donut-base-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: 1.8229

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss
4.1301	1.0	73	4.2545
1.8221	2.0	146	2.3995
0.9406	3.0	219	1.9901
1.3708	4.0	292	1.7363
0.9771	5.0	365	1.6777
0.5417	6.0	438	1.6835
0.9799	7.0	511	1.6810
0.8556	8.0	584	1.6444
0.4318	9.0	657	1.6896
0.3058	10.0	730	1.7384
0.6697	11.0	803	1.7513
0.3883	12.0	876	1.7887
0.166	13.0	949	1.8229

Framework versions

• Transformers 4.34.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1