Use custom environment in Azure Machine Learning Designer

G Cocci 211 Reputation points Microsoft Employee
2022-03-18T09:53:51.527+00:00

Hello,

I would like to know if there is the possibility to use a custom environment (created from the AML portal) for the execution of a Python Script Step in the Azure Machine Learning Designer (only using the designer, not using azureml sdk to publish the pipeline from the code).

Thanks,
G

Azure Machine Learning
Azure Machine Learning
An Azure machine learning service for building and deploying models.
2,611 questions
{count} votes

Accepted answer
  1. GiftA-MSFT 11,151 Reputation points
    2022-03-28T19:05:42.677+00:00

    Thanks for your feedback. Based on your comments above, it seems you want to configure a custom environment in AML designer and install unsupported python libraries. These are the supported Custom Environments. However, in AML designer, the execute python script component enables you to write custom python scripts and install python libraries. This particular document shows how to configure execute python script. You can install packages that aren't in the preinstalled list by using the following command:

    import os  
    os.system(f"pip install scikit-misc")  
    

    --- *Kindly Accept Answer if the information helps. Thanks.*

    2 people found this answer helpful.

0 additional answers

Sort by: Most helpful