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Outdated suggestions cannot be applied.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.
The maxdiff function used in assess_convergence allocates which causes a non negligible time increase in a large-ish LBFGS problem for me.
#889 switched maxdiff to a mapreduce based implementation to support gpu arrays.
Unfortunately the mapreduce now allocates for standard arrays.
This change restores the previous allocation free code for normal Arrays while still allowing gpu arrays using mapreduce