I am trying to parallelize one process in Python which is part of a SAS Intelligent decision Tool (SAS Viya 3.5). The python code file is part of a decision flow and the calculation I am trying to run in parallel is a calculation that includes numpy arrays.
I have tried to using concurrent.futures and its predecessor multiprocess. Both libraries require the script being executed to be directly executed in the interpreter (__name__ == '__main__') , which (based on what I've tried so far) is not possible as the script is nested in a decision flow object.
Has anyone any recommendation on how to parallelize python code within a decision flow running in Micro Analytics Service (MAS)? If not, are you aware of any other way to increase the performance of calculations with numpy arrays?
Thank you in advance!