search

MicrosoftDocs / azure-docs

Open source documentation of Microsoft Azure
https://docs.microsoft.com/azure
Creative Commons Attribution 4.0 International
10.29k stars 21.47k forks source link

R Studio on Azure #72723

Closed Jimaras08 closed 3 years ago

Jimaras08 commented 3 years ago

Hi there!

What is the best way of programmatically spinning up an R Studio environment on demand with specific dependencies, e.g. R version, CPU, packages, ... ?

Azure ML CI is a managed VM. Thus, you can't choose the R version you want. One may fire up another CI within R Studio for using a different R environments but that seems to decrease rapid experimentation.

Is DSVM as an unmanaged VM the way to go here? One may create DSVM environments via Azure CLI by using ARM (templates). Perhaps there is something that allows to specify the R version?

AzureVM provides the vm_config function which allows to specify the image to deploy.

Any other ideas?

Thank you!

CHEEKATLAPRADEEP-MSFT commented 3 years ago

@dmangonakis Please help with the link to the document on which the above document feedback request is created.

In case your query is not related to document content, open a new forum thread in Microsoft Q&A or Stack Overflow as it could be better place to get help on your scenario. These forum community members could provide their expert guidance on your scenario based on their experience. Thanks.

Jimaras08 commented 3 years ago

https://docs.microsoft.com/en-us/machine-learning-server/install/machine-learning-server-in-the-cloud

2021-03-25 12_55_25-Machine Learning Server in the Cloud (Virtual Machines) _ Microsoft Docs

ram-msft commented 3 years ago

@dmangonakis Thanks, Please use the R Studio from the compute instance as shown below or you can use the VS code remote compute instance. We can see using the private preview for Azure ML allows the inclusion of a stratup script.

image

image

ram-msft commented 3 years ago

@dmangonakis We will now proceed to close this thread. If there are further questions regarding this matter, please tag me in your reply. We will gladly continue the discussion and we will reopen the issue.