test

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

• Loss: 0.5474
• Precision: 0.8890
• Recall: 0.9071
• F1: 0.8980
• Accuracy: 0.8665

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	0.5976	0.7412	0.8296	0.7829	0.8001
No log	2.67	200	0.5019	0.8259	0.8698	0.8473	0.8269
No log	4.0	300	0.4829	0.8701	0.8982	0.8839	0.8540
No log	5.33	400	0.4490	0.8829	0.9141	0.8982	0.8725
0.5303	6.67	500	0.5120	0.8721	0.9046	0.8881	0.8574
0.5303	8.0	600	0.5212	0.8802	0.9011	0.8905	0.8644
0.5303	9.33	700	0.5447	0.8918	0.9086	0.9001	0.8559
0.5303	10.67	800	0.5304	0.8875	0.9056	0.8965	0.8713
0.5303	12.0	900	0.5496	0.8878	0.9081	0.8978	0.8630
0.1291	13.33	1000	0.5474	0.8890	0.9071	0.8980	0.8665

Framework versions

• Transformers 4.39.0.dev0
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2