layoutlmv3-finetuned-invoice

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: 0.2568
• Precision: 0.7955
• Recall: 0.6931
• F1: 0.7407
• Accuracy: 0.9524

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: 2000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	9.0909	100	0.8724	0.0270	0.0099	0.0145	0.7931
No log	18.1818	200	0.3880	0.4299	0.4554	0.4423	0.9126
No log	27.2727	300	0.2870	0.6	0.4158	0.4912	0.9229
No log	36.3636	400	0.3227	0.6389	0.4554	0.5318	0.9242
0.6024	45.4545	500	0.3251	0.6092	0.5248	0.5638	0.9280
0.6024	54.5455	600	0.2188	0.6842	0.6436	0.6633	0.9422
0.6024	63.6364	700	0.2146	0.7159	0.6238	0.6667	0.9447
0.6024	72.7273	800	0.2138	0.8202	0.7228	0.7684	0.9563
0.6024	81.8182	900	0.2128	0.7927	0.6436	0.7104	0.9499
0.0428	90.9091	1000	0.2400	0.7753	0.6832	0.7263	0.9512
0.0428	100.0	1100	0.2498	0.7821	0.6040	0.6816	0.9434
0.0428	109.0909	1200	0.2614	0.7805	0.6337	0.6995	0.9447
0.0428	118.1818	1300	0.2742	0.7821	0.6040	0.6816	0.9447
0.0428	127.2727	1400	0.2744	0.7471	0.6436	0.6915	0.9473
0.0091	136.3636	1500	0.2568	0.7955	0.6931	0.7407	0.9524
0.0091	145.4545	1600	0.2711	0.7701	0.6634	0.7128	0.9486
0.0091	154.5455	1700	0.3043	0.7778	0.6238	0.6923	0.9434
0.0091	163.6364	1800	0.2746	0.7683	0.6238	0.6885	0.9434
0.0091	172.7273	1900	0.2646	0.7955	0.6931	0.7407	0.9524
0.0056	181.8182	2000	0.2681	0.7955	0.6931	0.7407	0.9524

Framework versions

• Transformers 4.41.0.dev0
• Pytorch 2.2.2+cpu
• Datasets 2.19.0
• Tokenizers 0.19.1