README

This repository contains necessary files for the Azure-MLflow integration. The next steps will walk you through the initialization process and the testing procedure. After completion your environment would be ready to play with Azure ML for the logging purposes.

Setup

Install poetry

With your preferred method install poetry tool. Please note, that installing poetry with other that recommended method (curl) could cause troubles. Using pip, homebrew or other similar method could cause PATH problems and lead to errors while setting-up some packages. See installation instructions for poetry.

Install dependencies

Just run:

make install

Environment

You should export environment variables from .envrc file located in the main project directory, but first, fill the missing Azure parameters -> SUBSCRIPTION_ID, WORKSPACE_ID and RESOURCE_GROUP. Run:

source .envrc

Highly recommended to use direnv for automation of this procedure.

Testing

To test the connection with your Azure ML Workspace run:

make test_mlflow

Sources

• panda.jpg - Cute icons created by Smashicons - Flaticon