Skip to content

Latest commit

 

History

History
47 lines (28 loc) · 791 Bytes

README.md

File metadata and controls

47 lines (28 loc) · 791 Bytes

Handwritten digits recognition

This a Flask web app to predict handwritten digits using a deep convolutional neural network built with Keras and trained on the MNIST dataset.

demo gif

Training the model

To re-train a different model or see the existing model, check the training notebook

Running the app locally

Prerequisites

  • Docker
  • Docker Compose

Installation

  1. Clone the repo:
git clone https://github.com/ihebu/digit-recognizer.git
  1. Navigate to the project directory:
cd digit-recognizer
  1. Build the docker images
docker-compose build
  1. Start running the containers
docker-compose up

Usage

to use the app, simply go to http://localhost