Employee-Attrition

The aim of this Project is to predict whether an Employee would face Attrition in this Lockdown period. The model was made as part of a Hackathon Organised by Consulting and Analytics Club, IIT Guwahati.

Competition Link: https://www.kaggle.com/c/summeranalytics2020

Working

The model used is an ensemble of Random Forest Classifier, XgBoost Classifier and LightGBM Classifier and achieves an AUC score of 0.804 on the test dataset. The web application is deployed with Flask as backend and the frontend was built using Bootstrap and some custom CSS.

Demo

Link to Complete Demo: Drive Link

Usage

1. Create a Virtual Environment preferably with Anaconda

conda create -n myenv

2. Activate the virtual environment

conda activate myenv

2. Install the Requirements file

pip install -r requirements.txt

3. Change directory to the cloned directory

4. Run the app.py file using the following command

python app.py

References

1. Deploy Machine Learning Model using Flask by Krish Naik
2. Flask Tutorials by Corey Schafer