WAXAL-mas-lug-w2v-bert-2.0
This model is a fine-tuned version of facebook/w2v-bert-2.0 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4480
- Cer: 0.0798
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: 5e-05
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use 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: 500
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 5.6731 | 1.4925 | 300 | 0.5391 | 0.1132 |
| 1.1121 | 2.9851 | 600 | 0.4349 | 0.0972 |
| 0.8173 | 4.4776 | 900 | 0.4065 | 0.0927 |
| 0.6398 | 5.9701 | 1200 | 0.3909 | 0.0838 |
| 0.3470 | 7.4627 | 1500 | 0.4056 | 0.0805 |
| 0.2364 | 8.9552 | 1800 | 0.4233 | 0.0804 |
| 0.2364 | 10.0 | 2010 | 0.4480 | 0.0798 |
Framework versions
- Transformers 5.13.0
- Pytorch 2.13.0+cu130
- Datasets 5.0.0
- Tokenizers 0.22.2