donut_marriage_sm_560-140_aug_002

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: 0.0472

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: 25
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.3237	1.0	280	0.2879
0.3617	2.0	560	0.1548
0.1783	3.0	840	0.1054
0.0673	4.0	1120	0.0931
0.0738	5.0	1400	0.0890
0.0333	6.0	1680	0.0780
0.018	7.0	1960	0.0740
0.0606	8.0	2240	0.0681
0.0148	9.0	2520	0.0637
0.015	10.0	2800	0.0589
0.0074	11.0	3080	0.0576
0.0003	12.0	3360	0.0543
0.0277	13.0	3640	0.0561
0.0067	14.0	3920	0.0571
0.0003	15.0	4200	0.0592
0.0005	16.0	4480	0.0541
0.0025	17.0	4760	0.0563
0.0032	18.0	5040	0.0503
0.0109	19.0	5320	0.0498
0.0003	20.0	5600	0.0501
0.0025	21.0	5880	0.0504
0.0007	22.0	6160	0.0480
0.0002	23.0	6440	0.0476
0.0029	24.0	6720	0.0474
0.0003	25.0	7000	0.0472

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0