LayoutLM_4

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: 0.6673
• Precision: 0.675
• Recall: 0.3576
• F1: 0.4675
• Accuracy: 0.8559

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-06
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 1500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	7.14	100	1.2248	0.0	0.0	0.0	0.7818
No log	14.29	200	0.9800	0.0	0.0	0.0	0.7818
No log	21.43	300	0.8988	0.0	0.0	0.0	0.7818
No log	28.57	400	0.8416	0.0	0.0	0.0	0.7818
1.0601	35.71	500	0.8025	0.0	0.0	0.0	0.7818
1.0601	42.86	600	0.7719	0.0	0.0	0.0	0.7818
1.0601	50.0	700	0.7428	0.75	0.0397	0.0755	0.7902
1.0601	57.14	800	0.7225	0.5714	0.0530	0.0970	0.7972
1.0601	64.29	900	0.7107	0.6923	0.1192	0.2034	0.8140
0.6088	71.43	1000	0.6954	0.6444	0.1921	0.2959	0.8308
0.6088	78.57	1100	0.6861	0.6727	0.2450	0.3592	0.8392
0.6088	85.71	1200	0.6800	0.6719	0.2848	0.4	0.8462
0.6088	92.86	1300	0.6694	0.6901	0.3245	0.4414	0.8517
0.6088	100.0	1400	0.6684	0.675	0.3576	0.4675	0.8573
0.5237	107.14	1500	0.6673	0.675	0.3576	0.4675	0.8559

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3