medium_cross.en
This model is a fine-tuned version of crossdelenna/medium_cross.en on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3034
- Wer: 15.1384
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: 1e-05
- train_batch_size: 22
- eval_batch_size: 22
- seed: 42
- optimizer: Use OptimizerNames.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: 10
- training_steps: 1051
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.664 | 1.2411 | 350 | 0.3998 | 18.2094 |
| 0.4625 | 2.4823 | 700 | 0.3244 | 16.0633 |
| 0.3703 | 3.7234 | 1050 | 0.3034 | 15.1384 |
Framework versions
- Transformers 4.49.0.dev0
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0