mlflow-minio-psql

Mlflow server

Deploy Mlflow server with Minio and PostgreSQL backend.

Usage

1. Deploy docker-compose up -d
2. Go Minio webUI localhost:9001 and create the bucket bucket-models
3. Test uploading a pickle file

$ cd src
$ python -m venv .env
$ source .env/bin/activate
$ pip install -r requirements.txt
$ python train.py
$ python load.py

Extra commands

# start server
$ mlflow server --host 127.0.0.1 --port 8080
# with sqlite and local artifacts
$ mlflow server --host 127.0.0.1 --port 8080 --backend-store-uri sqlite:///mlflow.db
# with sqlite and artifacts stored on s3
$ mlflow server --host 127.0.0.1 --port 8080 --backend-store-uri sqlite:///mlflow.db --artifacts-destination s3://<bucket>/<path>
# host 0.0.0.0 for docker and k8s
$ mlflow server --host 0.0.0.0 --port 8080 --artifacts-destination s3://<bucket>/<path> --expose-prometheus --workers 1

Referenes

MLflow Common setups

MLflow Authentication

Mlflow CLI

Artifact Stores

Backend Stores

SQLAlchemy Databases engines