layoutlmv3-finetuned-invoice_ConControl_v1

This model is a fine-tuned version of microsoft/layoutlmv3-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1550
• Precision: 0.8826
• Recall: 0.7390
• F1: 0.8044
• Accuracy: 0.9788

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.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: 300

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	8	0.5664	0.0	0.0	0.0	0.9383
No log	2.0	16	0.3789	0.0	0.0	0.0	0.9383
No log	3.0	24	0.3673	0.0	0.0	0.0	0.9383
No log	4.0	32	0.3596	0.0	0.0	0.0	0.9383
No log	5.0	40	0.3381	0.0	0.0	0.0	0.9383
No log	6.0	48	0.2985	0.0	0.0	0.0	0.9383
No log	7.0	56	0.2590	0.5833	0.0949	0.1633	0.9456
No log	8.0	64	0.2310	0.6629	0.3932	0.4936	0.9585
No log	9.0	72	0.2125	0.5837	0.5085	0.5435	0.9639
No log	10.0	80	0.2003	0.6130	0.5424	0.5755	0.9672
No log	11.0	88	0.1858	0.6667	0.5831	0.6221	0.9696
No log	12.0	96	0.1734	0.7188	0.6237	0.6679	0.9731
No log	13.0	104	0.1719	0.7510	0.6542	0.6993	0.9739
No log	14.0	112	0.1677	0.8048	0.6847	0.7399	0.9756
No log	15.0	120	0.1616	0.8008	0.6949	0.7441	0.9754
No log	16.0	128	0.1587	0.8038	0.7085	0.7532	0.9757
No log	17.0	136	0.1528	0.8275	0.7153	0.7673	0.9783
No log	18.0	144	0.1591	0.8554	0.7220	0.7831	0.9779
No log	19.0	152	0.1449	0.868	0.7356	0.7963	0.9791
No log	20.0	160	0.1568	0.8611	0.7356	0.7934	0.9786
No log	21.0	168	0.1509	0.8645	0.7356	0.7949	0.9788
No log	22.0	176	0.1465	0.8715	0.7356	0.7978	0.9786
No log	23.0	184	0.1446	0.8577	0.7356	0.7920	0.9783
No log	24.0	192	0.1469	0.8645	0.7356	0.7949	0.9788
No log	25.0	200	0.1513	0.8685	0.7390	0.7985	0.9788
No log	26.0	208	0.1554	0.8617	0.7390	0.7956	0.9788
No log	27.0	216	0.1511	0.8651	0.7390	0.7971	0.9786
No log	28.0	224	0.1566	0.872	0.7390	0.8	0.9788
No log	29.0	232	0.1544	0.8685	0.7390	0.7985	0.9788
No log	30.0	240	0.1510	0.8622	0.7424	0.7978	0.9791
No log	31.0	248	0.1589	0.8755	0.7390	0.8015	0.9788
No log	32.0	256	0.1550	0.8826	0.7390	0.8044	0.9788
No log	33.0	264	0.1498	0.8656	0.7424	0.7993	0.9791
No log	34.0	272	0.1539	0.8725	0.7424	0.8022	0.9793
No log	35.0	280	0.1542	0.8690	0.7424	0.8007	0.9793
No log	36.0	288	0.1538	0.8690	0.7424	0.8007	0.9793
No log	37.0	296	0.1543	0.8725	0.7424	0.8022	0.9793
No log	37.5	300	0.1543	0.8725	0.7424	0.8022	0.9793

Framework versions

• Transformers 4.50.0
• Pytorch 2.6.0+cu118
• Datasets 3.4.1
• Tokenizers 0.21.1