0.5429 |
| 0.5239 | 1.1429 | 800 | 1.2099 | 0.5129 |
| 1.6308 | 1.7143 | 1200 | 0.9599 | 0.6550 |
| 1.1922 | 2.2857 | 1600 | 0.9010 | 0.7007 |
| 0.538 | 2.8571 | 2000 | 1.0914 | 0.6636 |
| 0.1937 | 3.4286 | 2400 | 0.8024 | 0.7700 |
| 0.7111 | 4.0 | 2800 | 0.7487 | 0.7750 |
| 0.8463 | 4.5714 | 3200 | 0.7362 | 0.7943 |
| 0.2734 | 5.1429 | 3600 | 0.7874 | 0.7971 |
| 0.2645 | 5.7143 | 4000 | 1.1901 | 0.7350 |
| 0.5437 | 6.2857 | 4400 | 0.8219 | 0.8100 |
| 0.0947 | 6.8571 | 4800 | 0.8187 | 0.8193 |
| 1.2019 | 7.4286 | 5200 | 0.9061 | 0.8207 |
| 0.0436 | 8.0 | 5600 | 0.8861 | 0.8286 |
| 0.0642 | 8.5714 | 6000 | 0.9070 | 0.8271 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1