BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
dipanarya1
Obsidian | Level 7

Hello Guys,

 

Can anyone please provide details on below queries:

 

- Can we include a pre-built Python Model directly into MIP. If yes, Please provide steps for performing for same. If No, What is alternate process to achieve this

- How can we integrate Python Model which is using Python 3.0 or above on SAS environment.

- With Regards to MIP,

                -              Do we need to create any set of input variables or Input SAS Dataset to integrate  Python Model? if yes please provide details

                -              what are the ways to validate execution of the Python Model in SAS MIP?

                -              Please provide details on how can we troubleshoot the python integration

                -              Please provide details steps How can we execute next process which consume the Python model output

 

Thanks,

Dipan Arya

1 ACCEPTED SOLUTION

Accepted Solutions
cwid05
SAS Employee

First, MIP can only run models implemented in MIP. All models in MIP are executing PROC HPRISK code.

 

It depends on what you mean by "pre-built" Python model. 

 

If you mean previously implemented model, then the answer is no. I know some customers are executing existing models (Base SAS, Python, etc.) directly from RFW.

 

If you mean atomic models estimated in Python, then the answer is yes! In fact, there is virtually no difference in the work you have to do to implement a model in MIP that was estimated in Python, SAS, or any other statistical language. All models (regardless of where they were estimated) require the same metadata setup to be implemented in MIP. 

 

Hope this clears up any confusion. Let us know if you have any other questions.

 

-Chris Widman

View solution in original post

1 REPLY 1
cwid05
SAS Employee

First, MIP can only run models implemented in MIP. All models in MIP are executing PROC HPRISK code.

 

It depends on what you mean by "pre-built" Python model. 

 

If you mean previously implemented model, then the answer is no. I know some customers are executing existing models (Base SAS, Python, etc.) directly from RFW.

 

If you mean atomic models estimated in Python, then the answer is yes! In fact, there is virtually no difference in the work you have to do to implement a model in MIP that was estimated in Python, SAS, or any other statistical language. All models (regardless of where they were estimated) require the same metadata setup to be implemented in MIP. 

 

Hope this clears up any confusion. Let us know if you have any other questions.

 

-Chris Widman