donut-finetuned – AI Model by PLUTO1439 | AlphaNeural AI

You can deploy this model and start earning money today!

PLUTO1439/
donut-finetuned

like

0

vision-encoder-decoder

generated_from_trainer

naver-clova-ix/donut-base

endpoints_compatible

Views

No views yet

donut-finetuned

This model is a fine-tuned version of naver-clova-ix/donut-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.5362

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: 4
eval_batch_size: 4
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 30
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
2.6522	1.0	297	4.0072
1.5818	2.0	594	2.8662
1.2532	3.0	891	2.3891
0.9056	4.0	1188	2.0890
0.9489	5.0	1485	1.9141
0.9369	6.0	1782	1.7803
0.7761	7.0	2079	1.6971
0.8836	8.0	2376	1.6392
0.6898	9.0	2673	1.6093
0.6968	10.0	2970	1.5654
0.7277	11.0	3267	1.5260
0.4091	12.0	3564	1.5575
0.5144	13.0	3861	1.5411
0.5459	14.0	4158	1.5362

Framework versions

Transformers 4.51.1
Pytorch 2.5.1+cu124
Datasets 3.5.0
Tokenizers 0.21.0