Small buffer optimization for InvertedAxes #275
Merged
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
adamreichold
force-pushed
the
sbo-inverted-axes
branch
from
a07acc3 to
5553518
Compare
davidhewitt
approved these changes
Feb 14, 2022
src/array.rs
Outdated
| if len < 32 { | ||
| Self::Short(0) | ||
| } else { | ||
| Self::Long(Vec::new()) |
There was a problem hiding this comment.
Is it worth perhaps adding a test with a 40-dimension array, to exercise this code path?
There was a problem hiding this comment.
I tried and actually can't as NumPy seems to be limited to at most 32 dimensions:
In [12]: a = np.zeros(sum(((1,) for _ in range(32)), ()) + (3,))
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
Input In [12], in <module>
----> 1 a = np.zeros(sum(((1,) for _ in range(32)), ()) + (3,))
ValueError: maximum supported dimension for an ndarray is 32, found 33So it would seem that I do not need the Long representation at all?
Since InvertedAxes is basically a set of small integers, we can avoid dynamic allocations by using a word-sized bitmap in the common case of up to 32 axes. This should also not unduly increase code slice as there is no dynamic spilling into the large representation, but the decision is rather taken once based on the dimensionality of the array.
…r InvertedAxes as NumPy does not support it in the first place.
adamreichold
force-pushed
the
sbo-inverted-axes
branch
from
5553518 to
35e0bf2
Compare
davidhewitt
approved these changes
Feb 15, 2022
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Outdated suggestions cannot be applied.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.
Since
InvertedAxesis basically a set of small integers, we can avoid dynamic allocations by using a word-sized bitmap in the common case of up to 32 axes.This should also not unduly increase code slice as there is no dynamic spilling into the large representation, but the decision is rather taken once based on the dimensionality of the array.