https://github.com/amine-akrout/mental_health_risk

Training and deploying LightGBM Model with MLFlow, fastapi, App Engine and github actions
https://github.com/amine-akrout/mental_health_risk

appengine ci-cd docker fastapi github-actions google-cloud machine-learning mlflow mlops pycaret python

Last synced: 1 day ago
JSON representation

Training and deploying LightGBM Model with MLFlow, fastapi, App Engine and github actions

• Host: GitHub
• URL: https://github.com/amine-akrout/mental_health_risk
• Owner: amine-akrout
• Created: 2022-01-07T15:14:27.000Z (about 3 years ago)
• Default Branch: master
• Last Pushed: 2024-02-28T11:19:27.000Z (11 months ago)
• Last Synced: 2024-02-28T12:33:32.463Z (11 months ago)
• Topics: appengine, ci-cd, docker, fastapi, github-actions, google-cloud, machine-learning, mlflow, mlops, pycaret, python
• Language: CSS
• Homepage:
• Size: 1.71 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# MLOps with MLflow, FastApi, App Engine and Github Actions

[![Deploy to App Engine](https://github.com/amine-akrout/mental_health_risk/actions/workflows/app_engine.yml/badge.svg)](https://github.com/amine-akrout/mental_health_risk/actions/workflows/app_engine.yml)

## Data Information

Data used in the project can be found on [UCI Machine Learning Repository](https://archive.ics.uci.edu/ml/datasets/Maternal+Health+Risk+Data+Set)

*Data has been collected from different hospitals, community clinics, maternal health cares from the rural areas of Bangladesh through the IoT based risk monitoring system.*

### Attribute Information:

- Age: Any ages in years when a women during pregnant.  

- SystolicBP: Upper value of Blood Pressure in mmHg, another significant attribute during pregnancy.

- DiastolicBP: Lower value of Blood Pressure in mmHg, another significant attribute during pregnancy.

- BS: Blood glucose levels is in terms of a molar concentration, mmol/L.

- HeartRate: A normal resting heart rate in beats per minute.

- Risk Level: Predicted Risk Intensity Level during pregnancy considering the previous attribute.

## Workflow Architecture

![workflow_architecture](./demo/workflow_architecture.png) 

## Requirements

* Python 3.8

* Docker

* Google Cloud Plateform account

## Quick Start

* Clone the repository


git clone https://github.com/amine-akrout/mental_health_risk



* Create a virtual and install requirements


python -m venv

pip install -r requirements.txt



* Train LightGBM using Pycaret and log metrics and artifacts with MLflow


python ./model.py



**MLflow experiment**

![demo_webapp](./demo/mlflow.JPG)  

## Test locally

To test the web app locally using docker, start by building the image from the Dockerfile


docker build --pull --rm -f "Dockerfile" -t mentalhealthrisk:latest "."




docker run -p 8080:8080 mentalhealthrisk



the Web app should be runnining on http://localhost:8080/


## Deploy to Google Cloud App Engine


gcloud app deploy



## CI/CD workflow

Using Github actions and [app_engine.yml](https://github.com/amine-akrout/mental_health_risk/blob/master/.github/workflows/app_engine.yml), we could continuously deploy the web app by simply using the term "deploy" in the commit message when pushing to main branch

## Demo

![demo_webapp](./demo/webapp.gif)  

## References:

* Data source: Marzia Ahmed, Email: [email protected]

Institution: Daffodil International University, Dhaka, Bangladesh.

* Deploy App Engine [Continuous deployment to Google App Engine using GitHub Actions](https://github.com/google-github-actions/deploy-appengine)