layoutlmv3-finetuned-v6

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: 0.1889
• Precision: 0.9549
• Recall: 0.9322
• F1: 0.9434
• Accuracy: 0.9767

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.3158	100	0.0637	0.9864	0.9864	0.9864	0.9831
No log	2.6316	200	0.1441	0.9613	0.9254	0.9430	0.9640
No log	3.9474	300	0.0723	0.9685	0.9390	0.9535	0.9682
No log	5.2632	400	0.1235	0.9414	0.9254	0.9333	0.9682
0.031	6.5789	500	0.1814	0.9685	0.9390	0.9535	0.9809
0.031	7.8947	600	0.1842	0.9549	0.9322	0.9434	0.9767
0.031	9.2105	700	0.1873	0.9549	0.9322	0.9434	0.9767
0.031	10.5263	800	0.1938	0.9549	0.9322	0.9434	0.9725
0.031	11.8421	900	0.1938	0.9549	0.9322	0.9434	0.9746
0.0003	13.1579	1000	0.1889	0.9549	0.9322	0.9434	0.9767

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0