layoutlmv3-finetuned-invoice

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

• Loss: 0.0019
• Precision: 0.9980
• Recall: 0.9980
• F1: 0.9980
• Accuracy: 0.9998

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	2.0	100	0.1069	0.946	0.9594	0.9527	0.9943
No log	4.0	200	0.0229	0.972	0.9858	0.9789	0.9971
No log	6.0	300	0.0158	0.972	0.9858	0.9789	0.9971
No log	8.0	400	0.0113	0.972	0.9858	0.9789	0.9971
0.1416	10.0	500	0.0103	0.9800	0.9919	0.9859	0.9979
0.1416	12.0	600	0.0047	0.9980	0.9959	0.9970	0.9996
0.1416	14.0	700	0.0035	0.9939	0.9959	0.9949	0.9994
0.1416	16.0	800	0.0044	0.9980	0.9959	0.9970	0.9996
0.1416	18.0	900	0.0027	0.9980	0.9959	0.9970	0.9996
0.0049	20.0	1000	0.0019	0.9980	0.9980	0.9980	0.9998
0.0049	22.0	1100	0.0017	1.0	1.0	1.0	1.0
0.0049	24.0	1200	0.0041	0.9960	0.9980	0.9970	0.9996
0.0049	26.0	1300	0.0033	0.9960	0.9980	0.9970	0.9996
0.0049	28.0	1400	0.0029	0.9960	0.9980	0.9970	0.9996
0.0029	30.0	1500	0.0018	0.9960	0.9980	0.9970	0.9996
0.0029	32.0	1600	0.0019	0.9960	0.9980	0.9970	0.9996
0.0029	34.0	1700	0.0016	0.9980	0.9980	0.9980	0.9998
0.0029	36.0	1800	0.0017	0.9980	0.9980	0.9980	0.9998
0.0029	38.0	1900	0.0018	0.9980	0.9980	0.9980	0.9998
0.0019	40.0	2000	0.0014	0.9980	0.9980	0.9980	0.9998

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1