The writing deskReports on what I buildTwo entries
Report No. 1 · Machine learning · 2024

Teaching a CNN to read handwriting

An optical character recognition model built on EMNIST for my Arkwright Engineering Scholarship, reaching 82% accuracy on characters outside the training set.

By Niran RadiaPython · TensorFlowSequential CNN

The dataset is more than 700,000 images spanning 62 classes of handwritten character, each one paired with its label.

The report covers the aims, how the data was prepared, the augmentation and scheduling used in training, and what the model still gets wrong.

Read the full report
Correspondence invitedcontact@niranradia.comgithub.com/niranradia
Return to the front pageSet in Bodoni Moda and Newsreader.© 2026 Niran Radia