nexon_jan_2023

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

• Loss: 0.0380
• Precision: 0.9756
• Recall: 0.9302
• F1: 0.9524
• Accuracy: 0.9971

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: 1e-05
• train_batch_size: 2
• eval_batch_size: 2
• 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	16.67	100	0.1998	0.6286	0.5116	0.5641	0.9571
No log	33.33	200	0.0616	0.9756	0.9302	0.9524	0.9971
No log	50.0	300	0.0439	0.9756	0.9302	0.9524	0.9971
No log	66.67	400	0.0404	0.9756	0.9302	0.9524	0.9971
0.1151	83.33	500	0.0389	0.9756	0.9302	0.9524	0.9971
0.1151	100.0	600	0.0380	0.9756	0.9302	0.9524	0.9971
0.1151	116.67	700	0.0378	0.9756	0.9302	0.9524	0.9971
0.1151	133.33	800	0.0379	0.9756	0.9302	0.9524	0.9971
0.1151	150.0	900	0.0378	0.9756	0.9302	0.9524	0.9971
0.009	166.67	1000	0.0378	0.9756	0.9302	0.9524	0.9971
0.009	183.33	1100	0.0378	0.9756	0.9302	0.9524	0.9971
0.009	200.0	1200	0.0379	0.9756	0.9302	0.9524	0.9971
0.009	216.67	1300	0.0379	0.9756	0.9302	0.9524	0.9971
0.009	233.33	1400	0.0379	0.9756	0.9302	0.9524	0.9971
0.0064	250.0	1500	0.0380	0.9756	0.9302	0.9524	0.9971

Framework versions

• Transformers 4.27.0.dev0
• Pytorch 1.13.1+cu116
• Datasets 2.2.2
• Tokenizers 0.13.2