TimeSeries Forecasting with Microservice Architecture using ReactJs and Flask

This is a personal project to implement Microservice Architecture for TimeSeries forecasting of github repo data. This application collects following data(commits, release, issues, branches, pull) of githbub repository using the github3.py library in the main flask server. After data is fully retrived from the github then it is stored in csv files and json files. There are 3 distinct flask server running except for the main flask server with different timeseries forecasting models i.e.(LSTM, Prophet, Statsmodel). The main flask server sends request to the respective forecasting flask servers to generate forecast based on the users request.

Attach the Architecture Diagram here.

Architecture

![App Architecture]()

Tech Stack

Client: React, ContextApi, Material UI

Server: Flask, CORS, Tensorflow, FB prophet, Statsmodel

Installation

5 Steps to install this project

Step 1

Locate to React app directory and open that directory in the termainl. Run the following commands.

  npm install 
  npm start

Step 2

Locate to backends directory and open that directory in the termainl. Create virtual enviornments and run the following commands.

  activate virtual env
  pip install -r requirements 

Step 3

Locate to fbprophet directory and open that directory in the termainl. Create virtual enviornments and run the following commands.

  activate virtual env
  pip install -r requirements 

Step 4

Locate to lstmapp directory and open that directory in the termainl. Create virtual enviornments and run the following commands.

  activate virtual env
  pip install -r requirements 

Step 5

Locate to Statsmodel directory and open that directory in the termainl. Create virtual enviornments and run the following commands.

  activate virtual env
  pip install -r requirements 

Run Locally

Clone the project

  git clone https://github.com/bvedang/React-Flask-Microservice.git

Go to the project directory

  cd React-Flask-Microservice

Follow the installation instruction. Open 5 different Terminal/Powershell/CMD

Navigate to lstmapp directory in project directory

Terminal/Powershell/CMD : 1 LSTM Flask Server

  activate virtual env
  python app.py

Navigate to fbprophet directory in project directory

Terminal/Powershell/CMD : 2 Prophet Flask Server

  activate virtual env
  python app.py

Navigate to Statsmodel directory in project directory

Terminal/Powershell/CMD : 3 Statsmodel Flask Server

  activate virtual env
  python app.py

Navigate to backends directory in project directory

Terminal/Powershell/CMD : 4 Main Flask Server

  activate virtual env
  python app.py

Navigate to React app directory in project directory

Start the ReactJs server

  npm run start