Skip to content

Latest commit

 

History

History
84 lines (56 loc) · 1.54 KB

ReadMe_ENG.md

File metadata and controls

84 lines (56 loc) · 1.54 KB

Data Prediction Server

This project consists of a Flask-based server that provides forecasts about the departure of customers from the RZD. It includes a Reactos web interface and a Python backend that uses a pre-trained model for forecasting.

Project Structure

├── site/ # Frontend code
└── server/ # Backend code
ㅤㅤㅤ├── dist/ # Compiled code
ㅤㅤㅤ├── functions.py
ㅤㅤㅤ├── model.py
ㅤㅤㅤ└── runserver.py \

Technology Stack

  • ReactJS
  • SASS, Bootstrap
  • Python 3.8
  • Flask, Pandas, Numpy
  • LightGBM, XGBoost, CatBoost.

Installation

  1. Clone this repository:
    git clone <repository-url>
    cd <project-directory>
  1. Install frontend dependencies:
    cd site
    npm install
  1. Install backend dependencies:
    cd ../server
    pip install -r requirements.txt

Building the Project

  1. Build the frontend:
    cd site
    npm run build
    npm run postbuild
  1. Build the backend:
    cd ../server
    pyinstaller --name runserver --onefile runserver.py

Running the Application

After building, you can run the application by executing the compiled runserver file in the server/dist directory.

On Windows:

server\dist\runserver.exe

On macOS/Linux:

./server/dist/runserver

The server will start and automatically open a web browser to http://127.0.0.1:5000/index.html.

API Endpoints

  • GET /data/<location>: Returns the top 20 companies for which the forecast of leaving RZD is the highest