---
title: "Number Spotter App | Neural Net Magic"
url: "https://ridwaanhall.com/projects/number-spotter-app-neural-net-magic"
markdown: "https://ridwaanhall.com/projects/number-spotter-app-neural-net-magic.md"
summary: "A Python desktop app where a Multi-Layer Perceptron recognises digits you draw."
---

# Number Spotter App | Neural Net Magic

> A Python desktop app where a Multi-Layer Perceptron recognises digits you draw.

- Category: Machine Learning
- Status: Completed
- Started: October 14, 2023
- Built with: GUI Framework, Python, Neural Networks

## About

### The problem

Neural networks are hard to build intuition for when training happens out of sight and the only output is a final score.

### The approach

I built a desktop GUI where a Multi-Layer Perceptron recognises digits you draw, with seven sliders for its settings and live feedback during training.

- Drawing input with live digit predictions.
- Seven sliders that change the network's configuration.
- Training progress shown as epochs and error.

### The result

The app makes training visible and interactive, a hands-on way to learn how a network's settings affect what it learns.

## Features

- **Adjustable Settings**: Seven sliders change the network and update predictions.
- **MLP Recognition**: A Multi-Layer Perceptron recognises digits 0 to 9.
- **Training View**: Epoch counts and error shown as the network learns.

## Tags

Neural Network, MLP, Handwritten Digit Recognition, Python, GUI, Education, Real-time Prediction, Training Visualization, Machine Learning

## Elsewhere

- This page for people: https://ridwaanhall.com/projects/number-spotter-app-neural-net-magic
- Index of every page as Markdown: [llms.txt](https://ridwaanhall.com/llms.txt)
