The last few years I have seen several options for integrating pyhton and SAS. These include: Calling SAS from python installed on a local machine or virtual machine, using IOM\SASpy Executing python code from SAS using proc FCMP SWAT to connect to CAS in i VIYA environment Groovy I have been using some saspy, as well as just using SAS to prep data, and read the sas files into python directly or through csv. But in this case, I have a local python installation. This works well for ad hoc task, as I can be quite self sufficient in installing and managing packages. If we were to use python in production, and not just for ad hoc analysis or visualizations, the setup would have to be different (with python installed on the SAS server, or something to that effect). This could allow for using python in production, running code in batch or on request from an application. But in that case, the end user would not be able to install a python package at their own leisure I would assume. Unless there is some container sollution. I can't find a good overview of all the options, including how the different setups look on the server side. Roughly I have the impression that we could either go for local python, with flexibility and no good options for running python in a production settting, or installing python on a VIYA server, where someone with server access would have to manage packages and make sure nothing breaks. Am I missing somehting? And do you know any good overview of all these options for integrating python with SAS?
... View more