baoule_stt_2
This model is a fine-tuned version of abdouaziiz/baoule on the abdouaziz/baoule_waxal_concat dataset. It achieves the following results on the evaluation set:
- Loss: 0.2106
- Wer: 0.1831
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: 12
- eval_batch_size: 12
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 24
- optimizer: Use 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_steps: 50
- training_steps: 48000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.3006 | 4.2105 | 2000 | 0.2106 | 0.1831 |
| 0.0347 | 8.4211 | 4000 | 0.2498 | 0.1819 |
| 0.0197 | 12.6316 | 6000 | 0.2616 | 0.1627 |
| 0.015 | 16.8421 | 8000 | 0.2623 | 0.1588 |
| 0.0108 | 21.0526 | 10000 | 0.2774 | 0.1574 |
Framework versions
- Transformers 4.46.0
- Pytorch 2.9.0+cu128
- Datasets 4.3.0
- Tokenizers 0.20.3