[WIP] Improve generation performance in manybody.jl #121
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Outdated suggestions cannot be applied.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.
Performance optimizations:
mat[i, j] = k
is extremely slow and unoptimal when done in a loop. It can be avoided by creating three separate lists for values and indices and callingsparse(is, js, vs, size1, size2)
in the end.coefficient
function does not allocate new arrays (or copy old ones) and accepts any iterable as indices - even single number values (they are iterable, too!).Combined together this yields a 300x performance boost (23s -> 75ms for a full boson basis with 2 particles on 36 modes) for the
manybodyoperator
function.Still, this is nearly not enough for large systems (for example, when there are 3 particles,
manybodypoerator
runs for 12 seconds), so we have to do something better. My simplest proposal is making the occupations vectors sparse to make finding the coefficients easier, but probably there is an even better idea somewhere.