Replace multiprocessing with joblib, parallelize ESS and TLE #24
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR addresses issue #21
I used this test code to demonstrate the issue and benchmark the solution:
LocalEstimator ESS
On commit 427b1881b before my changes, I time the kNN computation and the ESS computation separately and vary n_jobs. I observe the following using a computer with 40 cpu.
After the PR changes I observe
Conclusions:
The ESS estimator code can be parallelized effectively. The calls to kNN computation have a lot of variance in terms of timing, and are not clearly benefiting from n_jobs>1
GlobalEstimator LPCA
Before these changes
With this PR
Conclusions:
PCA already used parallelism through the
GlobalEstimator
code, butjoblib
provides an additional speedup overmultiprocessing
Additional results on TLE which I added parallelism for, and FisherS which is a GlobalEstimator that has strange parallel behaviour already, are provided in this google doc
https://docs.google.com/document/d/1Jo9VPFQwBaDP9Xj-8xiA_ueULNstXqF554VfFr6Wmeo/edit?usp=sharing