blob: 98d2c4276c3282aa58441c8f5a3b6a344dac0d8b (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
|
PORTNAME= ideep
DISTVERSIONPREFIX= v
DISTVERSION= 2.0.0-119
DISTVERSIONSUFFIX= -gb57539e
CATEGORIES= math # neural-networks
MAINTAINER= yuri@FreeBSD.org
COMMENT= Chainer module for numpy-like API and DNN acceleration using MKL-DNN
LICENSE= APACHE20
LICENSE_FILE= ${WRKSRC}/LICENSE
TEST_DEPENDS= ${LOCALBASE}/include/rapidcheck.h:devel/rapidcheck
USES= cmake compiler:c++11-lang
USE_GITHUB= yes
GH_ACCOUNT= intel
USE_LDCONFIG= yes
do-test: # tests broken: https://github.com/intel/ideep/issues/34 and assume bundled mkl-dnn
@cd ${BUILD_WRKSRC} && \
${SETENV} ${CONFIGURE_ENV} ${CMAKE_BIN} ${CMAKE_ARGS} -DBUILD_TESTING:BOOL=ON ${CMAKE_SOURCE_PATH} && \
${SETENV} ${MAKE_ENV} ${MAKE_CMD} ${MAKE_ARGS} ${ALL_TARGET} && \
${SETENV} ${MAKE_ENV} ${MAKE_CMD} ${MAKE_ARGS} test
.include <bsd.port.mk>
|
