Skip to content

Installation & Setup

This guide will help you set up your machine to run mlflow-experience.

1. Prerequisites

  • Python: Version 3.11 or 3.12 (Python 3.12 recommended).
  • uv: We recommend uv because it installs dependencies in seconds and manages virtual environments automatically.
  • Git: To clone the repository.
  • Docker (Optional): Required only if you want to run the self-hosted MLflow + PostgreSQL + MinIO stack. You can also run MLflow locally without Docker!

2. Clone the Repository

Open your terminal and clone the repository:

git clone https://github.com/sid-146/mlflow-experience.git
cd mlflow-experience

3. Install Dependencies

If you have uv installed, simply run:

uv sync

This single command will: 1. Create a .venv virtual environment if it doesn't already exist. 2. Install all required packages pinned in uv.lock.

To install uv if you don't have it yet: 

# Windows (PowerShell)
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

# macOS / Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

Option B: Using Standard pip

If you prefer using standard Python venv and pip:

python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt

python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

4. Verify Your Environment

Test that the CLI tool runs and displays help:

uv run python -m main --help

python -m main --help

You should see: 

usage: main.py [-h] --config CONFIG [--logging-level LOGGING_LEVEL]

options:
  -h, --help            show this help message and exit
  --config CONFIG       Config File Path
  --logging-level LOGGING_LEVEL
                        Debug Level

5. Setting up MLflow

You have two simple choices for MLflow tracking:

Choice 1: Run MLflow Locally (Zero Docker Needed)

You can run an MLflow server directly in a separate terminal:

uv run mlflow server --backend-store-uri sqlite:///mlflow.db --default-artifact-root ./mlruns --port 5000

Open http://localhost:5000 in your browser. All your experiment runs and metrics will show up here.

Choice 2: Run via Docker Compose

If you have Docker Desktop installed and want a production-like stack with PostgreSQL and MinIO S3 storage:

docker compose up -d