How to correctly generate training data?

Azazel0203 / ocr_captcha

This project creates a basic web service for solving image-based CAPTCHAs. Using the Flask framework, it allows users to upload CAPTCHA images and employs an Optical Character Recognition (OCR) pipeline to extract the embedded text.

https://ocr.aadarshkr.me/

MIT License

7 stars 2 forks source link

How to correctly generate training data? #1

Closed Roboxkin closed 2 months ago

Roboxkin commented 2 months ago

Hello! How to correctly generate training data?

Please tell us in more detail what files are needed and what structure Thank you

P.S. Not all dependencies are installed P.P.S. labels? How to create them correctly?

ANTIBOT_MODULE_captcha

Azazel0203 commented 2 months ago

Hello! How to correctly generate training data?

Please tell us in more detail what files are needed and what structure Thank you

P.S. Not all dependencies are installed P.P.S. labels? How to create them correctly?

For Images formatted like this The csv file should be formatted like this

And regarding file structure.. Thats under the 'data' folder

Roboxkin commented 2 months ago

Thank you very much for the answer! Thank you very much for the work done and the project provided! I hope that you will not abandon it and will develop it. This is the best thing I've seen on GIT