This is a text recognition model trained on MLT 2019 data. Currently the model is able to recognise the characters in an image upto an accuracy of 51.75% (Based on 5 fold Cross validation).
The data on which this model was trained can be accessed here.
https://drive.google.com/drive/folders/1gQeEafYGVEPR2-ue0TL436Kr2pLkmeyX?usp=sharing
This model uses TF2CV https://pypi.org/project/tf2cv/
In order to install it run the command pip install tf2cv
Credits:
https://github.com/AakashKumarNain for his awesome tutorial on Keras OCR.
https://github.com/bckenstler/CLR/blob/master/clr_callback.py This implementation literally saved my life more times than I count.
Built as part of my internship with Offnote Labs.