metadata
license: mit
base_model: naver-clova-ix/donut-base
tags:
- generated_from_trainer
datasets:
- imagefolder
model-index:
- name: donut-base-full_text_wt_val_1008
results: []
donut-base-full_text_wt_val_1008
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: 0.1215
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: 8
- 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 |
|---|---|---|---|
| 0.3636 | 1.0 | 504 | 0.2522 |
| 0.2144 | 2.0 | 1008 | 0.1862 |
| 0.1898 | 3.0 | 1512 | 0.1540 |
| 0.0952 | 4.0 | 2016 | 0.1459 |
| 0.1096 | 5.0 | 2520 | 0.1319 |
| 0.0742 | 6.0 | 3024 | 0.1265 |
| 0.074 | 7.0 | 3528 | 0.1240 |
| 0.0692 | 8.0 | 4032 | 0.1209 |
| 0.0671 | 9.0 | 4536 | 0.1218 |
| 0.0376 | 10.0 | 5040 | 0.1215 |
Framework versions
- Transformers 4.38.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1