Facundo Santiago

HI @ecm200! Just a quick update on this thread, the scenario you mentioned is currently supported in the last version of our integration with MLflow. Actually, in June we introduced...

@manishpyy, if you already have the `azureml-mlflow` plug-in installed, then you just need to make MLflow to point to your workspace tracking URL. Authentication will be handled for you. Once...

Hi @bhogasena, the Azure ML plugin for MLflow tries different authentication methods in a sequence base until one that works successfully is identified. For that, it uses the package `azure-identity`....

Hi @franperezlopez. Unfortunately, we don't currently support MLflow SDK for R for tracking artifacts or models, we only support it for metrics and parameters. You can capture artifacts by saving...

Hi @franperezlopez. The open-source version of MLflow supports R for model registration so you should be able to use it without any problem.

@manishpyy we just added an example about [how to use the MLflow API in Azure ML in Python (Jupyter Notebook)](https://github.com/Azure/azureml-examples/blob/main/notebooks/using-mlflow/using-rest-api/using_mlflow_rest_api.ipynb).

Hi @manishpyy, I think you might be using the MLflow SDK now since you mentioned using the functions `download_artifacts` (which is not available in the REST API - the closer...

> Hi @akshaya-a @santiagxf. Following your instructions, I am able to access the RestAPI that is created on the Tracking Server on my AzureML workspace. However, I want to access...

> Hi @santiagxf , could you help me with an example code to log artifacts for the azureml, since I tried the log_artifacts('run_id', 'local_artifact_path'), and I got this error KeyError:...

> @akshaya-a @santiagxf I thought as much. Thanks a lot for clarifying this. As for the documentation I referred to, https://www.mlflow.org/docs/latest/tracking.html#tracking-ui doesn't explicitly say anything about AzureML, but does say...