layoutlmv3-funsd-soumiya

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: 0.7204
• Precision: 0.8912
• Recall: 0.9116
• F1: 0.9013
• Accuracy: 0.8349

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
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
1.3549	1.3333	100	1.0456	0.6145	0.6801	0.6456	0.6119
0.5821	2.6667	200	0.5192	0.8036	0.8718	0.8363	0.8169
0.4692	4.0	300	0.5310	0.8663	0.8977	0.8817	0.8287
0.3176	5.3333	400	0.5153	0.8789	0.9046	0.8916	0.8336
0.2169	6.6667	500	0.5890	0.8758	0.9041	0.8898	0.8292
0.2001	8.0	600	0.6333	0.8643	0.8957	0.8797	0.8174
0.1315	9.3333	700	0.6813	0.8851	0.9111	0.8979	0.8332
0.1135	10.6667	800	0.6965	0.8859	0.9061	0.8959	0.8353
0.1064	12.0	900	0.7204	0.8912	0.9116	0.9013	0.8349
0.0875	13.3333	1000	0.7205	0.8828	0.9091	0.8957	0.8353

Framework versions

• Transformers 4.44.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.19.1