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: 0.5799
• Precision: 0.8808
• Recall: 0.9106
• F1: 0.8955
• Accuracy: 0.8507

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	1.3333	100	0.6686	0.7452	0.8251	0.7831	0.7535
No log	2.6667	200	0.4724	0.8064	0.8713	0.8376	0.8389
No log	4.0	300	0.4922	0.8612	0.8942	0.8774	0.8481
No log	5.3333	400	0.4632	0.8587	0.8997	0.8787	0.8521
0.544	6.6667	500	0.4850	0.8632	0.9031	0.8827	0.8474
0.544	8.0	600	0.5024	0.8744	0.8992	0.8866	0.8451
0.544	9.3333	700	0.5394	0.8768	0.9155	0.8957	0.8565
0.544	10.6667	800	0.5647	0.8800	0.9146	0.8970	0.8550
0.544	12.0	900	0.5798	0.8847	0.9106	0.8974	0.8545
0.1288	13.3333	1000	0.5799	0.8808	0.9106	0.8955	0.8507

Framework versions

• Transformers 4.41.0.dev0
• Pytorch 2.1.1+cu118
• Datasets 2.15.0
• Tokenizers 0.19.1