App mainly consists of 2 Parts.
- Image Classification
- Object Detection
Image classification is divided into 3 parts.
- catVSdog classification - user can upload an image and classify it between cat and dog.
- rickVSmorty classification - user can upload an image and classify it between rick and morty.
- custom Image classifier - user can train his own custom Image classification model just by uploading images into desired classes, and can evaluate the trained model by uploading images.
In Object Detection user can detect objects present in an Image just by uploading the image. This Object Detection is done using a ML model developed using Tensorflow-ObjectDetection API.
- ReactJs - JavaScript library for building user interfaces.
- Material UI - React component and styling library based on 'material design' by google.
- Typescript - Superset of javascript with support for types.
- Flask - Python library for building and managing servers.
- Tensorflow - Python library for building Machine Learning models.
- Keras - Python library, subset of Tensorflow, for building Machine Learning models.
- OpenCV - Python library for manipulating images.
- NodeJS JavaScript Runtime.
- Node Package Manager (npm).
- Git.
- python.
A step by step series of examples that tell you how to get a development env running
git clone https://github.com/StrAnGe-7805/MegaMLProject.git
Frontend libraries Installation
cd frontend
npm install
Backend Libraries Installation
# Backend comes with preinstalled libraries. If you run into any trouble run the following commands to install required libraries.
cd Backend
source bin/activate
pip install -r requirements.txt
To start ReactJs client(frontend)
cd frontend
npm start
To start Flask Server(Backend)
cd Backend
source bin/activate
python app.py
Download catVSdog and rickVSmorty models from Google Drive into paths mentioned below.
- catVSdog -
Backend/Image_Classification/catVSdog/
- rickVSmorty -
Backend/Image_Classification/rickVSmorty/