layoutlmv3-finetuned-sroie

This model is a fine-tuned version of layoutlmv3 on the mp-02/sroie dataset. It achieves the following results on the evaluation set:

• Loss: 0.0651
• Precision: 0.9233
• Recall: 0.9579
• F1: 0.9403
• Accuracy: 0.9815

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-06
• train_batch_size: 6
• eval_batch_size: 6
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 1500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	2.3810	250	0.0957	0.9075	0.9304	0.9188	0.9752
0.1943	4.7619	500	0.0699	0.9260	0.9456	0.9357	0.9805
0.1943	7.1429	750	0.0657	0.9291	0.9513	0.9400	0.9817
0.0485	9.5238	1000	0.0651	0.9233	0.9579	0.9403	0.9815
0.0485	11.9048	1250	0.0661	0.9155	0.9625	0.9384	0.9808
0.0397	14.2857	1500	0.0660	0.9161	0.9632	0.9391	0.9810

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1