opencv/modules/dnn/CMakeLists.txt
Yashas Samaga B L 613c12e590 Merge pull request #14827 from YashasSamaga:cuda4dnn-csl-low
CUDA backend for the DNN module

* stub cuda4dnn design

* minor fixes for tests and doxygen

* add csl public api directory to module headers

* add low-level CSL components

* add high-level CSL components

* integrate csl::Tensor into backbone code

* switch to CPU iff unsupported; otherwise, fail on error

* add fully connected layer

* add softmax layer

* add activation layers

* support arbitary rank TensorDescriptor

* pass input wrappers to `initCUDA()`

* add 1d/2d/3d-convolution

* add pooling layer

* reorganize and refactor code

* fixes for gcc, clang and doxygen; remove cxx14/17 code

* add blank_layer

* add LRN layer

* add rounding modes for pooling layer

* split tensor.hpp into tensor.hpp and tensor_ops.hpp

* add concat layer

* add scale layer

* add batch normalization layer

* split math.cu into activations.cu and math.hpp

* add eltwise layer

* add flatten layer

* add tensor transform api

* add asymmetric padding support for convolution layer

* add reshape layer

* fix rebase issues

* add permute layer

* add padding support for concat layer

* refactor and reorganize code

* add normalize layer

* optimize bias addition in scale layer

* add prior box layer

* fix and optimize normalize layer

* add asymmetric padding support for pooling layer

* add event API

* improve pooling performance for some padding scenarios

* avoid over-allocation of compute resources to kernels

* improve prior box performance

* enable layer fusion

* add const layer

* add resize layer

* add slice layer

* add padding layer

* add deconvolution layer

* fix channelwise  ReLU initialization

* add vector traits

* add vectorized versions of relu, clipped_relu, power

* add vectorized concat kernels

* improve concat_with_offsets performance

* vectorize scale and bias kernels

* add support for multi-billion element tensors

* vectorize prior box kernels

* fix address alignment check

* improve bias addition performance of conv/deconv/fc layers

* restructure code for supporting multiple targets

* add DNN_TARGET_CUDA_FP64

* add DNN_TARGET_FP16

* improve vectorization

* add region layer

* improve tensor API, add dynamic ranks

1. use ManagedPtr instead of a Tensor in backend wrapper
2. add new methods to tensor classes
  - size_range: computes the combined size of for a given axis range
  - tensor span/view can be constructed from a raw pointer and shape
3. the tensor classes can change their rank at runtime (previously rank was fixed at compile-time)
4. remove device code from tensor classes (as they are unused)
5. enforce strict conditions on tensor class APIs to improve debugging ability

* fix parametric relu activation

* add squeeze/unsqueeze tensor API

* add reorg layer

* optimize permute and enable 2d permute

* enable 1d and 2d slice

* add split layer

* add shuffle channel layer

* allow tensors of different ranks in reshape primitive

* patch SliceOp to allow Crop Layer

* allow extra shape inputs in reshape layer

* use `std::move_backward` instead of `std::move` for insert in resizable_static_array

* improve workspace management

* add spatial LRN

* add nms (cpu) to region layer

* add max pooling with argmax ( and a fix to limits.hpp)

* add max unpooling layer

* rename DNN_TARGET_CUDA_FP32 to DNN_TARGET_CUDA

* update supportBackend to be more rigorous

* remove stray include from preventing non-cuda build

* include op_cuda.hpp outside condition #if

* refactoring, fixes and many optimizations

* drop DNN_TARGET_CUDA_FP64

* fix gcc errors

* increase max. tensor rank limit to six

* add Interp layer

* drop custom layers; use BackendNode

* vectorize activation kernels

* fixes for gcc

* remove wrong assertion

* fix broken assertion in unpooling primitive

* fix build errors in non-CUDA build

* completely remove workspace from public API

* fix permute layer

* enable accuracy and perf. tests for DNN_TARGET_CUDA

* add asynchronous forward

* vectorize eltwise ops

* vectorize fill kernel

* fixes for gcc

* remove CSL headers from public API

* remove csl header source group from cmake

* update min. cudnn version in cmake

* add numerically stable FP32 log1pexp

* refactor code

* add FP16 specialization to cudnn based tensor addition

* vectorize scale1 and bias1 + minor refactoring

* fix doxygen build

* fix invalid alignment assertion

* clear backend wrappers before allocateLayers

* ignore memory lock failures

* do not allocate internal blobs

* integrate NVTX

* add numerically stable half precision log1pexp

* fix indentation, following coding style,  improve docs

* remove accidental modification of IE code

* Revert "add asynchronous forward"

This reverts commit 1154b9da9da07e9b52f8a81bdcea48cf31c56f70.

* [cmake] throw error for unsupported CC versions

* fix rebase issues

* add more docs, refactor code, fix bugs

* minor refactoring and fixes

* resolve warnings/errors from clang

* remove haveCUDA() checks from supportBackend()

* remove NVTX integration

* changes based on review comments

* avoid exception when no CUDA device is present

* add color code for CUDA in Net::dump
2019-10-21 14:28:00 +03:00

142 lines
5.4 KiB
CMake

if(WINRT)
ocv_module_disable(dnn)
endif()
if(NOT HAVE_PROTOBUF)
ocv_module_disable(opencv_dnn)
endif()
set(the_description "Deep neural network module. It allows to load models from different frameworks and to make forward pass")
ocv_add_dispatched_file_force_all("layers/layers_common" AVX AVX2 AVX512_SKX)
ocv_add_module(dnn opencv_core opencv_imgproc WRAP python java js)
ocv_option(OPENCV_DNN_OPENCL "Build with OpenCL support" HAVE_OPENCL AND NOT APPLE)
if(OPENCV_DNN_OPENCL AND HAVE_OPENCL)
add_definitions(-DCV_OCL4DNN=1)
endif()
ocv_option(OPENCV_DNN_CUDA "Build with CUDA support" HAVE_CUDA AND HAVE_CUBLAS AND HAVE_CUDNN)
if(OPENCV_DNN_CUDA AND HAVE_CUDA AND HAVE_CUBLAS AND HAVE_CUDNN)
add_definitions(-DCV_CUDA4DNN=1)
endif()
ocv_cmake_hook_append(INIT_MODULE_SOURCES_opencv_dnn "${CMAKE_CURRENT_LIST_DIR}/cmake/hooks/INIT_MODULE_SOURCES_opencv_dnn.cmake")
if(MSVC)
add_definitions( -D_CRT_SECURE_NO_WARNINGS=1 )
ocv_warnings_disable(CMAKE_CXX_FLAGS /wd4244 /wd4267 /wd4018 /wd4355 /wd4800 /wd4251 /wd4996 /wd4146
/wd4305 /wd4127 /wd4100 /wd4512 /wd4125 /wd4389 /wd4510 /wd4610
/wd4702 /wd4456 /wd4457 /wd4065 /wd4310 /wd4661 /wd4506
)
else()
ocv_warnings_disable(CMAKE_CXX_FLAGS -Wno-deprecated -Wmissing-prototypes -Wmissing-declarations -Wshadow
-Wunused-parameter -Wsign-compare
)
endif()
if(HAVE_CUDA)
ocv_warnings_disable(CMAKE_CXX_FLAGS -Wundef)
endif()
if(NOT HAVE_CXX11)
ocv_warnings_disable(CMAKE_CXX_FLAGS -Wno-undef) # LANG_CXX11 from protobuf files
endif()
if(APPLE_FRAMEWORK)
ocv_warnings_disable(CMAKE_CXX_FLAGS -Wshorten-64-to-32)
endif()
if(ANDROID)
add_definitions(-DDISABLE_POSIX_MEMALIGN -DTH_DISABLE_HEAP_TRACKING)
endif()
if(NOT BUILD_PROTOBUF)
add_definitions(-DOPENCV_DNN_EXTERNAL_PROTOBUF=1)
endif()
add_definitions(-DHAVE_PROTOBUF=1)
#suppress warnings in autogenerated caffe.pb.* files
ocv_warnings_disable(CMAKE_CXX_FLAGS
/wd4125 /wd4267 /wd4127 /wd4244 /wd4512 /wd4702
/wd4456 /wd4510 /wd4610 /wd4800
/wd4701 /wd4703 # potentially uninitialized local/pointer variable 'value' used
/wd4505 # unreferenced local function has been removed
-wd858 -wd2196
-Winvalid-offsetof # Apple Clang (attr_value.pb.cc)
)
if(PROTOBUF_UPDATE_FILES)
file(GLOB proto_files "${CMAKE_CURRENT_LIST_DIR}/src/tensorflow/*.proto" "${CMAKE_CURRENT_LIST_DIR}/src/caffe/opencv-caffe.proto" "${CMAKE_CURRENT_LIST_DIR}/src/onnx/opencv-onnx.proto")
set(PROTOBUF_GENERATE_CPP_APPEND_PATH ON) # required for tensorflow
protobuf_generate_cpp(fw_srcs fw_hdrs ${proto_files})
else()
file(GLOB fw_srcs "${CMAKE_CURRENT_LIST_DIR}/misc/tensorflow/*.cc" "${CMAKE_CURRENT_LIST_DIR}/misc/caffe/opencv-caffe.pb.cc" "${CMAKE_CURRENT_LIST_DIR}/misc/onnx/opencv-onnx.pb.cc")
file(GLOB fw_hdrs "${CMAKE_CURRENT_LIST_DIR}/misc/tensorflow/*.h" "${CMAKE_CURRENT_LIST_DIR}/misc/caffe/opencv-caffe.pb.h" "${CMAKE_CURRENT_LIST_DIR}/misc/onnx/opencv-onnx.pb.h")
set(fw_inc "${CMAKE_CURRENT_LIST_DIR}/misc/caffe" "${CMAKE_CURRENT_LIST_DIR}/misc/tensorflow" "${CMAKE_CURRENT_LIST_DIR}/misc/onnx")
endif()
set(include_dirs ${fw_inc})
set(sources_options "")
set(libs libprotobuf ${LAPACK_LIBRARIES})
if(OPENCV_DNN_OPENCL AND HAVE_OPENCL)
list(APPEND include_dirs ${OPENCL_INCLUDE_DIRS})
else()
set(sources_options EXCLUDE_OPENCL)
endif()
if(OPENCV_DNN_CUDA AND HAVE_CUDA AND HAVE_CUBLAS AND HAVE_CUDNN)
list(APPEND include_dirs ${CUDA_TOOLKIT_INCLUDE} ${CUDNN_INCLUDE_DIRS})
set(CC_LIST ${CUDA_ARCH_BIN})
separate_arguments(CC_LIST)
foreach(cc ${CC_LIST})
if(cc VERSION_LESS 5.3)
message(FATAL_ERROR "CUDA backend for DNN module requires CC 5.3 or higher. Please remove unsupported architectures from CUDA_ARCH_BIN option.")
endif()
endforeach()
unset(CC_LIST)
else()
set(sources_options ${sources_options} EXCLUDE_CUDA)
endif()
ocv_module_include_directories(${include_dirs})
if(CMAKE_CXX_COMPILER_ID STREQUAL "GNU")
ocv_append_source_files_cxx_compiler_options(fw_srcs "-Wno-suggest-override") # GCC
elseif(CMAKE_CXX_COMPILER_ID STREQUAL "Clang")
ocv_append_source_files_cxx_compiler_options(fw_srcs "-Wno-inconsistent-missing-override") # Clang
endif()
ocv_glob_module_sources(${sources_options} SOURCES ${fw_srcs})
ocv_create_module(${libs} ${INF_ENGINE_TARGET})
ocv_add_samples()
ocv_add_accuracy_tests(${INF_ENGINE_TARGET})
set(perf_path "${CMAKE_CURRENT_LIST_DIR}/perf")
file(GLOB_RECURSE perf_srcs "${perf_path}/*.cpp")
file(GLOB_RECURSE perf_hdrs "${perf_path}/*.hpp" "${perf_path}/*.h")
ocv_add_perf_tests(${INF_ENGINE_TARGET}
FILES test_common "${CMAKE_CURRENT_LIST_DIR}/test/test_common.hpp" "${CMAKE_CURRENT_LIST_DIR}/test/test_common.impl.hpp"
FILES Src ${perf_srcs}
FILES Include ${perf_hdrs}
)
ocv_option(${the_module}_PERF_CAFFE "Add performance tests of Caffe framework" OFF)
ocv_option(${the_module}_PERF_CLCAFFE "Add performance tests of clCaffe framework" OFF)
if(BUILD_PERF_TESTS)
if (${the_module}_PERF_CAFFE)
find_package(Caffe QUIET)
if (Caffe_FOUND)
add_definitions(-DHAVE_CAFFE=1)
ocv_target_link_libraries(opencv_perf_dnn caffe)
endif()
elseif(${the_module}_PERF_CLCAFFE)
find_package(Caffe QUIET)
if (Caffe_FOUND)
add_definitions(-DHAVE_CLCAFFE=1)
ocv_target_link_libraries(opencv_perf_dnn caffe)
endif()
endif()
endif()