distributed computation via maps interface. #19
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…he main BayesianOptimisation object. Changed a couple loops in the initialisation and maximisation routines to take advantage of possible multi-threading if a suitable map function is passed as mapf. Also allowed for the possibility that in the case of a multi-threaded mapf, one might want to evaluate a couple different options during the optimisation loop at each iteration.
|
This looks very interesting. I'll take a closer look over the the weekend, but I like the idea. Thanks. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Hi,
I've been thinking a little bit about how to adapt this code to run across a large distributed compute cluster (particularly using scoop - http://scoop.readthedocs.org/en/0.7/). I stopped a little short of specifically re-writing to depend on scoop, but I did instead re-do a few bits and pieces to allow the user to define a custom map function through which the initialisation and optimisation routine are run instead of loops. The user can then pick whatever future of map they fancy (which in my case is scoop)
I also made a couple changes to the maximisation routine to make it a bit more flexible in the case of using a futurised map. For example, you might if you're running across a cluster, want to run a few evaluations in parallel at each iteration. I've allowed for that by adding an n_evals argument to maximise().
Anyhow, thought some of these changes might be useful in general.
BW