Neural Network from Scratch to Predict Handwritten Numbers

https://neuralhandwriting.streamlit.app

Overview

This repository contains the source code for a Streamlit app and a neural network trained on the MNIST Dataset.

Deployment Instructions

Prerequisites

• Google Cloud SDK installed
• Google Cloud Project created
• Required libraries installed (pip install -r requirements.txt)

Configuration

1. Set up your Google Cloud SDK by running:

   gcloud init

2. Deploy the app to Google App Engine:

   gcloud app deploy

3. Access your app using the provided URL after a successful deployment.

Local Development

1. Install dependencies:

   pip install -r requirements.txt

2. Run the Streamlit app locally:

   streamlit run app.py

3. Access the app in your browser at http://localhost:8501.

App Structure

• app.py: Main Streamlit app script.
• training_neural_network.ipynb: Script to train the neural network
• app.yaml: Configuration file for Google App Engine.
• requirements.txt: List of Python dependencies.

Feedback and Contributions

Feel free to provide feedback, report issues, or contribute to the project. We welcome your input!