Open-source layoutlmv3-finetuned-invoice model - Efficiently extract invoice information with excellent marking and classification performance

Layoutlmv3 Finetuned Invoice

Developed by oussama

A fine-tuned invoice information extraction model based on LayoutLMv3-base on the SROIE dataset, excelling in token classification tasks

Text Recognition

Transformers

#Invoice information extraction #Document layout understanding #High-precision OCR

Downloads 52

Release Time : 6/23/2022

Model Overview

This model is specifically designed to extract structured information from invoice documents, such as key fields like date, amount, and supplier

Model Features

High-precision invoice information extraction

Achieved perfect performance on the SROIE dataset with precision, recall, and F1-score all at 1.0

Layout-aware

Combines text content and visual layout information for comprehensive understanding

End-to-end training

Supports end-to-end processing from raw documents to structured information

Model Capabilities

Invoice document analysis

Key information extraction

Structured data generation

Document layout understanding

Use Cases

Financial automation

Invoice processing automation

Automatically extracts key information such as supplier, amount, and date from invoices

Reduces manual entry and improves processing efficiency

Document digitization

Paper invoice digitization

Converts scanned paper invoices into structured data

Enables digital management and archiving of documents

🚀 layoutlmv3-finetuned-invoice

This model is a fine - tuned version of [microsoft/layoutlmv3 - base](https://huggingface.co/microsoft/layoutlmv3 - base) on the sroie dataset, achieving excellent performance on evaluation metrics.

🚀 Quick Start

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

Loss: 0.0018
Precision: 1.0
Recall: 1.0
F1: 1.0
Accuracy: 1.0

📚 Documentation

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

🔧 Technical Details

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	0.0967	0.958	0.9716	0.9648	0.9956
No log	4.0	200	0.0222	0.972	0.9858	0.9789	0.9971
No log	6.0	300	0.0171	0.972	0.9858	0.9789	0.9971
No log	8.0	400	0.0136	0.972	0.9858	0.9789	0.9971
0.1307	10.0	500	0.0117	0.964	0.9777	0.9708	0.9962
0.1307	12.0	600	0.0099	0.972	0.9858	0.9789	0.9971
0.1307	14.0	700	0.0094	0.972	0.9858	0.9789	0.9971
0.1307	16.0	800	0.0071	0.9918	0.9838	0.9878	0.9983
0.1307	18.0	900	0.0026	0.9980	0.9980	0.9980	0.9998
0.0089	20.0	1000	0.0018	1.0	1.0	1.0	1.0
0.0089	22.0	1100	0.0016	1.0	1.0	1.0	1.0
0.0089	24.0	1200	0.0015	1.0	0.9980	0.9990	0.9998
0.0089	26.0	1300	0.0015	0.9980	0.9980	0.9980	0.9998
0.0089	28.0	1400	0.0014	0.9980	0.9980	0.9980	0.9998
0.0025	30.0	1500	0.0011	1.0	1.0	1.0	1.0
0.0025	32.0	1600	0.0012	1.0	1.0	1.0	1.0
0.0025	34.0	1700	0.0011	1.0	1.0	1.0	1.0
0.0025	36.0	1800	0.0010	1.0	1.0	1.0	1.0
0.0025	38.0	1900	0.0010	1.0	1.0	1.0	1.0
0.0019	40.0	2000	0.0011	1.0	1.0	1.0	1.0

Framework versions

Transformers 4.21.0.dev0
Pytorch 1.11.0+cu113
Datasets 2.2.2
Tokenizers 0.12.1

📄 Additional Information

Property	Details
Tags	generated_from_trainer
Datasets	sroie
Metrics	precision, recall, f1, accuracy
Model Name	layoutlmv3 - finetuned - invoice
Task	Token Classification
Dataset Name	sroie
Precision	1.0
Recall	1.0
F1	1.0
Accuracy	1.0

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご