README.md

metadata

title: Oasis - mlflow server
emoji: 📈
colorFrom: indigo
colorTo: gray
sdk: docker
pinned: false
short_description: A mlflow server for the Oasis project

Oasis - mlflow server

This is a Docker image for running an mlflow server, specifically designed for the Oasis project. It includes all necessary dependencies and configurations to run the server efficiently.

Prerequisites

Before using this Docker image, ensure you have the following prerequisites installed:

• A database server (e.g., PostgreSQL) for the backend store.
• An object storage service (e.g., AWS S3) for the artifact store.

How to use this image

To use this Docker image, you can pull it from the Docker registry and run it with the following command:

docker run -p 7860:7860 \
    -e BACKEND_STORE_URI=postgresql://user:password@db:5432/mlflow \
    -e ARTIFACT_STORE_URI=s3://my-bucket/mlflow \
    -e MLFLOW_SERVER_PORT=7860 \
    -e AWS_ACCESS_KEY_ID=your_access_key_id \
    -e AWS_SECRET_ACCESS_KEY=your_secret_access_key \
    oasis-mlflow

Configuration

You can configure the mlflow server by setting the following environment variables:

• BACKEND_STORE_URI: The URI for the backend store (e.g., PostgreSQL database).
• ARTIFACT_STORE_URI: The URI for the artifact store (e.g., S3 bucket).
• MLFLOW_SERVER_PORT: The port for the mlflow server (optional, defaults to 7860).
• AWS_ACCESS_KEY_ID: Your AWS access key ID (if using S3).
• AWS_SECRET_ACCESS_KEY: Your AWS secret access key (if using S3).

Example configuration

Here is an example of how to configure the mlflow server using environment variables:

export BACKEND_STORE_URI=postgresql://user:password@db:5432/mlflow
export ARTIFACT_STORE_URI=s3://my-bucket/mlflow
export MLFLOW_SERVER_PORT=7860