layout3

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

• Loss: 0.6334
• Precision: 0.8935
• Recall: 0.9131
• F1: 0.9032
• Accuracy: 0.8586

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: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.33	100	0.6874	0.7820	0.8073	0.7944	0.7841
No log	2.67	200	0.4485	0.8321	0.8838	0.8571	0.8474
No log	4.0	300	0.4403	0.8579	0.9086	0.8825	0.8414
No log	5.33	400	0.4593	0.8452	0.9056	0.8743	0.8341
0.5531	6.67	500	0.4881	0.8732	0.9170	0.8946	0.8575
0.5531	8.0	600	0.5332	0.8761	0.9101	0.8928	0.8547
0.5531	9.33	700	0.5910	0.8894	0.9106	0.8999	0.8517
0.5531	10.67	800	0.5914	0.8909	0.9131	0.9019	0.8557
0.5531	12.0	900	0.6127	0.9001	0.9180	0.9090	0.8614
0.1245	13.33	1000	0.6334	0.8935	0.9131	0.9032	0.8586

Framework versions

• Transformers 4.32.0
• Pytorch 2.0.0+cu118
• Datasets 2.17.1
• Tokenizers 0.13.2