metadata
license: mit
base_model: naver-clova-ix/donut-base
tags:
- generated_from_trainer
datasets:
- imagefolder
model-index:
- name: donut-base3-sroie
results: []
donut-base3-sroie
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: 1.8932
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: 3
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 12
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.4525 | 1.0 | 49 | 5.8289 |
| 3.7395 | 2.0 | 98 | 3.4073 |
| 3.093 | 3.0 | 147 | 2.8774 |
| 3.5816 | 4.0 | 196 | 2.5408 |
| 2.2642 | 5.0 | 245 | 2.3801 |
| 2.0238 | 6.0 | 294 | 2.1951 |
| 1.7201 | 7.0 | 343 | 2.1006 |
| 1.5596 | 8.0 | 392 | 2.0265 |
| 2.0967 | 9.0 | 441 | 1.9439 |
| 1.9109 | 10.0 | 490 | 1.9280 |
| 1.5129 | 11.0 | 539 | 1.9024 |
| 1.5883 | 12.0 | 588 | 1.8932 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1