layoutlmv3-finetuned-cedulas_v3

This model is a fine-tuned version of microsoft/layoutlmv3-base on the data_cedulas_layoutv3 dataset. It achieves the following results on the evaluation set:

• Loss: 0.0832
• Precision: 0.8992
• Recall: 0.9068
• F1: 0.9030
• Accuracy: 0.9817

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: 5e-06
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 2500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	3.12	250	0.7409	0.2850	0.2729	0.2788	0.8614
0.9048	6.25	500	0.3660	0.6222	0.6559	0.6386	0.9393
0.9048	9.38	750	0.2132	0.7492	0.7593	0.7542	0.9544
0.2923	12.5	1000	0.1467	0.7830	0.7949	0.7889	0.9661
0.2923	15.62	1250	0.1172	0.8114	0.8237	0.8175	0.9701
0.1445	18.75	1500	0.1013	0.8560	0.8763	0.8660	0.9766
0.1445	21.88	1750	0.0952	0.8811	0.8915	0.8863	0.9794
0.0956	25.0	2000	0.0876	0.8923	0.8983	0.8953	0.9807
0.0956	28.12	2250	0.0840	0.9005	0.9051	0.9028	0.9811
0.0766	31.25	2500	0.0832	0.8992	0.9068	0.9030	0.9817

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3