donut_pdf_ocr
This model was trained from scratch on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.0443
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: 2e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.0829 | 1.0 | 47 | 0.1157 |
| 0.0184 | 2.0 | 94 | 0.1660 |
| 0.0533 | 3.0 | 141 | 0.0765 |
| 0.0765 | 4.0 | 188 | 0.0530 |
| 0.101 | 5.0 | 235 | 0.0481 |
| 0.0936 | 6.0 | 282 | 0.0494 |
| 0.1032 | 7.0 | 329 | 0.0524 |
| 0.0033 | 8.0 | 376 | 0.0460 |
| 0.0185 | 9.0 | 423 | 0.0440 |
| 0.0044 | 10.0 | 470 | 0.0443 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3