LayoutLM_5

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

• Loss: 0.3586
• Precision: 0.8344
• Recall: 0.8344
• F1: 0.8344
• Accuracy: 0.9343

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: 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: 2000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	3.7	100	0.8644	0.0	0.0	0.0	0.7818
No log	7.41	200	0.6214	0.7857	0.0728	0.1333	0.8
No log	11.11	300	0.4714	0.7303	0.4305	0.5417	0.8657
No log	14.81	400	0.4046	0.7955	0.6954	0.7420	0.9189
0.6176	18.52	500	0.3755	0.8194	0.7815	0.8000	0.9301
0.6176	22.22	600	0.3611	0.7935	0.8146	0.8039	0.9245
0.6176	25.93	700	0.3679	0.7848	0.8212	0.8026	0.9245
0.6176	29.63	800	0.3292	0.8289	0.8344	0.8317	0.9357
0.6176	33.33	900	0.3408	0.8289	0.8344	0.8317	0.9315
0.1555	37.04	1000	0.3479	0.8141	0.8411	0.8274	0.9315
0.1555	40.74	1100	0.3491	0.8247	0.8411	0.8328	0.9357
0.1555	44.44	1200	0.3704	0.7888	0.8411	0.8141	0.9245
0.1555	48.15	1300	0.3591	0.8194	0.8411	0.8301	0.9315
0.1555	51.85	1400	0.3420	0.8344	0.8344	0.8344	0.9343
0.0746	55.56	1500	0.3546	0.8421	0.8477	0.8449	0.9357
0.0746	59.26	1600	0.3442	0.8421	0.8477	0.8449	0.9371
0.0746	62.96	1700	0.3687	0.8205	0.8477	0.8339	0.9357
0.0746	66.67	1800	0.3743	0.8258	0.8477	0.8366	0.9343
0.0746	70.37	1900	0.3626	0.8301	0.8411	0.8355	0.9343
0.0502	74.07	2000	0.3586	0.8344	0.8344	0.8344	0.9343

Framework versions

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