guix-commits
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

11/17: gnu: Add python-pyfma.


From: guix-commits
Subject: 11/17: gnu: Add python-pyfma.
Date: Mon, 4 Dec 2023 16:50:34 -0500 (EST)

civodul pushed a commit to branch master
in repository guix.

commit 880310007288f9901514ae779809c7b4915dfcc9
Author: Felix Gruber <felgru@posteo.net>
AuthorDate: Sat Nov 4 15:02:35 2023 +0000

    gnu: Add python-pyfma.
    
    * gnu/packages/python-science.scm (python-pyfma): New variable.
    
    Change-Id: I354d3d56f2799c8c9ac9b868fcacbf88dcc5c25d
    Signed-off-by: Ludovic Courtès <ludo@gnu.org>
---
 gnu/packages/python-science.scm | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/gnu/packages/python-science.scm b/gnu/packages/python-science.scm
index e836235a6b..604ef6d08e 100644
--- a/gnu/packages/python-science.scm
+++ b/gnu/packages/python-science.scm
@@ -1891,6 +1891,31 @@ Python, from the Sheffield machine learning group.  GPy 
implements a range of
 machine learning algorithms based on GPs.")
     (license license:bsd-3)))
 
+(define-public python-pyfma
+  (package
+    (name "python-pyfma")
+    (version "0.1.6")
+    (source (origin
+              (method git-fetch)   ;for tests
+              (uri (git-reference
+                    (url "https://github.com/nschloe/pyfma";)
+                    (commit version)))
+              (file-name (git-file-name name version))
+              (sha256
+               (base32
+                "12i68jj9n1qj9phjnj6f0kmfhlsd3fqjlk9p6d4gs008azw5m8yn"))))
+    (build-system pyproject-build-system)
+    (propagated-inputs (list python-numpy))
+    (native-inputs (list pybind11 python-pytest))
+    (home-page "https://github.com/nschloe/pyfma";)
+    (synopsis "Fused multiply-add for Python")
+    (description "@code{pyfma} provides an implementation of fused
+multiply-add which computes @code{(x*y) + z} with a single rounding.
+This is useful for dot products, matrix multiplications, polynomial
+evaluations (e.g., with Horner's rule), Newton's method for evaluating
+functions, convolutions, artificial neural networks etc.")
+    (license license:expat)))
+
 (define-public python-pydicom
   (package
     (name "python-pydicom")



reply via email to

[Prev in Thread] Current Thread [Next in Thread]