An employee churn classification model is invaluable for organizations seeking to retain their valuable workforce and maintain operational efficiency. It helps predict which employees are at risk of leaving the company, enabling proactive intervention through targeted strategies. This project aims to create a classification model that can help in classifying employee churn. Out of the models tuned and compared, Decision Tree Classifier is selected as the best model with an F1-score of 0.818.
Data visualization | Model building | Model deployment
Run locally
- Run
docker pull python:3.10.4-slim
to download the python image image from the Docker Hub registry to local machine. - Run
docker build -t <name>:<tag-optional> .
to build docker image. - Run
docker run -it -p <port>:<port> <name>:<tag-optional>
to build docker image. Can use 8081 as port if available in system.
Deploy
- On the same folder, run
eb init -p docker <name>
to initialize an Elastic Beanstalk environment for a Docker application. Make sure an AWS account has been created asaws-access-id
andaws-secret-key
would be prompted. - Run
eb create <name>
to create a new environment the application. - Run
eb terminate <name>
to stop and delete existing Elastic Beanstalk environment after finish testing.
Dataset: Employee dataset