kpriv/freebsd-ports
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

math/py-rvlib: Add py-rvlib 0.0.5

Browse source 
Anyone who has used Distributions.jl will tell you how nice the interface is
relative to the "exotic" (the most polite word we can think of) interface to
distributions exposed by scipy.stats. Distributions.jl also brings better
performace, particularly when its methods are used inside loops.

For these reason we've put together rvlib, which mimics the interface of
Distributions.jl, while at the same time attaining similar performance by
exploiting numba.

This package was inspired by Joshua Adelman's (@synapticarbors) blog post
describing how to call the Rmath C library from numba using CFFI, and utilizes
his build script to set up the CFFI interface.

WWW: https://github.com/QuantEcon/rvlib
This commit is contained in:
Po-Chuan Hsieh 2021-09-21 11:06:10 +08:00
parent 4074909b34
commit cc7cd6a770
No known key found for this signature in database
GPG key ID: 9A4BD10F002DD04B
4 changed files with 46 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
math/Makefile
View file

@ -904,6 +904,7 @@
SUBDIR += py-random2
SUBDIR += py-roman
SUBDIR += py-rpy2
SUBDIR += py-rvlib
SUBDIR += py-scikit-umfpack
SUBDIR += py-scs
SUBDIR += py-seaborn

28
math/py-rvlib/Makefile Normal file
View file

@ -0,0 +1,28 @@
# Created by: Po-Chuan Hsieh <sunpoet@FreeBSD.org>
PORTNAME= rvlib
PORTVERSION= 0.0.5
CATEGORIES= math python
MASTER_SITES= CHEESESHOP
PKGNAMEPREFIX= ${PYTHON_PKGNAMEPREFIX}
MAINTAINER= sunpoet@FreeBSD.org
COMMENT= Probability distributions mimicking Distrbutions.jl
LICENSE= BSD3CLAUSE
LICENSE_FILE= ${WRKSRC}/LICENSE.txt
BUILD_DEPENDS= ${PYTHON_PKGNAMEPREFIX}cffi>=1.0.0:devel/py-cffi@${PY_FLAVOR} \
${PYTHON_PKGNAMEPREFIX}yaml>=0:devel/py-yaml@${PY_FLAVOR}
RUN_DEPENDS= ${PYTHON_PKGNAMEPREFIX}cffi>=1.0.0:devel/py-cffi@${PY_FLAVOR} \
${PYTHON_PKGNAMEPREFIX}numba>=0.49:devel/py-numba@${PY_FLAVOR} \
${PYTHON_PKGNAMEPREFIX}numpy>=0,1:math/py-numpy@${PY_FLAVOR} \
${PYTHON_PKGNAMEPREFIX}yaml>=0:devel/py-yaml@${PY_FLAVOR}
USES= python:3.6+
USE_PYTHON= autoplist concurrent distutils
post-install:
${FIND} ${STAGEDIR}${PYTHON_SITELIBDIR} -name '*.so' -exec ${STRIP_CMD} {} +
.include <bsd.port.mk>

3
math/py-rvlib/distinfo Normal file
View file

@ -0,0 +1,3 @@
TIMESTAMP = 1632036828
SHA256 (rvlib-0.0.5.tar.gz) = d79b45bedff32653ed4662da1a273d3b3d70ea960ad4fd8d74509888d7f7f45d
SIZE (rvlib-0.0.5.tar.gz) = 183776

14
math/py-rvlib/pkg-descr Normal file
View file

@ -0,0 +1,14 @@
Anyone who has used Distributions.jl will tell you how nice the interface is
relative to the "exotic" (the most polite word we can think of) interface to
distributions exposed by scipy.stats. Distributions.jl also brings better
performace, particularly when its methods are used inside loops.
For these reason we've put together rvlib, which mimics the interface of
Distributions.jl, while at the same time attaining similar performance by
exploiting numba.
This package was inspired by Joshua Adelman's (@synapticarbors) blog post
describing how to call the Rmath C library from numba using CFFI, and utilizes
his build script to set up the CFFI interface.
WWW: https://github.com/QuantEcon/rvlib