Source: miopen Section: devel Homepage: https://github.com/ROCm/MIOpen Priority: optional Standards-Version: 4.7.0 Vcs-Git: https://salsa.debian.org/rocm-team/miopen.git Vcs-Browser: https://salsa.debian.org/rocm-team/miopen Maintainer: Debian ROCm Team Uploaders: Cordell Bloor , Xuanteng Huang , Christian Kastner , Build-Depends: debhelper-compat (= 13), cmake, rocm-cmake, clang-17, hipcc, libhiprtc-builtins5, clang-tools-17, libclang-common-17-dev, libboost-filesystem-dev, libboost-system-dev, libbz2-dev, librocblas-dev, libsqlite3-dev, nlohmann-json3-dev, pkgconf, libhalf-dev, zlib1g-dev, libgtest-dev, libgmock-dev, chrpath, Rules-Requires-Root: no Package: libmiopen1 Section: libs Architecture: amd64 arm64 ppc64el Depends: ${misc:Depends}, ${shlibs:Depends}, libstdc++-dev, libhiprtc-builtins5, libclang-common-17-dev, Description: ROCm library for machine learning primitives - library MIOpen is a library for machine learning on AMD GPUs. It provides high-performance primitives for a number of operations, including convolutions, batch normalization, pooling layers, and softmax layers for Recurrent Neural Networks. HIP and OpenCL are both supported as programming models. . This package provides the AMD ROCm MIOpen library. Package: libmiopen-dev Section: libdevel Architecture: amd64 arm64 ppc64el Depends: libmiopen1 (= ${binary:Version}), ${misc:Depends}, ${shlibs:Depends}, Description: ROCm library for machine learning primitives - headers MIOpen is a library for machine learning on AMD GPUs. It provides high-performance primitives for a number of operations, including convolutions, batch normalization, pooling layers, and softmax layers for Recurrent Neural Networks. HIP and OpenCL are both supported as programming models. . This package provides the AMD ROCm MIOpen development headers. Package: libmiopen1-tests Section: libdevel Architecture: amd64 arm64 ppc64el Depends: libmiopen1 (= ${binary:Version}), ${misc:Depends}, ${shlibs:Depends}, Description: ROCm library for machine learning primitives - tests MIOpen is a library for machine learning on AMD GPUs. It provides high-performance primitives for a number of operations, including convolutions, batch normalization, pooling layers, and softmax layers for Recurrent Neural Networks. HIP and OpenCL are both supported as programming models. . This package provides the AMD ROCm MIOpen driver and unit tests executable.