multilingual-xlsr-128-F1-New
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9051
- Accuracy: 0.7864
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 |
|---|---|---|---|---|
| 1.1074 | 0.5714 | 400 | 1.0071 | 0.5864 |
| 1.226 | 1.1429 | 800 | 1.0225 | 0.6279 |
| 0.7884 | 1.7143 | 1200 | 0.8577 | 0.6800 |
| 1.1173 | 2.2857 | 1600 | 0.9340 | 0.6986 |
| 0.7001 | 2.8571 | 2000 | 0.8230 | 0.7179 |
| 0.6121 | 3.4286 | 2400 | 0.8605 | 0.7221 |
| 0.5152 | 4.0 | 2800 | 0.7163 | 0.7771 |
| 0.5272 | 4.5714 | 3200 | 0.7712 | 0.7657 |
| 0.2069 | 5.1429 | 3600 | 0.8816 | 0.7464 |
| 0.3369 | 5.7143 | 4000 | 0.8336 | 0.7757 |
| 0.749 | 6.2857 | 4400 | 0.7737 | 0.7936 |
| 0.127 | 6.8571 | 4800 | 0.8558 | 0.7771 |
| 0.2323 | 7.4286 | 5200 | 0.8413 | 0.7871 |
| 0.4587 | 8.0 | 5600 | 0.8553 | 0.7886 |
| 0.4738 | 8.5714 | 6000 | 0.9051 | 0.7864 |
Framework versions
- Transformers 4.56.1
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.0