https://bugzilla.redhat.com/show_bug.cgi?id=2240398
Bug ID: 2240398
Summary: Review Request: pocketfft - A PyTorch FFT math header
Product: Fedora
Version: rawhide
OS: Linux
Status: NEW
Component: Package Review
Severity: medium
Assignee: nobody(a)fedoraproject.org
Reporter: trix(a)redhat.com
QA Contact: extras-qa(a)fedoraproject.org
CC: package-review(a)lists.fedoraproject.org
Target Milestone: ---
Classification: Fedora
Spec URL: https://trix.fedorapeople.org/pocketfft.spec
SRPM URL:
https://trix.fedorapeople.org/pocketfft-1.0%5egit20230214.076cb3d-1.fc40.sr…
PocketFFT for C++
=================
This is a heavily modified implementation of FFTPack [1,2], with the following
advantages:
- Strictly C++11 compliant
- More accurate twiddle factor computation
- Worst case complexity for transform sizes with large prime factors is
`N*log(N)`, because Bluestein's algorithm [3] is used for these cases.
- Supports multidimensional arrays and selection of the axes to be transformed.
- Supports `float`, `double`, and `long double` types.
- Supports fully complex and half-complex (i.e. complex-to-real and
real-to-complex) FFTs. For half-complex transforms, several conventions for
representing the complex-valued side are supported (reduced-size complex
array, FFTPACK-style half-complex format and Hartley transform).
- Supports discrete cosine and sine transforms (Types I-IV)
- Makes use of CPU vector instructions when performing 2D and
higher-dimensional
transforms, if they are available.
- Has a small internal cache for transform plans, which speeds up repeated
transforms of the same length (most significant for 1D transforms).
- Has optional multi-threading support for multidimensional transforms
The intended use in PyTorch is shown in this commit of in the wip spec
https://github.com/trixirt/pytorch-fedora/commit/51922a40f512b8272eaa3c6831…
Reproducible: Always
--
You are receiving this mail because:
You are on the CC list for the bug.
You are always notified about changes to this product and component
https://bugzilla.redhat.com/show_bug.cgi?id=2240398
Report this comment as SPAM: https://bugzilla.redhat.com/enter_bug.cgi?product=Bugzilla&format=report-sp…