zuhorer / mlflow_model_tracking

Simplifying ML Model Deployment and Tracking with Docker and MLflow

Simplifying ML Model Deployment and Tracking with Docker and MLflow

Introduction This project aims to simplify the deployment, tracking, and management of machine learning models using Flask, Docker, and MLflow. It provides an easy-to-use Flask API for model serving and utilizes MLflow for model tracking and performance metrics.

Table of Contents

• Introduction
• Features
• Prerequisites
• Installation
• Usage
• Contributing
• Features
• Flask API for model serving
• MLflow for model tracking and performance metrics
• Dockerized setup for hassle-free deployment

Prerequisites

• Python 3.10
• Docker
• Docker Compose

Installation

• Clone the repository:

  git clone https://github.com/zuhorer/mlflow_model_tracking.git

• Navigate to the project directory:

  cd your-repo-name

• Build the Docker images:

  docker-compose build

• Usage

  Start the containers:

    docker-compose up
  

  Access the Flask API at http://localhost:3000 and the MLflow UI at http://localhost:5000.

  To make predictions, send a POST request to http://localhost:3000/predict.

  To track models, navigate to the MLflow UI and check the metrics and parameters.