epochs

Projects · Intermediate · 3–4 hours

Recognise handwriting with PyTorch

Train a small convolutional network that reads handwritten digits, and learn the PyTorch workflow every ML team uses.

What you'll build

A CNN in PyTorch that recognises handwritten digits with about 98% accuracy on images it has never seen, then saves and reloads to make predictions.

You'll need

Python 3 and NumPy where the code imports it. scikit-learn's built-in digits dataset: 1,797 handwritten digits of 8 × 8 pixels, with nothing to download. Install with pip install torch scikit-learn.

Learn first

This project is part of Pro

The first two projects are free. Pro opens every lesson and every project.

See Pro plansFrom ₹119 a month. No auto-renewal.