diff options
| author | Yuri Victorovich <yuri@FreeBSD.org> | 2020-04-14 15:30:20 +0000 |
|---|---|---|
| committer | Yuri Victorovich <yuri@FreeBSD.org> | 2020-04-14 15:30:20 +0000 |
| commit | f58fbc30ce6c74310ccc26bffcf38b4cc7eb2005 (patch) | |
| tree | bbb34131b25079b1ce0ecfd9d7d6532e7fcec487 /misc/ncnn | |
| parent | net/kdenetwork: readd kopete as an option (diff) | |
Reported by: portscout (update)
Notes
Notes:
svn path=/head/; revision=531698
Diffstat (limited to 'misc/ncnn')
| -rw-r--r-- | misc/ncnn/Makefile | 26 | ||||
| -rw-r--r-- | misc/ncnn/distinfo | 3 | ||||
| -rw-r--r-- | misc/ncnn/pkg-descr | 11 | ||||
| -rw-r--r-- | misc/ncnn/pkg-plist | 24 |
4 files changed, 64 insertions, 0 deletions
diff --git a/misc/ncnn/Makefile b/misc/ncnn/Makefile new file mode 100644 index 000000000000..ee570be55c54 --- /dev/null +++ b/misc/ncnn/Makefile @@ -0,0 +1,26 @@ +# $FreeBSD$ + +PORTNAME= ncnn +DISTVERSION= 20200413 +CATEGORIES= misc # machine-learning + +MAINTAINER= yuri@FreeBSD.org +COMMENT= High-performance neural network inference framework + +LICENSE= BSD3CLAUSE +LICENSE_FILE= ${WRKSRC}/LICENSE.txt + +USES= cmake compiler:c11 +USE_GITHUB= yes +GH_ACCOUNT= Tencent + +OPTIONS_DEFINE= VULKAN +OPTIONS_DEFAULT= VULKAN + +VULKAN_DESC= Vulkan compute support +VULKAN_CMAKE_BOOL= NCNN_VULKAN +VULKAN_BUILD_DEPENDS= glslangValidator:devel/glslang \ + ${LOCALBASE}/include/vulkan/vulkan.h:devel/vulkan-headers \ + ${LOCALBASE}/lib/libvulkan.so:graphics/vulkan-loader + +.include <bsd.port.mk> diff --git a/misc/ncnn/distinfo b/misc/ncnn/distinfo new file mode 100644 index 000000000000..ec1f979b2f86 --- /dev/null +++ b/misc/ncnn/distinfo @@ -0,0 +1,3 @@ +TIMESTAMP = 1586877414 +SHA256 (Tencent-ncnn-20200413_GH0.tar.gz) = 1ed523bda96cfdf5b5619c6480987b0811e3c4ef62b5d3fe2e8bdca9b8b2fc90 +SIZE (Tencent-ncnn-20200413_GH0.tar.gz) = 10520160 diff --git a/misc/ncnn/pkg-descr b/misc/ncnn/pkg-descr new file mode 100644 index 000000000000..6fb369b8f07a --- /dev/null +++ b/misc/ncnn/pkg-descr @@ -0,0 +1,11 @@ +ncnn is a high-performance neural network inference computing framework +optimized for mobile platforms. ncnn is deeply concerned about its deployment +and use on mobile phones from the beginning of its design. ncnn does not have +third party dependencies. It is cross-platform, and runs faster than all known +open-source frameworks on mobile phone CPUs. Developers can easily deploy deep +learning algorithm models to mobile platforms by using the efficient ncnn +implementation. They can create intelligent apps, and bring the artificial +intelligence to your fingertips. ncnn is currently being used in many Tencent +applications, such as QQ, Qzone, WeChat, Pitu and so on. + +WWW: https://github.com/Tencent/ncnn diff --git a/misc/ncnn/pkg-plist b/misc/ncnn/pkg-plist new file mode 100644 index 000000000000..f11212be296d --- /dev/null +++ b/misc/ncnn/pkg-plist @@ -0,0 +1,24 @@ +include/ncnn/allocator.h +include/ncnn/benchmark.h +include/ncnn/blob.h +include/ncnn/command.h +include/ncnn/cpu.h +include/ncnn/datareader.h +include/ncnn/gpu.h +include/ncnn/layer.h +include/ncnn/layer_shader_type.h +include/ncnn/layer_shader_type_enum.h +include/ncnn/layer_type.h +include/ncnn/layer_type_enum.h +include/ncnn/mat.h +include/ncnn/modelbin.h +include/ncnn/net.h +include/ncnn/opencv.h +include/ncnn/option.h +include/ncnn/paramdict.h +include/ncnn/pipeline.h +include/ncnn/platform.h +lib/cmake/ncnn/ncnn-%%CMAKE_BUILD_TYPE%%.cmake +lib/cmake/ncnn/ncnn.cmake +lib/cmake/ncnn/ncnnConfig.cmake +lib/libncnn.a |