Skip to content

Latest commit

 

History

History
5 lines (4 loc) · 531 Bytes

File metadata and controls

5 lines (4 loc) · 531 Bytes

CharCNN

A convolutional neural network (CNN) that can detect English characters/letters (written by hand) made using Scikit-learn, Keras, and Tensorflow (displayed with OpenCV2).

Some large files like the dataset need to be downloaded externally. requirements.txt has all the packages needed for this to run, and you must recompile the model first for it to succeed.