Thinking of using ipyparallel to develop a machine learning algorithm on a cluster, mainly using pandas, scikit-learn and numpy.
What are the recommended debugging techniques. Is it possible to attach PyCharm to one of the engines to debug the function being executed?
Or is it possible to run your function where all client, hub, scheduler, single engine is in the same python instance so that you can step through debugging points in PyCharm?
Thanks
