The current parallelization approach using multiprocessing seems to be a bit cumbersome, and I hope to switch to the much simpler and more lightweightjoblib which seems to be the standard in the ML community (e.g., scikit-learn). I will be making a PR soon.
The current parallelization approach using
multiprocessing
seems to be a bit cumbersome, and I hope to switch to the much simpler and more lightweightjoblib
which seems to be the standard in the ML community (e.g., scikit-learn). I will be making a PR soon.This is also linked to #134.