https://github.com/abdallaabker/mlops_canadian_forest_fire_prediction

MLOps Zoomcamp Project
https://github.com/abdallaabker/mlops_canadian_forest_fire_prediction

azure-cloud evidently fastapi grafana mlflow prefect

Last synced: about 1 month ago
JSON representation

MLOps Zoomcamp Project

• Host: GitHub
• URL: https://github.com/abdallaabker/mlops_canadian_forest_fire_prediction
• Owner: AbdallaAbker
• License: mit
• Created: 2024-07-13T21:21:55.000Z (4 months ago)
• Default Branch: main
• Last Pushed: 2024-07-21T18:50:23.000Z (4 months ago)
• Last Synced: 2024-10-11T06:03:49.361Z (about 1 month ago)
• Topics: azure-cloud, evidently, fastapi, grafana, mlflow, prefect
• Language: Jupyter Notebook
• Homepage: https://fire-forest-webapp.azurewebsites.net/
• Size: 9.41 MB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
## Welcome to my MLOps Zoomcamp Project :wave:

Project: MLOps Zoomcamp - Canadian Forest Fire Prediction

Project Overview:

The task is to predict the risk of forest fires in various provinces of Canada based on environmental and geographical factors. The problem is a multi-class classification problem where the target variable has four categories: No Fire, Low Risk, Medium Risk, and High Risk. The goal is to develop a machine learning model that can accurately classify the fire risk level given the feature inputs.

Project Use Case:

- Resource allocation for firefighting efforts.

- Implementing precautionary measures during high-risk periods.

This project focuses less on experimentation and more on illustrating various tools and practices in MLOps.

Tools and Technology:

- Cloud: Azure

- Experiment Tracking & Model Registry: MLflow, Azure Blob Container

- Workflow Orchestration: Prefect, Azure Blob Container

- Model Deployment: Azure Container Registry, FastAPI, HTML, CSS, Azure Web App, Streamlit

- Monitoring: Evidently, Grafana, PostgreSQL

- Best Engineering Practices: CI/CD Pipeline (GitHub Actions), Version Control (Git), Unit Tests, Integration Tests, Linting, Code Formatting, Pre-commit Hooks

- Containerization: Docker, Docker Compose

Guide to the project:

- Programming language: Python version 3.10.14

- Operating System: Linux - Ubuntu 20.04

- Clone the project:

    git clone https://github.com/AbdallaAbker/MLOps_Canadian_Forest_Fire_Prediction.git

- Navigate into the project's main directory:

    cd MLOps_Canadian_Forest_Fire_Prediction

- Create a virtual environment and activate it:

    python3 -m venv .venv

    source .venv/bin/activate

- Install the requirements:

    pip install requirements.txt

    (This will install all the necessary dependencies for the project)

Main Directories:

- notebook

- experiment_tracking_model_registry

- workflow_orchestration

- model_deployment

- monitoring

![alt text]()