How to set Spark configuration in ProcessingStep? #2724
Answered
by
neilmcguigan
neilmcguigan
asked this question in
Help
-
It's easy to set Spark configuration when you RUN a PySparkProcessor:
However, I do not see a way to do so in a SageMaker pipeline, via
How can I do so? TIA |
Beta Was this translation helpful? Give feedback.
Answered by
neilmcguigan
Oct 21, 2021
Replies: 1 comment
-
Figured it out. The answer is to use get_run_args():
|
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
neilmcguigan
# for free
to join this conversation on GitHub.
Already have an account?
# to comment
Figured it out. The answer is to use get_run_args():