bank_statement_extractor-v2

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: 0.0708
• Precision: 0.9963
• Recall: 0.9963
• F1: 0.9963
• Accuracy: 0.9968

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: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	5.0	50	1.6411	0.7203	0.6886	0.7041	0.7302
No log	10.0	100	0.7567	0.9599	0.9634	0.9616	0.9619
No log	15.0	150	0.3852	0.9927	0.9927	0.9927	0.9937
No log	20.0	200	0.2156	0.9927	0.9927	0.9927	0.9937
No log	25.0	250	0.1415	0.9963	0.9963	0.9963	0.9968
No log	30.0	300	0.1087	0.9963	0.9963	0.9963	0.9968
No log	35.0	350	0.0896	0.9963	0.9963	0.9963	0.9968
No log	40.0	400	0.0779	0.9963	0.9963	0.9963	0.9968
No log	45.0	450	0.0720	0.9963	0.9963	0.9963	0.9968
0.4783	50.0	500	0.0708	0.9963	0.9963	0.9963	0.9968

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1