Conversation
|
Hi @calebzulawski, We now have a working version of R2C FFT. This version is clearly sub-optimal as I made correctness the main priority. I'll measure and tune the implementation for now. In the interim, I recall you mentioned, inter alia, ideas about the public API and the use of traits. I'm curious to hear your thoughts. Thank you! Best, |
|
Looks good! I think starting with correctness is the way to go. Regarding traits, I think I will probably submit a PR in the future after this has been merged. |
|
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #26 +/- ##
==========================================
+ Coverage 99.16% 99.33% +0.16%
==========================================
Files 8 9 +1
Lines 841 1053 +212
==========================================
+ Hits 834 1046 +212
Misses 7 7 ☔ View full report in Codecov by Sentry. |
- Benchmark throughput/time of Real-to-Complex (R2C) versus the Complex-to-Complex (C2C) FFT on real input data
Hi @calebzulawski, I believe we're finally at a point where we can scrutinize the implementation, make necessary changes, and merge. Please feel free to point out any deficiencies and I can work on fixing them. Thank you!! |
- This is a quick, but hacky and inefficient way to use the twiddle factors generated by the `Planner`, but it works. - Add more details to assert statements for debugging
@calebzulawski Draft for #23