donut_synDB_wplus
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: 0.0887
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: 7e-05
- train_batch_size: 5
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 6
- total_train_batch_size: 30
- 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 |
---|---|---|---|
0.4536 | 1.29 | 42 | 0.1805 |
0.2279 | 1.93 | 63 | 0.1423 |
0.1556 | 2.57 | 84 | 0.0746 |
0.1186 | 3.21 | 105 | 0.0720 |
0.0971 | 3.86 | 126 | 0.0954 |
0.083 | 4.5 | 147 | 0.0759 |
0.0695 | 5.14 | 168 | 0.0821 |
0.0707 | 5.79 | 189 | 0.0887 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 24
Unable to determine this model’s pipeline type. Check the
docs
.