Home

Awesome

How to Deploy a Keras Model to Production

Overview

This is the code for this video on Youtube by Siraj Raval. We're going to build a model that recognizes handwritten digit images (MNIST). Developed using TensorFlow and the super simple Keras Library. Wrapped into a Webapp using Flask Micro Framework.

Dependencies

sudo pip install -r requirements.txt

Usage

Once dependencies are installed, just run this to see it in your browser.

python app.py

That's it! It's serving a saved Keras model to you via Flask.

Credits

The credits for this code go to moinudeen. I've merely created a wrapper to get people started.