Get Receipt Data with Hugging Face ML Model

This tutorial is about how to use fine-tuned Hugging Face model to extract data from scanned receipt documents. We are executing inference action - passing receipt image, along with words and coordinates to the model. As a result, we get back predictions - class labels assigned to each input. This helps to classify document elements and extract correct data. I share a hint on how to match input words with classified labels. Input words and coordinates are expected to be retrieved from separate OCR.

Colab:

GitHub:

0:00 Introduction
1:50 Sparrow
2:40 Demo in Colab
3:15 Dependencies
3:45 Dataset
5:00 Data structure
6:50 LayoutLMv2 Processor
7:50 LayoutLMv2 Model
8:50 Inference results
12:25 Getting data
14:30 Summary

CONNECT:
- Subscribe to this YouTube channel

#HuggingFace #PyTroch #Python

[sebabrataghosh]

How to create my own dataset which I can use for Layoutlmv2 model building.
Which format I have to keep the annotation dataset
Please tell me.

[venusdev]

This is interesting.
Can the iinvoice be in other languages like Portuguese and Spanish?

[drissdoukkali]

Hello,
Thank you for the great tutorials.
can you please explain how to train a invoice with multiple pages
Many Thanks

[AriefWijayaisMRAW]

How to get list of goods in a structured format ?

[SarikaLozy-yser]

Hey can you please tell, how to use hugging face library after performing OCR on invoice images dataset, as i have only the raw text from OCR .