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: 0.0026
• Precision: 0.9960
• Recall: 0.9980
• F1: 0.9970
• Accuracy: 0.9996

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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• training_steps: 2000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	2.0	100	0.1316	0.872	0.8844	0.8781	0.9865
No log	4.0	200	0.0248	0.972	0.9858	0.9789	0.9971
No log	6.0	300	0.0167	0.972	0.9858	0.9789	0.9971
No log	8.0	400	0.0134	0.972	0.9858	0.9789	0.9971
0.1408	10.0	500	0.0066	0.9899	0.9980	0.9939	0.9989
0.1408	12.0	600	0.0034	1.0	1.0	1.0	1.0
0.1408	14.0	700	0.0027	1.0	1.0	1.0	1.0
0.1408	16.0	800	0.0024	1.0	1.0	1.0	1.0
0.1408	18.0	900	0.0150	0.9897	0.9777	0.9837	0.9977
0.0051	20.0	1000	0.0030	1.0	0.9980	0.9990	0.9998
0.0051	22.0	1100	0.0035	0.9980	0.9959	0.9970	0.9996
0.0051	24.0	1200	0.0017	0.9980	0.9980	0.9980	0.9998
0.0051	26.0	1300	0.0023	0.9980	0.9980	0.9980	0.9998
0.0051	28.0	1400	0.0024	0.9960	0.9980	0.9970	0.9996
0.0028	30.0	1500	0.0023	0.9960	0.9980	0.9970	0.9996
0.0028	32.0	1600	0.0024	0.9960	0.9980	0.9970	0.9996
0.0028	34.0	1700	0.0026	0.9960	0.9980	0.9970	0.9996
0.0028	36.0	1800	0.0027	0.9960	0.9980	0.9970	0.9996
0.0028	38.0	1900	0.0026	0.9960	0.9980	0.9970	0.9996
0.0019	40.0	2000	0.0026	0.9960	0.9980	0.9970	0.9996

Framework versions

• Transformers 4.47.0.dev0
• Pytorch 2.4.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.1