go-mlflow is a Go client library for the MLflow API, including user & permission management as described in the MLFlow Authentication API.

Currently, most APIs are covered, missing mainly registered models APIs.

go-mlflow is compatible with modern Go releases in module mode, with Go installed:

go get github.com/codeocean/go-mlflow

will resolve and add the package to the current development module, along with its dependencies.

Alternatively the same can be achieved if you use import in a package:

import "github.com/codeocean/go-mlflow/mlflow"

and run go get without parameters.

import "github.com/codeocean/go-mlflow/mlflow"

Construct a new MLflow client:

client, err := mlflow.NewClient(nil, "http://localhost:5000")

or if needed, with admin credentials:

client, err := mlflow.NewClient(nil, "http://admin-user:admin-password@localhost:5000")

then use the various services on the client to access different parts of the MLflow API. For example:

// Get an experiment
experiment, err := client.Experiments.Get(context.Background(), "1")

The services of a client divide the API into logical chunks and correspond to the structure of the MLflow API documentation at https://mlflow.org/docs/latest/rest-api.html .

NOTE: Using the context package, one can easily pass cancellation signals and deadlines to various services of the client for handling a request. In case there is no context available, then context.Background() can be used as a starting point.