Jacob's Convolutional Neural Network (JCNN)

This repository is an exploration into how to build convolutional neural networks from the ground up. It is written in Python3.7 using only NumPy and Flask.

To create a virtual environment for project dependencies:

mkdir venv
python3 -m venv venv

Install all requirements:

pip install -r requirements.txt

Run the demo

Start the debug server:

python3 run.py

Then open index.html in your browser.

Train a model on mnist

python3 server/model/alpha_model.py