Wizardify

Table of Contents

• Wizardify
  • Table of Contents
  • About The Project
    • Built With
  • Implementation
  • Getting Started
    • Prerequisites
    • Installation
      • Optional
  • Tensorflow js
  • License

About The Project

A web service that uses Deep Learning AI to analyse pictures and categorize them. This machine-learning project was created to help answer the burning question:

What is my D&D -class? Go to Wizardify-App and find out!

The idea for the project was born from a need to learn about Machine Learning. After some thought, we landed on image analysis and a Dungeons & Dragons based concept. The basic idea was to create an AI model that could analyse a picture and categorize it based on the 12 classes of the game. This model would be deployed into a web app for easy access to all.

Built With

• Python
• Flask
• Bootstrap
• TensorFlow
• Keras

Implementation

The program is Python based web application that is built upon a Flask framework. Our Tensorflow and Keras based AI program is separate from the web application, but the model that is built with it is critical to the app.

The readme for the Tensorflow part can be seen here: TENSOR-README

Flask is used to work out the routing on the app. The front page of the web app can recieve images sent by the user, which are then routed to the model for comparison. After the results are in, the user is redirected to the results page. The visuals of the app use Flask Bootstrap.

Getting Started

To get a local copy up and running follow these simple steps.

Prerequisites

• Install Python 3.x

https://www.python.org/downloads/

Installation

1. Clone the repo

git clone https://github.com/harrinupponen/wizardify.git

2. Create virtual environment:

python3 -m venv venv
OR
python -m venv venv

3. Activate virtual environment

On macOS and Linux:

 source venv/bin/activate

On Windows:

 venv\Scripts\activate

4. Install requirements to new environment

pip install -r requirements.txt

Optional

If you make changes and add new packages remember to update requirements.txt

pip freeze > requirements.txt

Show content of requirements.txt

cat requirements.txt

Tensorflow js

We also made a version using Tensorflow.js, that can be found from TFJS-folder. This version is tested to be working with Heroku and it uses Node.js and express.js for routing. For further dependencies and information about the build, see package.json and server.js.

Prerequisites

• Install npm

npm install npm@latest -g

Installation

1. Clone the repo

git clone https://github.com/harrinupponen/wizardify.git

2. Install npm-packages

npm install

Start the service

npm start

The service will run at http://localhost:81

License

Distributed under the MIT License. See LICENSE for more information.