Vectorize signal generator functions #252
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
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #252 +/- ##
==========================================
+ Coverage 96.07% 96.14% +0.07%
==========================================
Files 31 31
Lines 3565 3581 +16
==========================================
+ Hits 3425 3443 +18
+ Misses 140 138 -2 ☔ View full report in Codecov by Sentry. |
Collaborator
Author
|
matt-graham
commented
Dec 13, 2024
Comment on lines
82
to
+90
| if precomps is None: | ||
| precomps = generate_precomputes(L, -mm, sampling, nside, L_lower) | ||
| precomps = generate_precomputes( | ||
| L=L, | ||
| spin=-mm, | ||
| sampling=sampling, | ||
| nside=nside, | ||
| forward=False, | ||
| L_lower=L_lower, | ||
| ) |
Collaborator
Author
matt-graham
commented
Dec 13, 2024
Comment on lines
+165
to
+174
| elif n > 0: | ||
| # Generate independent complex coefficients for positive n | ||
| flmn[N - 1 + n, min_el:L, L - 1] = complex_normal( | ||
| rng, L - min_el, var=2 | ||
| ) | ||
| # For m = 0, n > 0 | ||
| # flmn[N - 1 - n, el, L - 1] = (-1)**n * flmn[N - 1 + n, el, L - 1].conj | ||
| flmn[N - 1 - n, min_el:L, L - 1] = (-1) ** n * ( | ||
| flmn[N - 1 + n, min_el:L, L - 1].conj() | ||
| ) |
Collaborator
Author
There was a problem hiding this comment.
Compared to previous implementation, here for m = 0 and abs(n) > 0 we generate complex coefficients rather than real, as the conjugate symmetry condition does not seem to enforce reality in this case - not sure if this is correct though?
CosmoMatt
approved these changes
Dec 18, 2024
Collaborator
CosmoMatt
left a comment
CosmoMatt
left a comment
There was a problem hiding this comment.
Nice! We originally just implemented these functions for testing purposes so didn't bother optimising them, but as you say definitely worth quickly updating them now.
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.
When running some benchmarks with large
LI noticed that the signal generation functions ins2fft.utils.signal_generatorbecome very slow for largeLdue to the nested for loops.This switches to using a vectorized implementation.
Currently implemented forgenerate_flmonly.EDIT: Now implemented for both
generate_flmandgenerate_flmn.