donut_synDB_big
This model is a fine-tuned version of naver-clova-ix/donut-base on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.0569
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: 3
- total_train_batch_size: 12
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.3353 | 0.92 | 60 | 0.1525 |
0.1389 | 1.38 | 90 | 0.0705 |
0.1055 | 1.85 | 120 | 0.0595 |
0.0701 | 2.31 | 150 | 0.0727 |
0.0547 | 2.77 | 180 | 0.0750 |
0.0454 | 3.23 | 210 | 0.0714 |
0.0371 | 3.69 | 240 | 0.0609 |
0.0332 | 4.15 | 270 | 0.0629 |
0.0269 | 4.62 | 300 | 0.0583 |
0.0233 | 5.08 | 330 | 0.0601 |
0.0219 | 5.54 | 360 | 0.0576 |
0.0227 | 6.0 | 390 | 0.0569 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 23
Unable to determine this model’s pipeline type. Check the
docs
.