WIP to merge upstream changes as of 2018 Dec.

Merge remote-tracking branch 'upstream/master' into merge-upstream-99bd997

Conflicts:
	python/caffe/io.py
	src/caffe/layer_factory.cpp
	src/caffe/proto/caffe.proto

.github/ISSUE_TEMPLATE.md

@@ -1,19 +1,34 @@
-Please use the [caffe-users list](https://groups.google.com/forum/#!forum/caffe-users) for usage, installation, or modeling questions, or other requests for help.
-_Do not post such requests to Issues._ Doing so interferes with the development of Caffe.
+## Important - read before submitting
 
-Please read the [guidelines for contributing](https://github.com/BVLC/caffe/blob/master/CONTRIBUTING.md) before submitting this issue.
+*Please read the [guidelines for contributing](https://github.com/BVLC/caffe/blob/master/CONTRIBUTING.md) before submitting this issue!*
+
+*Please do not post installation, build, usage, or modeling questions, or other requests for help to Issues.*
+Use the [caffe-users list](https://groups.google.com/forum/#!forum/caffe-users) instead.
+This helps developers maintain a clear, uncluttered, and efficient view of the state of Caffe.
 
 ### Issue summary
 
 
 ### Steps to reproduce
 
-If you are having difficulty building Caffe or training a model, please ask the caffe-users mailing list. If you are reporting a build error that seems to be due to a bug in Caffe, please attach your build configuration (either Makefile.config or CMakeCache.txt) and the output of the make (or cmake) command.
 
-### Your system configuration
-Operating system:
-Compiler:
-CUDA version (if applicable):
-CUDNN version (if applicable):
-BLAS:
-Python or MATLAB version (for pycaffe and matcaffe respectively):
+### Tried solutions
+
+
+### System configuration
+
+* Operating system: 
+* Compiler: 
+* CUDA version (if applicable): 
+* CUDNN version (if applicable): 
+* BLAS: 
+* Python version (if using pycaffe): 
+* MATLAB version (if using matcaffe): 
+
+### Issue checklist
+
+- [ ] read the guidelines and removed the first paragraph
+- [ ] written a short summary and detailed steps to reproduce
+- [ ] explained how solutions to related problems failed (tick if found none)
+- [ ] filled system configuration
+- [ ] attached relevant logs/config files (tick if not applicable)

CMakeLists.txt

@@ -42,6 +42,9 @@ caffe_option(USE_LMDB "Build with lmdb" ON)
 caffe_option(ALLOW_LMDB_NOLOCK "Allow MDB_NOLOCK when reading LMDB files (only if necessary)" OFF)
 caffe_option(USE_OPENMP "Link with OpenMP (when your BLAS wants OpenMP and you get linker errors)" OFF)
 
+# This code is taken from https://github.com/sh1r0/caffe-android-lib
+caffe_option(USE_HDF5 "Build with hdf5" ON)
+
 # ---[ Dependencies
 include(cmake/Dependencies.cmake)
 

CONTRIBUTING.md

@@ -1,21 +1,63 @@
 # Contributing
 
+Below you will find a collection of guidelines for submitting issues as well as contributing code to the Caffe repository.
+Please read those before starting an issue or a pull request.
+
 ## Issues
 
 Specific Caffe design and development issues, bugs, and feature requests are maintained by GitHub Issues.
 
-_Please do not post usage, installation, or modeling questions, or other requests for help to Issues._
-Use the [caffe-users list](https://groups.google.com/forum/#!forum/caffe-users) instead. This helps developers maintain a clear, uncluttered, and efficient view of the state of Caffe.
-
-When reporting a bug, it's most helpful to provide the following information, where applicable:
+*Please do not post installation, build, usage, or modeling questions, or other requests for help to Issues.*
+Use the [caffe-users list](https://groups.google.com/forum/#!forum/caffe-users) instead.
+This helps developers maintain a clear, uncluttered, and efficient view of the state of Caffe.
+See the chapter [caffe-users](#caffe-users) below for guidance on posting to the users list.
 
-* What steps reproduce the bug?
-* Can you reproduce the bug using the latest [master](https://github.com/BVLC/caffe/tree/master), compiled with the `DEBUG` make option?
-* What hardware and operating system/distribution are you running?
+When reporting an issue, it's most helpful to provide the following information, where applicable:
+* How does the problem look like and what steps reproduce it?
+* Can you reproduce it using the latest [master](https://github.com/BVLC/caffe/tree/master), compiled with the `DEBUG` make option?
+* What hardware and software are you running? In particular:
+	* GPU make and model, if relevant,
+	* operating system/distribution,
+	* compiler; please also post which version (for example, with GCC run `gcc --version` to check),
+	* CUDA version, if applicable (run `nvcc --version` to check),
+	* cuDNN version, if applicable (version number is stored in `cudnn.h`, look for lines containing `CUDNN_MAJOR`, `CUDNN_MINOR` and `CUDNN_PATCHLEVEL`),
+	* BLAS library,
+	* Python version, if relevant,
+	* MATLAB version, if relevant.
+* **What have you already tried** to solve the problem? How did it fail? Are there any other issues related to yours?
+* If this is not a build-related issue, does your installation pass `make runtest`?
 * If the bug is a crash, provide the backtrace (usually printed by Caffe; always obtainable with `gdb`).
+* If you are reporting a build error that seems to be due to a bug in Caffe, please attach your build configuration (either Makefile.config or CMakeCache.txt) and the output of the make (or cmake) command.
+
+If only a small portion of the code/log is relevant to your issue, you may paste it directly into the post, preferably using Markdown syntax for code block: triple backtick ( \`\`\` ) to open/close a block.
+In other cases (multiple files, or long files), please **attach** them to the post - this greatly improves readability.
+
+If the problem arises during a complex operation (e.g. large script using pycaffe, long network prototxt), please reduce the example to the minimal size that still causes the error.
+Also, minimize influence of external modules, data etc. - this way it will be easier for others to understand and reproduce your issue, and eventually help you.
+Sometimes you will find the root cause yourself in the process.
 
 Try to give your issue a title that is succinct and specific. The devs will rename issues as needed to keep track of them.
 
+## Caffe-users
+
+Before you post to the [caffe-users list](https://groups.google.com/forum/#!forum/caffe-users), make sure you look for existing solutions.
+The Caffe community has encountered and found solutions to countless problems - benefit from the collective experience.
+Recommended places to look:
+* the [users list](https://groups.google.com/forum/#!forum/caffe-users) itself,
+* [`caffe`](https://stackoverflow.com/questions/tagged/caffe) tag on StackOverflow,
+* [GitHub issues](https://github.com/BVLC/caffe/issues) tracker (some problems have been answered there),
+* the public [wiki](https://github.com/BVLC/caffe/wiki),
+* the official [documentation](http://caffe.berkeleyvision.org/).
+
+Found a post/issue with your exact problem, but with no answer?
+Don't just leave a "me too" message - provide the details of your case.
+Problems with more available information are easier to solve and attract good attention.
+
+When posting to the list, make sure you provide as much relevant information as possible - recommendations for an issue report (see above) are a good starting point.  
+*Please make it very clear which version of Caffe you are using, especially if it is a fork not maintained by BVLC.*
+
+Formatting recommendations hold: paste short logs/code fragments into the post (use fixed-width text for them), **attach** long logs or multiple files.
+
 ## Pull Requests
 
 Caffe welcomes all contributions.

Makefile

@@ -178,11 +178,13 @@ ifneq ($(CPU_ONLY), 1)
 	LIBRARIES := cudart cublas curand
 endif
 
-LIBRARIES += glog gflags protobuf boost_system boost_filesystem m hdf5_hl hdf5
+LIBRARIES += glog gflags protobuf boost_system boost_filesystem m
 
 # handle IO dependencies
 USE_LEVELDB ?= 1
 USE_LMDB ?= 1
+# This code is taken from https://github.com/sh1r0/caffe-android-lib
+USE_HDF5 ?= 1
 USE_OPENCV ?= 1
 
 ifeq ($(USE_LEVELDB), 1)
@@ -191,6 +193,10 @@ endif
 ifeq ($(USE_LMDB), 1)
 	LIBRARIES += lmdb
 endif
+# This code is taken from https://github.com/sh1r0/caffe-android-lib
+ifeq ($(USE_HDF5), 1)
+	LIBRARIES += hdf5_hl hdf5
+endif
 ifeq ($(USE_OPENCV), 1)
 	LIBRARIES += opencv_core opencv_highgui opencv_imgproc
 
@@ -347,6 +353,10 @@ ifeq ($(ALLOW_LMDB_NOLOCK), 1)
 	COMMON_FLAGS += -DALLOW_LMDB_NOLOCK
 endif
 endif
+# This code is taken from https://github.com/sh1r0/caffe-android-lib
+ifeq ($(USE_HDF5), 1)
+	COMMON_FLAGS += -DUSE_HDF5
+endif
 
 # CPU-only configuration
 ifeq ($(CPU_ONLY), 1)
@@ -577,7 +587,7 @@ $(STATIC_NAME): $(OBJS) | $(LIB_BUILD_DIR)
 	@ echo AR -o $@
 	$(Q)ar rcs $@ $(OBJS)
 
-$(BUILD_DIR)/%.o: %.cpp | $(ALL_BUILD_DIRS)
+$(BUILD_DIR)/%.o: %.cpp $(PROTO_GEN_HEADER) | $(ALL_BUILD_DIRS)
 	@ echo CXX $<
 	$(Q)$(CXX) $< $(CXXFLAGS) -c -o $@ 2> $@.$(WARNS_EXT) \
 		|| (cat $@.$(WARNS_EXT); exit 1)
@@ -641,7 +651,7 @@ $(PROTO_BUILD_DIR)/%.pb.cc $(PROTO_BUILD_DIR)/%.pb.h : \
 $(PY_PROTO_BUILD_DIR)/%_pb2.py : $(PROTO_SRC_DIR)/%.proto \
 		$(PY_PROTO_INIT) | $(PY_PROTO_BUILD_DIR)
 	@ echo PROTOC \(python\) $<
-	$(Q)protoc --proto_path=$(PROTO_SRC_DIR) --python_out=$(PY_PROTO_BUILD_DIR) $<
+	$(Q)protoc --proto_path=src --python_out=python $<
 
 $(PY_PROTO_INIT): | $(PY_PROTO_BUILD_DIR)
 	touch $(PY_PROTO_INIT)
@@ -694,6 +704,6 @@ $(DISTRIBUTE_DIR): all py | $(DISTRIBUTE_SUBDIRS)
 	install -m 644 $(DYNAMIC_NAME) $(DISTRIBUTE_DIR)/lib
 	cd $(DISTRIBUTE_DIR)/lib; rm -f $(DYNAMIC_NAME_SHORT);   ln -s $(DYNAMIC_VERSIONED_NAME_SHORT) $(DYNAMIC_NAME_SHORT)
 	# add python - it's not the standard way, indeed...
-	cp -r python $(DISTRIBUTE_DIR)/python
+	cp -r python $(DISTRIBUTE_DIR)/
 
 -include $(DEPS)

Makefile.config.example

@@ -11,6 +11,8 @@
 # USE_OPENCV := 0
 # USE_LEVELDB := 0
 # USE_LMDB := 0
+# This code is taken from https://github.com/sh1r0/caffe-android-lib
+# USE_HDF5 := 0
 
 # uncomment to allow MDB_NOLOCK when reading LMDB files (only if necessary)
 #	You should not set this flag if you will be reading LMDBs with any
@@ -33,6 +35,7 @@ CUDA_DIR := /usr/local/cuda
 # CUDA architecture setting: going with all of them.
 # For CUDA < 6.0, comment the *_50 through *_61 lines for compatibility.
 # For CUDA < 8.0, comment the *_60 and *_61 lines for compatibility.
+# For CUDA >= 9.0, comment the *_20 and *_21 lines for compatibility.
 CUDA_ARCH := -gencode arch=compute_20,code=sm_20 \
 		-gencode arch=compute_20,code=sm_21 \
 		-gencode arch=compute_30,code=sm_30 \

cmake/ConfigGen.cmake

@@ -24,6 +24,18 @@ function(caffe_generate_export_configs)
     set(HAVE_CUDA FALSE)
   endif()
 
+  set(HDF5_IMPORTED OFF)
+  foreach(_lib ${HDF5_LIBRARIES} ${HDF5_HL_LIBRARIES})
+    if(TARGET ${_lib})
+      set(HDF5_IMPORTED ON)
+    endif()
+  endforeach()
+
+  # This code is taken from https://github.com/sh1r0/caffe-android-lib
+  if(USE_HDF5)
+    list(APPEND Caffe_DEFINITIONS -DUSE_HDF5)
+  endif()
+
   if(NOT HAVE_CUDNN)
     set(HAVE_CUDNN FALSE)
   endif()

cmake/Cuda.cmake

@@ -109,6 +109,12 @@ function(caffe_select_nvcc_arch_flags out_variable)
   set(__nvcc_flags "")
   set(__nvcc_archs_readable "")
 
+  string(COMPARE LESS "${CUDA_VERSION}" "9.0" iscudaolderthan90)
+  if(NOT iscudaolderthan90)
+    string(REPLACE "21(20)" "" __cuda_arch_bin "${__cuda_arch_bin}")
+    string(REPLACE "20" "" __cuda_arch_bin "${__cuda_arch_bin}")
+  endif()
+
   # Tell NVCC to add binaries for the specified GPUs
   foreach(__arch ${__cuda_arch_bin})
     if(__arch MATCHES "([0-9]+)\\(([0-9]+)\\)")
@@ -232,7 +238,7 @@ endfunction()
 ################################################################################################
 
 find_package(CUDA 5.5 QUIET)
-find_cuda_helper_libs(curand)  # cmake 2.8.7 compartibility which doesn't search for curand
+find_cuda_helper_libs(curand)  # cmake 2.8.7 compatibility which doesn't search for curand
 
 if(NOT CUDA_FOUND)
   return()

cmake/Dependencies.cmake

@@ -47,6 +47,14 @@ find_package(HDF5 COMPONENTS HL REQUIRED)
 list(APPEND Caffe_INCLUDE_DIRS PUBLIC ${HDF5_INCLUDE_DIRS})
 list(APPEND Caffe_LINKER_LIBS PUBLIC ${HDF5_LIBRARIES} ${HDF5_HL_LIBRARIES})
 
+# This code is taken from https://github.com/sh1r0/caffe-android-lib
+if(USE_HDF5)
+  find_package(HDF5 COMPONENTS HL REQUIRED)
+  include_directories(SYSTEM ${HDF5_INCLUDE_DIRS} ${HDF5_HL_INCLUDE_DIR})
+  list(APPEND Caffe_LINKER_LIBS ${HDF5_LIBRARIES} ${HDF5_HL_LIBRARIES})
+  add_definitions(-DUSE_HDF5)
+endif()
+
 # ---[ LMDB
 if(USE_LMDB)
   find_package(LMDB REQUIRED)

cmake/Misc.cmake

@@ -32,9 +32,10 @@ endif()
 set(CMAKE_INSTALL_RPATH_USE_LINK_PATH TRUE CACHE BOOLEAN "Use link paths for shared library rpath")
 set(CMAKE_MACOSX_RPATH TRUE)
 
-list(FIND CMAKE_PLATFORM_IMPLICIT_LINK_DIRECTORIES ${CMAKE_INSTALL_PREFIX}/lib __is_systtem_dir)
+list(FIND CMAKE_PLATFORM_IMPLICIT_LINK_DIRECTORIES
+     ${CMAKE_INSTALL_PREFIX}/${CMAKE_INSTALL_LIBDIR} __is_systtem_dir)
 if(${__is_systtem_dir} STREQUAL -1)
-  set(CMAKE_INSTALL_RPATH ${CMAKE_INSTALL_PREFIX}/lib)
+  set(CMAKE_INSTALL_RPATH ${CMAKE_INSTALL_PREFIX}/${CMAKE_INSTALL_LIBDIR})
 endif()
 
 # ---[ Funny target

cmake/Modules/FindMKL.cmake

@@ -9,7 +9,7 @@
 # This module defines the following variables:
 #
 #   MKL_FOUND            : True mkl is found
-#   MKL_INCLUDE_DIR      : unclude directory
+#   MKL_INCLUDE_DIR      : include directory
 #   MKL_LIBRARIES        : the libraries to link against.
 
 

cmake/ProtoBuf.cmake

@@ -78,7 +78,7 @@ function(caffe_protobuf_generate_cpp_py output_dir srcs_var hdrs_var python_var)
              "${output_dir}/${fil_we}_pb2.py"
       COMMAND ${CMAKE_COMMAND} -E make_directory "${output_dir}"
       COMMAND ${PROTOBUF_PROTOC_EXECUTABLE} --cpp_out    ${output_dir} ${_protoc_include} ${abs_fil}
-      COMMAND ${PROTOBUF_PROTOC_EXECUTABLE} --python_out ${output_dir} ${_protoc_include} ${abs_fil}
+      COMMAND ${PROTOBUF_PROTOC_EXECUTABLE} --python_out ${PROJECT_BINARY_DIR}/include --proto_path ${PROJECT_SOURCE_DIR}/src ${_protoc_include} ${abs_fil}
       DEPENDS ${abs_fil}
       COMMENT "Running C++/Python protocol buffer compiler on ${fil}" VERBATIM )
   endforeach()

cmake/Summary.cmake

@@ -119,6 +119,8 @@ function(caffe_print_configuration_summary)
   caffe_status("  USE_LMDB          :   ${USE_LMDB}")
   caffe_status("  USE_NCCL          :   ${USE_NCCL}")
   caffe_status("  ALLOW_LMDB_NOLOCK :   ${ALLOW_LMDB_NOLOCK}")
+  # This code is taken from https://github.com/sh1r0/caffe-android-lib
+  caffe_status("  USE_HDF5          :   ${USE_HDF5}")
   caffe_status("")
   caffe_status("Dependencies:")
   caffe_status("  BLAS              : " APPLE THEN "Yes (vecLib)" ELSE "Yes (${BLAS})")

docs/development.md

@@ -116,5 +116,5 @@ To get a list of all options `googletest` provides, simply pass the `--help` fla
 
 - **Run `make lint` to check C++ code.**
 - Wrap lines at 80 chars.
-- Follow [Google C++ style](http://google-styleguide.googlecode.com/svn/trunk/cppguide.xml) and [Google python style](http://google-styleguide.googlecode.com/svn/trunk/pyguide.html) + [PEP 8](http://legacy.python.org/dev/peps/pep-0008/).
+- Follow [Google C++ style](https://google.github.io/styleguide/cppguide.html) and [Google python style](https://google.github.io/styleguide/pyguide.html) + [PEP 8](http://legacy.python.org/dev/peps/pep-0008/).
 - Remember that “a foolish consistency is the hobgoblin of little minds,” so use your best judgement to write the clearest code for your particular case.

docs/install_apt.md

@@ -40,6 +40,7 @@ Continue with [compilation](installation.html#compilation).
 
     sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler
     sudo apt-get install --no-install-recommends libboost-all-dev
+    sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev
 
 **CUDA**: Install by `apt-get` or the NVIDIA `.run` package.
 The NVIDIA package tends to follow more recent library and driver versions, but the installation is more manual.
@@ -54,12 +55,6 @@ This can be skipped for CPU-only installation.
 
 CUDA 8 is required on Ubuntu 16.04.
 
-**Remaining dependencies, 14.04**
-
-Everything is packaged in 14.04.
-
-    sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev
-
 **Remaining dependencies, 12.04**
 
 These dependencies need manual installation in 12.04.

docs/installation.md

@@ -17,7 +17,7 @@ The official Makefile and `Makefile.config` build are complemented by a [communi
 - [RHEL / CentOS / Fedora installation](install_yum.html)
 - [Windows](https://github.com/BVLC/caffe/tree/windows) *see the Windows branch led by Guillaume Dumont*
 - [OpenCL](https://github.com/BVLC/caffe/tree/opencl) *see the OpenCL branch led by Fabian Tschopp*
-- [AWS AMI](https://github.com/bitfusionio/amis/tree/master/awsmrkt-bfboost-ubuntu14-cuda75-caffe) *pre-configured for AWS*
+- [AWS AMI](https://aws.amazon.com/marketplace/pp/B01M0AXXQB) *official deep learning amazon machine image from AWS*
 
 **Overview**:
 
@@ -42,14 +42,14 @@ Optional dependencies:
 
 * [OpenCV](http://opencv.org/) >= 2.4 including 3.0
 * IO libraries: `lmdb`, `leveldb` (note: leveldb requires `snappy`)
-* cuDNN for GPU acceleration (v6)
+* cuDNN for GPU acceleration (v7)
 
 Pycaffe and Matcaffe interfaces have their own natural needs.
 
 * For Python Caffe:  `Python 2.7` or `Python 3.3+`, `numpy (>= 1.7)`, boost-provided `boost.python`
 * For MATLAB Caffe: MATLAB with the `mex` compiler.
 
-**cuDNN Caffe**: for fastest operation Caffe is accelerated by drop-in integration of [NVIDIA cuDNN](https://developer.nvidia.com/cudnn). To speed up your Caffe models, install cuDNN then uncomment the `USE_CUDNN := 1` flag in `Makefile.config` when installing Caffe. Acceleration is automatic. The current version is cuDNN v6; older versions are supported in older Caffe.
+**cuDNN Caffe**: for fastest operation Caffe is accelerated by drop-in integration of [NVIDIA cuDNN](https://developer.nvidia.com/cudnn). To speed up your Caffe models, install cuDNN then uncomment the `USE_CUDNN := 1` flag in `Makefile.config` when installing Caffe. Acceleration is automatic. The current version is cuDNN v7; older versions are supported in older Caffe.
 
 **CPU-only Caffe**: for cold-brewed CPU-only Caffe uncomment the `CPU_ONLY := 1` flag in `Makefile.config` to configure and build Caffe without CUDA. This is helpful for cloud or cluster deployment.
 
@@ -80,7 +80,7 @@ The main requirements are `numpy` and `boost.python` (provided by boost). `panda
 
 You can install the dependencies with
 
-    for req in $(cat requirements.txt); do pip install $req; done
+    pip install -r requirements.txt
 
 but we suggest first installing the [Anaconda](https://store.continuum.io/cshop/anaconda/) Python distribution, which provides most of the necessary packages, as well as the `hdf5` library dependency.
 

docs/tutorial/interfaces.md

@@ -129,8 +129,8 @@ Use CPU:
 
 Use GPU and specify its gpu_id:
 
-    caffe.set_mode_gpu();
     caffe.set_device(gpu_id);
+    caffe.set_mode_gpu();
 
 #### Create a network and access its layers and blobs
 

docs/tutorial/layers.md

@@ -87,12 +87,13 @@ Layers:
 * [ELU](layers/elu.html) - exponential linear rectification.
 * [Sigmoid](layers/sigmoid.html)
 * [TanH](layers/tanh.html)
-* [Absolute Value](layers/abs.html)
+* [Absolute Value](layers/absval.html)
 * [Power](layers/power.html) - f(x) = (shift + scale * x) ^ power.
 * [Exp](layers/exp.html) - f(x) = base ^ (shift + scale * x).
 * [Log](layers/log.html) - f(x) = log(x).
 * [BNLL](layers/bnll.html) - f(x) = log(1 + exp(x)).
 * [Threshold](layers/threshold.html) - performs step function at user defined threshold.
+* [Clip](layers/clip.html) - clips a blob between a fixed minimum and maximum value.
 * [Bias](layers/bias.html) - adds a bias to a blob that can either be learned or fixed.
 * [Scale](layers/scale.html) - scales a blob by an amount that can either be learned or fixed.