2019-06-06 14:21:15 +02:00
|
|
|
#!/usr/bin/env bash
|
|
|
|
#
|
2021-02-01 06:31:20 +01:00
|
|
|
# /* ******************************************************************************
|
|
|
|
# *
|
|
|
|
# *
|
|
|
|
# * This program and the accompanying materials are made available under the
|
|
|
|
# * terms of the Apache License, Version 2.0 which is available at
|
|
|
|
# * https://www.apache.org/licenses/LICENSE-2.0.
|
|
|
|
# *
|
2021-02-01 09:47:29 +01:00
|
|
|
# * See the NOTICE file distributed with this work for additional
|
|
|
|
# * information regarding copyright ownership.
|
2021-02-01 06:31:20 +01:00
|
|
|
# * Unless required by applicable law or agreed to in writing, software
|
|
|
|
# * distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
|
|
|
|
# * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
|
|
|
|
# * License for the specific language governing permissions and limitations
|
|
|
|
# * under the License.
|
|
|
|
# *
|
|
|
|
# * SPDX-License-Identifier: Apache-2.0
|
|
|
|
# ******************************************************************************/
|
2019-06-06 14:21:15 +02:00
|
|
|
#
|
|
|
|
|
|
|
|
set -eu
|
|
|
|
|
|
|
|
# cd to the directory containing this script
|
|
|
|
DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
|
|
|
|
cd "$DIR"
|
|
|
|
|
2020-03-19 06:53:21 +01:00
|
|
|
setwindows_msys() {
|
|
|
|
if [[ $KERNEL == *"windows"* ]]; then
|
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -G \"MSYS Makefiles\""
|
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
|
|
|
setandroid_defaults() {
|
|
|
|
if [[ -z ${ANDROID_NDK:-} ]]; then
|
|
|
|
export ANDROID_NDK=$HOME/Android/android-ndk/
|
|
|
|
echo "No ANDROID_NDK variable set. Setting to default of $ANDROID_NDK"
|
|
|
|
else
|
|
|
|
echo "USING ANDROID NDK $ANDROID_NDK"
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [[ -z ${ANDROID_VERSION:-} ]]; then
|
|
|
|
export ANDROID_VERSION=21
|
|
|
|
echo "No ANDROID_VERSION variable set. Setting to default of $ANDROID_VERSION"
|
|
|
|
else
|
|
|
|
echo "USING ANDROID VERSION $ANDROID_VERSION"
|
|
|
|
# android needs static linking
|
|
|
|
|
|
|
|
fi
|
|
|
|
|
|
|
|
|
|
|
|
}
|
|
|
|
|
2019-06-06 14:21:15 +02:00
|
|
|
export CMAKE_COMMAND="cmake"
|
|
|
|
if which cmake3 &> /dev/null; then
|
|
|
|
export CMAKE_COMMAND="cmake3"
|
|
|
|
fi
|
|
|
|
export MAKE_COMMAND="make"
|
|
|
|
export MAKE_ARGUMENTS=
|
|
|
|
echo eval $CMAKE_COMMAND
|
|
|
|
|
|
|
|
[[ -z ${MAKEJ:-} ]] && MAKEJ=4
|
|
|
|
|
|
|
|
# Use > 1 to consume two arguments per pass in the loop (e.g. each
|
|
|
|
# argument has a corresponding value to go with it).
|
|
|
|
# Use > 0 to consume one or more arguments per pass in the loop (e.g.
|
|
|
|
# some arguments don't have a corresponding value to go with it such
|
|
|
|
# as in the --default example).
|
|
|
|
# note: if this is set to > 0 the /etc/hosts part is not recognized ( may be a bug )
|
|
|
|
PARALLEL="true"
|
|
|
|
OS=
|
|
|
|
CHIP=
|
|
|
|
BUILD=
|
|
|
|
COMPUTE=
|
|
|
|
ARCH=
|
|
|
|
LIBTYPE=
|
|
|
|
PACKAGING=
|
|
|
|
CHIP_EXTENSION=
|
|
|
|
CHIP_VERSION=
|
|
|
|
EXPERIMENTAL=
|
|
|
|
OPERATIONS=
|
|
|
|
CLEAN="false"
|
|
|
|
MINIFIER="false"
|
|
|
|
TESTS="false"
|
|
|
|
VERBOSE="false"
|
2019-11-13 15:15:18 +01:00
|
|
|
VERBOSE_ARG="VERBOSE=1"
|
2019-09-11 20:50:28 +02:00
|
|
|
HELPER=
|
2020-01-28 17:00:12 +01:00
|
|
|
CHECK_VECTORIZATION="OFF"
|
2021-03-05 02:59:02 +01:00
|
|
|
SYS_ROOT=
|
|
|
|
EXTRA_LINK_FLAGS=
|
2019-06-06 14:21:15 +02:00
|
|
|
NAME=
|
2021-03-05 02:59:02 +01:00
|
|
|
EXTRA_CUDA_FLAGS=
|
2020-03-19 06:53:21 +01:00
|
|
|
while [[ $# -gt 0 ]]
|
2019-06-06 14:21:15 +02:00
|
|
|
do
|
|
|
|
key="$1"
|
|
|
|
value="${2:-}"
|
|
|
|
#Build type (release/debug), packaging type, chip: cpu,cuda, lib type (static/dynamic)
|
|
|
|
case $key in
|
2019-09-11 20:50:28 +02:00
|
|
|
-h|--helper)
|
|
|
|
HELPER="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
2019-06-06 14:21:15 +02:00
|
|
|
-o|-platform|--platform)
|
|
|
|
OS="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
|
|
|
-b|--build-type)
|
|
|
|
BUILD="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
|
|
|
-p|--packaging)
|
|
|
|
PACKAGING="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
|
|
|
-c|--chip)
|
|
|
|
CHIP="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
|
|
|
-cc|--compute)
|
|
|
|
COMPUTE="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
|
|
|
-a|--arch)
|
|
|
|
ARCH="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
|
|
|
-l|--libtype)
|
|
|
|
LIBTYPE="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
|
|
|
-e|--chip-extension)
|
|
|
|
CHIP_EXTENSION="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
|
|
|
-v|--chip-version)
|
|
|
|
CHIP_VERSION="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
|
|
|
-x|--experimental)
|
|
|
|
EXPERIMENTAL="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
|
|
|
-g|--generator)
|
|
|
|
OPERATIONS="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
|
|
|
--name)
|
|
|
|
NAME="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
2020-01-28 17:00:12 +01:00
|
|
|
--check-vectorization)
|
|
|
|
CHECK_VECTORIZATION="ON"
|
|
|
|
;;
|
2019-06-06 14:21:15 +02:00
|
|
|
-j)
|
|
|
|
MAKEJ="$value"
|
|
|
|
shift # past argument
|
|
|
|
;;
|
|
|
|
clean)
|
|
|
|
CLEAN="true"
|
|
|
|
;;
|
|
|
|
-m|--minifier)
|
|
|
|
MINIFIER="true"
|
|
|
|
;;
|
|
|
|
-t|--tests)
|
|
|
|
TESTS="true"
|
|
|
|
;;
|
|
|
|
-V|--verbose)
|
|
|
|
VERBOSE="true"
|
|
|
|
;;
|
|
|
|
--default)
|
|
|
|
DEFAULT=YES
|
|
|
|
;;
|
|
|
|
*)
|
|
|
|
# unknown option
|
|
|
|
;;
|
|
|
|
esac
|
2020-03-19 06:53:21 +01:00
|
|
|
if [[ $# -gt 0 ]]; then
|
2019-06-06 14:21:15 +02:00
|
|
|
shift # past argument or value
|
|
|
|
fi
|
|
|
|
done
|
|
|
|
HOST=$(uname -s | tr [A-Z] [a-z])
|
|
|
|
KERNEL=$HOST-$(uname -m | tr [A-Z] [a-z])
|
|
|
|
if [ "$(uname)" == "Darwin" ]; then
|
|
|
|
HOST="macosx"
|
|
|
|
KERNEL="darwin-x86_64"
|
|
|
|
echo "RUNNING OSX CLANG"
|
|
|
|
elif [ "$(expr substr $(uname -s) 1 5)" == "MINGW" ] || [ "$(expr substr $(uname -s) 1 4)" == "MSYS" ]; then
|
|
|
|
HOST="windows"
|
|
|
|
KERNEL="windows-x86_64"
|
|
|
|
echo "Running windows"
|
|
|
|
elif [ "$(uname -m)" == "ppc64le" ]; then
|
|
|
|
if [ -z "$ARCH" ]; then
|
|
|
|
ARCH="power8"
|
|
|
|
fi
|
|
|
|
KERNEL="linux-ppc64le"
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ -z "$OS" ]; then
|
|
|
|
OS="$HOST"
|
|
|
|
fi
|
|
|
|
|
2020-03-20 04:14:03 +01:00
|
|
|
if [[ -z ${ANDROID_NDK:-} ]]; then
|
|
|
|
export ANDROID_NDK=$HOME/Android/android-ndk/
|
|
|
|
fi
|
2019-06-06 14:21:15 +02:00
|
|
|
|
|
|
|
case "$OS" in
|
|
|
|
linux-armhf)
|
2019-12-02 19:37:21 +01:00
|
|
|
if [ -z "$ARCH" ]; then
|
2021-02-01 06:31:20 +01:00
|
|
|
ARCH="armv7-a"
|
|
|
|
fi
|
|
|
|
if [ ! -z ${RPI_BIN+set} ]; then
|
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -D CMAKE_TOOLCHAIN_FILE=cmake/rpi.cmake"
|
2019-12-02 19:37:21 +01:00
|
|
|
fi
|
2021-02-01 06:31:20 +01:00
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -DSD_ARM_BUILD=true -DSD_SANITIZE=OFF "
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
|
|
|
|
linux-arm64)
|
2019-12-02 19:37:21 +01:00
|
|
|
if [ -z "$ARCH" ]; then
|
2019-06-06 14:21:15 +02:00
|
|
|
ARCH="armv8-a"
|
2019-12-02 19:37:21 +01:00
|
|
|
fi
|
2021-02-01 06:31:20 +01:00
|
|
|
if [ ! -z ${RPI_BIN+set} ]; then
|
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -D CMAKE_TOOLCHAIN_FILE=cmake/rpi.cmake"
|
|
|
|
fi
|
2020-03-02 10:49:41 +01:00
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -DSD_ARM_BUILD=true"
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
|
|
|
|
android-arm)
|
2019-12-02 19:37:21 +01:00
|
|
|
if [ -z "$ARCH" ]; then
|
2019-06-06 14:21:15 +02:00
|
|
|
ARCH="armv7-a"
|
2019-12-02 19:37:21 +01:00
|
|
|
fi
|
2020-03-19 06:53:21 +01:00
|
|
|
|
|
|
|
setandroid_defaults
|
|
|
|
|
2020-03-24 02:55:47 +01:00
|
|
|
# Note here for android 32 bit prefix on the binutils is different
|
|
|
|
# See https://developer.android.com/ndk/guides/other_build_systems
|
|
|
|
export ANDROID_BIN="$ANDROID_NDK/toolchains/arm-linux-androideabi/prebuilt/$KERNEL/"
|
2019-12-02 19:37:21 +01:00
|
|
|
export ANDROID_CPP="$ANDROID_NDK/sources/cxx-stl/llvm-libc++/"
|
2019-12-05 11:46:01 +01:00
|
|
|
export ANDROID_CC="$ANDROID_NDK/toolchains/llvm/prebuilt/$KERNEL/bin/clang"
|
2020-03-19 06:53:21 +01:00
|
|
|
export ANDROID_ROOT="$ANDROID_NDK/platforms/android-$ANDROID_VERSION/arch-arm/"
|
2020-03-02 10:49:41 +01:00
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -DCMAKE_TOOLCHAIN_FILE=cmake/android-arm.cmake -DSD_ANDROID_BUILD=true"
|
2020-03-19 06:53:21 +01:00
|
|
|
setwindows_msys
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
|
|
|
|
android-arm64)
|
2019-12-02 19:37:21 +01:00
|
|
|
if [ -z "$ARCH" ]; then
|
2019-06-06 14:21:15 +02:00
|
|
|
ARCH="armv8-a"
|
2019-12-02 19:37:21 +01:00
|
|
|
fi
|
2020-03-19 06:53:21 +01:00
|
|
|
|
|
|
|
setandroid_defaults
|
|
|
|
|
|
|
|
echo "BUILDING ANDROID ARM with KERNEL $KERNEL"
|
2019-12-02 19:37:21 +01:00
|
|
|
export ANDROID_BIN="$ANDROID_NDK/toolchains/aarch64-linux-android-4.9/prebuilt/$KERNEL/"
|
|
|
|
export ANDROID_CPP="$ANDROID_NDK/sources/cxx-stl/llvm-libc++/"
|
2019-12-05 11:46:01 +01:00
|
|
|
export ANDROID_CC="$ANDROID_NDK/toolchains/llvm/prebuilt/$KERNEL/bin/clang"
|
2020-03-19 06:53:21 +01:00
|
|
|
export ANDROID_ROOT="$ANDROID_NDK/platforms/android-$ANDROID_VERSION/arch-arm64/"
|
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -DCMAKE_TOOLCHAIN_FILE=cmake/android-arm64.cmake -DSD_ANDROID_BUILD=true"
|
|
|
|
setwindows_msys
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
|
|
|
|
android-x86)
|
2019-12-02 19:37:21 +01:00
|
|
|
if [ -z "$ARCH" ]; then
|
2019-06-06 14:21:15 +02:00
|
|
|
ARCH="i686"
|
2019-12-02 19:37:21 +01:00
|
|
|
fi
|
2020-03-19 06:53:21 +01:00
|
|
|
|
|
|
|
setandroid_defaults
|
|
|
|
export ANDROID_BIN="$ANDROID_NDK/toolchains/arm-linux-androideabi-4.9/prebuilt/$KERNEL/"
|
2019-12-02 19:37:21 +01:00
|
|
|
export ANDROID_CPP="$ANDROID_NDK/sources/cxx-stl/llvm-libc++/"
|
2019-12-05 11:46:01 +01:00
|
|
|
export ANDROID_CC="$ANDROID_NDK/toolchains/llvm/prebuilt/$KERNEL/bin/clang"
|
2020-03-19 06:53:21 +01:00
|
|
|
export ANDROID_ROOT="$ANDROID_NDK/platforms/android-$ANDROID_VERSION/arch-x86/"
|
2020-03-02 10:49:41 +01:00
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -DCMAKE_TOOLCHAIN_FILE=cmake/android-x86.cmake -DSD_ANDROID_BUILD=true"
|
2020-03-19 06:53:21 +01:00
|
|
|
setwindows_msys
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
|
|
|
|
android-x86_64)
|
2020-03-19 06:53:21 +01:00
|
|
|
|
2019-12-02 19:37:21 +01:00
|
|
|
if [ -z "$ARCH" ]; then
|
2019-06-06 14:21:15 +02:00
|
|
|
ARCH="x86-64"
|
2019-12-02 19:37:21 +01:00
|
|
|
fi
|
2020-03-19 06:53:21 +01:00
|
|
|
echo "BUILDING ANDROID x86_64"
|
|
|
|
|
|
|
|
setandroid_defaults
|
|
|
|
|
|
|
|
|
|
|
|
export ANDROID_BIN="$ANDROID_NDK/toolchains/arm-linux-androideabi-4.9/prebuilt/$KERNEL/"
|
2019-12-02 19:37:21 +01:00
|
|
|
export ANDROID_CPP="$ANDROID_NDK/sources/cxx-stl/llvm-libc++/"
|
2019-12-05 11:46:01 +01:00
|
|
|
export ANDROID_CC="$ANDROID_NDK/toolchains/llvm/prebuilt/$KERNEL/bin/clang"
|
2020-03-19 06:53:21 +01:00
|
|
|
export ANDROID_ROOT="$ANDROID_NDK/platforms/android-$ANDROID_VERSION/arch-x86_64/"
|
2020-03-02 10:49:41 +01:00
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -DCMAKE_TOOLCHAIN_FILE=cmake/android-x86_64.cmake -DSD_ANDROID_BUILD=true"
|
2020-03-19 06:53:21 +01:00
|
|
|
setwindows_msys
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
|
|
|
|
ios-x86_64)
|
2019-12-02 19:37:21 +01:00
|
|
|
LIBTYPE="static"
|
|
|
|
ARCH="x86-64"
|
|
|
|
if xcrun --sdk iphoneos --show-sdk-version &> /dev/null; then
|
|
|
|
export IOS_VERSION="$(xcrun --sdk iphoneos --show-sdk-version)"
|
|
|
|
else
|
2019-06-06 14:21:15 +02:00
|
|
|
export IOS_VERSION="10.3"
|
2019-12-02 19:37:21 +01:00
|
|
|
fi
|
|
|
|
XCODE_PATH="$(xcode-select --print-path)"
|
|
|
|
export IOS_SDK="$XCODE_PATH/Platforms/iPhoneSimulator.platform/Developer/SDKs/iPhoneSimulator$IOS_VERSION.sdk"
|
2020-03-02 10:49:41 +01:00
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -DCMAKE_TOOLCHAIN_FILE=cmake/ios-x86_64.cmake --debug-trycompile -DSD_IOS_BUILD=true"
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
|
|
|
|
ios-x86)
|
2019-12-02 19:37:21 +01:00
|
|
|
LIBTYPE="static"
|
|
|
|
ARCH="i386"
|
|
|
|
if xcrun --sdk iphoneos --show-sdk-version &> /dev/null; then
|
|
|
|
export IOS_VERSION="$(xcrun --sdk iphoneos --show-sdk-version)"
|
|
|
|
else
|
2019-06-06 14:21:15 +02:00
|
|
|
export IOS_VERSION="10.3"
|
2019-12-02 19:37:21 +01:00
|
|
|
fi
|
|
|
|
XCODE_PATH="$(xcode-select --print-path)"
|
|
|
|
export IOS_SDK="$XCODE_PATH/Platforms/iPhoneSimulator.platform/Developer/SDKs/iPhoneSimulator$IOS_VERSION.sdk"
|
2020-03-02 10:49:41 +01:00
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -DCMAKE_TOOLCHAIN_FILE=cmake/ios-x86.cmake --debug-trycompile -DSD_IOS_BUILD=true"
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
|
|
|
|
ios-arm64)
|
2019-12-02 19:37:21 +01:00
|
|
|
LIBTYPE="static"
|
|
|
|
ARCH="arm64"
|
|
|
|
if xcrun --sdk iphoneos --show-sdk-version &> /dev/null; then
|
|
|
|
export IOS_VERSION="$(xcrun --sdk iphoneos --show-sdk-version)"
|
|
|
|
else
|
2019-06-06 14:21:15 +02:00
|
|
|
export IOS_VERSION="10.3"
|
2019-12-02 19:37:21 +01:00
|
|
|
fi
|
|
|
|
XCODE_PATH="$(xcode-select --print-path)"
|
|
|
|
export IOS_SDK="$XCODE_PATH/Platforms/iPhoneOS.platform/Developer/SDKs/iPhoneOS$IOS_VERSION.sdk"
|
2020-03-02 10:49:41 +01:00
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -DCMAKE_TOOLCHAIN_FILE=cmake/ios-arm64.cmake --debug-trycompile -DSD_IOS_BUILD=true"
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
|
|
|
|
ios-arm)
|
2019-12-02 19:37:21 +01:00
|
|
|
LIBTYPE="static"
|
|
|
|
ARCH="armv7"
|
|
|
|
if xcrun --sdk iphoneos --show-sdk-version &> /dev/null; then
|
|
|
|
export IOS_VERSION="$(xcrun --sdk iphoneos --show-sdk-version)"
|
|
|
|
else
|
2019-06-06 14:21:15 +02:00
|
|
|
export IOS_VERSION="10.3"
|
2019-12-02 19:37:21 +01:00
|
|
|
fi
|
|
|
|
XCODE_PATH="$(xcode-select --print-path)"
|
|
|
|
export IOS_SDK="$XCODE_PATH/Platforms/iPhoneOS.platform/Developer/SDKs/iPhoneOS$IOS_VERSION.sdk"
|
2020-03-02 10:49:41 +01:00
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -DCMAKE_TOOLCHAIN_FILE=cmake/ios-arm.cmake --debug-trycompile -DSD_IOS_BUILD=true"
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
|
|
|
|
ios-armv7)
|
2019-12-02 19:37:21 +01:00
|
|
|
# change those 2 parameters and make sure the IOS_SDK exists
|
|
|
|
export iPhoneOS="iPhoneOS"
|
|
|
|
export IOS_VERSION="10.3"
|
|
|
|
LIBTYPE="static"
|
|
|
|
ARCH="armv7"
|
|
|
|
export IOS_SDK="/Applications/Xcode.app/Contents/Developer/Platforms/${iPhoneOS}.platform/Developer/SDKs/${iPhoneOS}${IOS_VERSION}.sdk"
|
2020-03-02 10:49:41 +01:00
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -DCMAKE_TOOLCHAIN_FILE=cmake/ios-armv7.cmake --debug-trycompile -DSD_IOS_BUILD=true"
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
|
|
|
|
linux*)
|
|
|
|
;;
|
|
|
|
|
|
|
|
macosx*)
|
2019-12-02 19:37:21 +01:00
|
|
|
export CC=clang
|
|
|
|
export CXX=clang++
|
|
|
|
PARALLEL="true"
|
2020-03-02 10:49:41 +01:00
|
|
|
export CMAKE_COMMAND="$CMAKE_COMMAND -DCMAKE_MACOSX_RPATH=ON -DSD_APPLE_BUILD=true"
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
|
|
|
|
windows*)
|
2019-11-13 15:15:18 +01:00
|
|
|
# Do something under Windows NT platform
|
|
|
|
if [ "$CHIP" == "cuda" ]; then
|
2019-06-06 14:21:15 +02:00
|
|
|
export CMAKE_COMMAND="cmake -G \"Ninja\""
|
|
|
|
export MAKE_COMMAND="ninja"
|
|
|
|
export CC="cl.exe"
|
|
|
|
export CXX="cl.exe"
|
|
|
|
PARALLEL="true"
|
2019-11-13 15:15:18 +01:00
|
|
|
VERBOSE_ARG="-v"
|
|
|
|
else
|
2019-06-06 14:21:15 +02:00
|
|
|
export CMAKE_COMMAND="cmake -G \"MSYS Makefiles\""
|
|
|
|
export MAKE_COMMAND="make"
|
|
|
|
export CC=/mingw64/bin/gcc
|
|
|
|
export CXX=/mingw64/bin/g++
|
|
|
|
PARALLEL="true"
|
2019-11-13 15:15:18 +01:00
|
|
|
fi
|
2019-06-06 14:21:15 +02:00
|
|
|
|
2019-11-13 15:15:18 +01:00
|
|
|
# Try some defaults for Visual Studio 2013 if user has not run vcvarsall.bat or something
|
|
|
|
if [ -z "${VCINSTALLDIR:-}" ]; then
|
2021-02-18 03:46:53 +01:00
|
|
|
echo "NEED TO SET DEFAULTS FOR VISUAL STUDIO, NO VCINSTALLDIR environment variable found"
|
|
|
|
export VisualStudioVersion=12.0
|
2019-06-06 14:21:15 +02:00
|
|
|
export VSINSTALLDIR="C:\\Program Files (x86)\\Microsoft Visual Studio $VisualStudioVersion"
|
|
|
|
export VCINSTALLDIR="$VSINSTALLDIR\\VC"
|
|
|
|
export WindowsSdkDir="C:\\Program Files (x86)\\Windows Kits\\8.1"
|
|
|
|
export Platform=X64
|
|
|
|
export INCLUDE="$VCINSTALLDIR\\INCLUDE;$WindowsSdkDir\\include\\shared;$WindowsSdkDir\\include\\um"
|
|
|
|
export LIB="$VCINSTALLDIR\\LIB\\amd64;$WindowsSdkDir\\lib\\winv6.3\\um\\x64"
|
|
|
|
export LIBPATH="$VCINSTALLDIR\\LIB\\amd64;$WindowsSdkDir\\References\\CommonConfiguration\\Neutral"
|
|
|
|
export PATH="$PATH:$VCINSTALLDIR\\BIN\\amd64:$WindowsSdkDir\\bin\\x64:$WindowsSdkDir\\bin\\x86"
|
2019-11-13 15:15:18 +01:00
|
|
|
fi
|
|
|
|
# Make sure we are using 64-bit MinGW-w64
|
|
|
|
export PATH=/mingw64/bin/:/mingw64/lib:$PATH
|
|
|
|
# export GENERATOR="MSYS Makefiles"
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
esac
|
|
|
|
|
|
|
|
if [ -z "$BUILD" ]; then
|
|
|
|
BUILD="release"
|
|
|
|
|
|
|
|
fi
|
|
|
|
|
2021-03-05 02:59:02 +01:00
|
|
|
if [ -z "$SYS_ROOT" ]; then
|
|
|
|
export SYS_ROOT=""
|
|
|
|
fi
|
|
|
|
|
|
|
|
|
2019-06-06 14:21:15 +02:00
|
|
|
if [ -z "$CHIP" ]; then
|
|
|
|
CHIP="cpu"
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ -z "$LIBTYPE" ]; then
|
|
|
|
LIBTYPE="dynamic"
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ -z "$PACKAGING" ]; then
|
|
|
|
PACKAGING="none"
|
|
|
|
fi
|
|
|
|
|
2021-03-05 02:59:02 +01:00
|
|
|
|
2019-06-06 14:21:15 +02:00
|
|
|
|
|
|
|
if [ "$CHIP_EXTENSION" == "avx512" ] || [ "$ARCH" == "avx512" ]; then
|
|
|
|
CHIP_EXTENSION="avx512"
|
|
|
|
ARCH="skylake-avx512"
|
|
|
|
elif [ "$CHIP_EXTENSION" == "avx2" ] || [ "$ARCH" == "avx2" ]; then
|
|
|
|
CHIP_EXTENSION="avx2"
|
|
|
|
ARCH="x86-64"
|
|
|
|
elif [ "$CHIP_EXTENSION" == "x86_64" ] || [ "$ARCH" == "x86_64" ]; then
|
|
|
|
CHIP_EXTENSION="x86_64"
|
|
|
|
ARCH="x86-64"
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ -z "$ARCH" ]; then
|
|
|
|
ARCH="x86-64"
|
|
|
|
fi
|
|
|
|
|
2021-03-05 02:59:02 +01:00
|
|
|
if [ -z "$COMPUTE" ]; then
|
|
|
|
if [ "$ARCH" == "x86-64" ]; then
|
|
|
|
COMPUTE="5.0 5.2 5.3 6.0 6.2 8.0"
|
|
|
|
else
|
|
|
|
COMPUTE="5.0 5.2 5.3 6.0 6.2"
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
|
2019-06-06 14:21:15 +02:00
|
|
|
OPERATIONS_ARG=
|
|
|
|
|
|
|
|
if [ -z "$OPERATIONS" ]; then
|
2020-03-02 10:49:41 +01:00
|
|
|
OPERATIONS_ARG="-DSD_ALL_OPS=true"
|
2019-06-06 14:21:15 +02:00
|
|
|
else
|
|
|
|
OPERATIONS_ARG=$OPERATIONS
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ -z "$EXPERIMENTAL" ]; then
|
|
|
|
EXPERIMENTAL="no"
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ "$CHIP" == "cpu" ]; then
|
2020-03-02 10:49:41 +01:00
|
|
|
BLAS_ARG="-DSD_CPU=true -DBLAS=TRUE"
|
2019-06-06 14:21:15 +02:00
|
|
|
else
|
2020-03-02 10:49:41 +01:00
|
|
|
BLAS_ARG="-DSD_CUDA=true -DBLAS=TRUE"
|
2019-06-06 14:21:15 +02:00
|
|
|
fi
|
|
|
|
|
|
|
|
if [ -z "$NAME" ]; then
|
|
|
|
if [ "$CHIP" == "cpu" ]; then
|
|
|
|
NAME="nd4jcpu"
|
|
|
|
else
|
|
|
|
NAME="nd4jcuda"
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ "$LIBTYPE" == "dynamic" ]; then
|
2020-03-19 06:53:21 +01:00
|
|
|
SHARED_LIBS_ARG="-DSD_SHARED_LIB=ON -DSD_STATIC_LIB=OFF"
|
2019-06-06 14:21:15 +02:00
|
|
|
else
|
2020-03-19 06:53:21 +01:00
|
|
|
SHARED_LIBS_ARG="-DSD_SHARED_LIB=OFF -DSD_STATIC_LIB=ON"
|
2019-06-06 14:21:15 +02:00
|
|
|
fi
|
|
|
|
|
|
|
|
if [ "$BUILD" == "release" ]; then
|
|
|
|
BUILD_TYPE="-DCMAKE_BUILD_TYPE=Release"
|
|
|
|
else
|
|
|
|
BUILD_TYPE="-DCMAKE_BUILD_TYPE=Debug"
|
|
|
|
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ "$PACKAGING" == "none" ]; then
|
|
|
|
PACKAGING_ARG="-DPACKAGING=none"
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ "$PACKAGING" == "rpm" ]; then
|
|
|
|
PACKAGING_ARG="-DPACKAGING=rpm"
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ "$PACKAGING" == "deb" ]; then
|
|
|
|
PACKAGING_ARG="-DPACKAGING=deb"
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ "$PACKAGING" == "msi" ]; then
|
|
|
|
PACKAGING_ARG="-DPACKAGING=msi"
|
|
|
|
fi
|
|
|
|
|
2019-09-11 20:50:28 +02:00
|
|
|
EXPERIMENTAL_ARG="";
|
2020-03-02 10:49:41 +01:00
|
|
|
MINIFIER_ARG="-DSD_BUILD_MINIFIER=false"
|
|
|
|
TESTS_ARG="-DSD_BUILD_TESTS=OFF"
|
|
|
|
NAME_ARG="-DSD_LIBRARY_NAME=$NAME"
|
2019-06-06 14:21:15 +02:00
|
|
|
|
|
|
|
if [ "$EXPERIMENTAL" == "yes" ]; then
|
2020-03-02 10:49:41 +01:00
|
|
|
EXPERIMENTAL_ARG="-DSD_EXPERIMENTAL=yes"
|
2019-06-06 14:21:15 +02:00
|
|
|
fi
|
|
|
|
|
|
|
|
if [ "$MINIFIER" == "true" ]; then
|
2020-03-02 10:49:41 +01:00
|
|
|
MINIFIER_ARG="-DSD_BUILD_MINIFIER=true"
|
2019-06-06 14:21:15 +02:00
|
|
|
fi
|
|
|
|
|
|
|
|
if [ "$TESTS" == "true" ]; then
|
2020-03-02 10:49:41 +01:00
|
|
|
MINIFIER_ARG="-DSD_BUILD_MINIFIER=true"
|
|
|
|
TESTS_ARG="-DSD_BUILD_TESTS=ON"
|
2019-06-06 14:21:15 +02:00
|
|
|
fi
|
|
|
|
|
2021-03-05 02:59:02 +01:00
|
|
|
|
|
|
|
if [ "$SYS_ROOT" != "" ]; then
|
|
|
|
EXTRA_SYSROOT="-DCMAKE_SYSROOT=$SYS_ROOT"
|
|
|
|
else
|
|
|
|
EXTRA_SYSROOT=""
|
|
|
|
fi
|
|
|
|
|
2020-03-02 10:49:41 +01:00
|
|
|
ARCH_ARG="-DSD_ARCH=$ARCH -DSD_EXTENSION=$CHIP_EXTENSION"
|
2019-06-06 14:21:15 +02:00
|
|
|
|
Development updates (#9098)
* RL4J: Add generic update rule (#502)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Shyrma reduce (#481)
* - start working on improving of cpu legacy code for reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving legacy loops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - still working on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - testing speed run of new reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - working on improvement of default loop for reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - update signatures of stuff which calls reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - make corrections in cuda reduce kernels
Signed-off-by: Yurii <iuriish@yahoo.com>
* - change loop for default case in broadcast legacy ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment some shape stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment unnecessary prints in RNGtests
Signed-off-by: Yurii <iuriish@yahoo.com>
* - finish to resolve conflicts after master has been merged
Signed-off-by: Yurii <iuriish@yahoo.com>
* - get rid of some compilation mistakes of cuda stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor changes
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further search for bug causing crash on java test
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add scalar case in reduce_ ... exec stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor corrections in NAtiveOps.cu
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add switch to scalar case execReduceXD functions
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in ConstantShapeHelper::createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
* - correct cuda mirrorPad
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in cuda createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
* Add support for CUDA 11.0 (#492)
* Add support for CUDA 11.0
* libnd4j tweaks for CUDA 11
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* bindings update, again?
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update versions of JavaCPP Presets for FFmpeg, OpenBLAS, and NumPy
* update API to match CUDA 8
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update version of JavaCPP Presets for CPython
* C++ updated for cuDNN 8.0
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* 128-bit alignment for workspaces
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* change seed in 1 test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Fix dependecy duplication in python4j-parent pom
* Fix group id for in python4j-numpy
* few tests tweaked
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Remove macosx-x86_64-gpu from nd4j-tests-tensorflow
* few minor tweaks for IndexReduce
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one test removed
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* RL4J: Add SyncTrainer and AgentLearnerBuilder for a few algorithms (#504)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* * Update versions of JavaCPP Presets for OpenCV, FFmpeg, and MKL
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Fix L2NormalizeVertex and eclipse#9054 (#513)
* update
* Fix L2NormalizeVertex
Fix eclipse#9054
* RL4J: Add async training and advantage actor-critic (#507)
* Added async training & Advantage Actor Critic
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Fix compiler error
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Renamed ActorCriticPolicy back to ACPolicy
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
* Python GIL overhaul (#517)
* Development updates (#9053)
* RL4J: Add generic update rule (#502)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Shyrma reduce (#481)
* - start working on improving of cpu legacy code for reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving legacy loops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - still working on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - testing speed run of new reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - working on improvement of default loop for reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - update signatures of stuff which calls reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - make corrections in cuda reduce kernels
Signed-off-by: Yurii <iuriish@yahoo.com>
* - change loop for default case in broadcast legacy ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment some shape stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment unnecessary prints in RNGtests
Signed-off-by: Yurii <iuriish@yahoo.com>
* - finish to resolve conflicts after master has been merged
Signed-off-by: Yurii <iuriish@yahoo.com>
* - get rid of some compilation mistakes of cuda stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor changes
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further search for bug causing crash on java test
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add scalar case in reduce_ ... exec stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor corrections in NAtiveOps.cu
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add switch to scalar case execReduceXD functions
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in ConstantShapeHelper::createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
* - correct cuda mirrorPad
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in cuda createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
* Add support for CUDA 11.0 (#492)
* Add support for CUDA 11.0
* libnd4j tweaks for CUDA 11
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* bindings update, again?
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update versions of JavaCPP Presets for FFmpeg, OpenBLAS, and NumPy
* update API to match CUDA 8
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update version of JavaCPP Presets for CPython
* C++ updated for cuDNN 8.0
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* 128-bit alignment for workspaces
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* change seed in 1 test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Fix dependecy duplication in python4j-parent pom
* Fix group id for in python4j-numpy
* few tests tweaked
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Remove macosx-x86_64-gpu from nd4j-tests-tensorflow
* few minor tweaks for IndexReduce
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one test removed
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* RL4J: Add SyncTrainer and AgentLearnerBuilder for a few algorithms (#504)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* Removed dead code (#9057)
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* performance improvement (#9055)
* performance improvement
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* revert some changes
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Development updates (#9064)
* Update versions of JavaCPP Presets for OpenCV, FFmpeg, and MKL
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Cherry pick rl4j changes from most recent KonduitAI/deeplearning4j PR
* Update cherry pick again from last master revision.
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
Co-authored-by: dariuszzbyrad <dariusz.zbyrad@gmail.com>
* Ag pythongiloverhaul (#518)
* Development updates (#9053)
* RL4J: Add generic update rule (#502)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Shyrma reduce (#481)
* - start working on improving of cpu legacy code for reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving legacy loops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - still working on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - testing speed run of new reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - working on improvement of default loop for reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - update signatures of stuff which calls reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - make corrections in cuda reduce kernels
Signed-off-by: Yurii <iuriish@yahoo.com>
* - change loop for default case in broadcast legacy ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment some shape stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment unnecessary prints in RNGtests
Signed-off-by: Yurii <iuriish@yahoo.com>
* - finish to resolve conflicts after master has been merged
Signed-off-by: Yurii <iuriish@yahoo.com>
* - get rid of some compilation mistakes of cuda stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor changes
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further search for bug causing crash on java test
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add scalar case in reduce_ ... exec stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor corrections in NAtiveOps.cu
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add switch to scalar case execReduceXD functions
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in ConstantShapeHelper::createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
* - correct cuda mirrorPad
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in cuda createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
* Add support for CUDA 11.0 (#492)
* Add support for CUDA 11.0
* libnd4j tweaks for CUDA 11
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* bindings update, again?
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update versions of JavaCPP Presets for FFmpeg, OpenBLAS, and NumPy
* update API to match CUDA 8
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update version of JavaCPP Presets for CPython
* C++ updated for cuDNN 8.0
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* 128-bit alignment for workspaces
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* change seed in 1 test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Fix dependecy duplication in python4j-parent pom
* Fix group id for in python4j-numpy
* few tests tweaked
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Remove macosx-x86_64-gpu from nd4j-tests-tensorflow
* few minor tweaks for IndexReduce
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one test removed
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* RL4J: Add SyncTrainer and AgentLearnerBuilder for a few algorithms (#504)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* Removed dead code (#9057)
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* performance improvement (#9055)
* performance improvement
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* revert some changes
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Development updates (#9064)
* Update versions of JavaCPP Presets for OpenCV, FFmpeg, and MKL
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Cherry pick rl4j changes from most recent KonduitAI/deeplearning4j PR
* Update cherry pick again from last master revision.
* Re update python4j
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
Co-authored-by: dariuszzbyrad <dariusz.zbyrad@gmail.com>
* Bump formatter-maven-plugin from 2.0.0 to 2.12.1 (#505)
Bumps [formatter-maven-plugin](https://github.com/revelc/formatter-maven-plugin) from 2.0.0 to 2.12.1.
- [Release notes](https://github.com/revelc/formatter-maven-plugin/releases)
- [Changelog](https://github.com/revelc/formatter-maven-plugin/blob/formatter-maven-plugin-2.12.1/CHANGELOG.md)
- [Commits](https://github.com/revelc/formatter-maven-plugin/compare/formatter-maven-plugin-2.0.0...formatter-maven-plugin-2.12.1)
Signed-off-by: dependabot-preview[bot] <support@dependabot.com>
Co-authored-by: dependabot-preview[bot] <27856297+dependabot-preview[bot]@users.noreply.github.com>
* Ag fix9060 (#519)
* Development updates (#9053)
* RL4J: Add generic update rule (#502)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Shyrma reduce (#481)
* - start working on improving of cpu legacy code for reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving legacy loops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - still working on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - testing speed run of new reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - working on improvement of default loop for reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - update signatures of stuff which calls reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - make corrections in cuda reduce kernels
Signed-off-by: Yurii <iuriish@yahoo.com>
* - change loop for default case in broadcast legacy ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment some shape stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment unnecessary prints in RNGtests
Signed-off-by: Yurii <iuriish@yahoo.com>
* - finish to resolve conflicts after master has been merged
Signed-off-by: Yurii <iuriish@yahoo.com>
* - get rid of some compilation mistakes of cuda stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor changes
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further search for bug causing crash on java test
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add scalar case in reduce_ ... exec stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor corrections in NAtiveOps.cu
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add switch to scalar case execReduceXD functions
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in ConstantShapeHelper::createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
* - correct cuda mirrorPad
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in cuda createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
* Add support for CUDA 11.0 (#492)
* Add support for CUDA 11.0
* libnd4j tweaks for CUDA 11
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* bindings update, again?
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update versions of JavaCPP Presets for FFmpeg, OpenBLAS, and NumPy
* update API to match CUDA 8
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update version of JavaCPP Presets for CPython
* C++ updated for cuDNN 8.0
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* 128-bit alignment for workspaces
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* change seed in 1 test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Fix dependecy duplication in python4j-parent pom
* Fix group id for in python4j-numpy
* few tests tweaked
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Remove macosx-x86_64-gpu from nd4j-tests-tensorflow
* few minor tweaks for IndexReduce
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one test removed
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* RL4J: Add SyncTrainer and AgentLearnerBuilder for a few algorithms (#504)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* Removed dead code (#9057)
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* performance improvement (#9055)
* performance improvement
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* revert some changes
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Development updates (#9064)
* Update versions of JavaCPP Presets for OpenCV, FFmpeg, and MKL
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Added support for the archunit (#9062)
* Added support for the archunit
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Updated pom files
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Datavec code cleaup (#9071)
* removed unnecessary semicolons
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Use standard charset object
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Removed unused imports
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* WIP: Fix Conv1d causal case
* Add inital tests
* Update Conv1d tests to be a bit more robust
* Remove redundant test
* Reset from master
* Remove cuda definition (left over)
* Update rl4j again
* Update pom.xml
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
Co-authored-by: dariuszzbyrad <dariusz.zbyrad@gmail.com>
* Fixes 9061 (#521)
* Get rid of edge case in validation
* Added support for the archunit (#9062)
* Added support for the archunit
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Updated pom files
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Using embedded copying of an array instead of manual (#9073)
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Datavec bulk operation (#9075)
* Bulk operation can be used instead of iteration inspection
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Redundant 'Collection.addAll()' call inspection
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Removed infinitely loop (#9076)
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
Co-authored-by: dariuszzbyrad <dariusz.zbyrad@gmail.com>
* Revert "Merge eclipse changes" (#526)
* Revert rl4j to 72f5c18c830f62df2c04fbf8dc7b1353cc2d3182 (#527)
* RL4J: Add generic update rule (#502)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Shyrma reduce (#481)
* - start working on improving of cpu legacy code for reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving legacy loops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - still working on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - testing speed run of new reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - working on improvement of default loop for reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - update signatures of stuff which calls reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - make corrections in cuda reduce kernels
Signed-off-by: Yurii <iuriish@yahoo.com>
* - change loop for default case in broadcast legacy ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment some shape stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment unnecessary prints in RNGtests
Signed-off-by: Yurii <iuriish@yahoo.com>
* - finish to resolve conflicts after master has been merged
Signed-off-by: Yurii <iuriish@yahoo.com>
* - get rid of some compilation mistakes of cuda stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor changes
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further search for bug causing crash on java test
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add scalar case in reduce_ ... exec stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor corrections in NAtiveOps.cu
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add switch to scalar case execReduceXD functions
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in ConstantShapeHelper::createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
* - correct cuda mirrorPad
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in cuda createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
* Add support for CUDA 11.0 (#492)
* Add support for CUDA 11.0
* libnd4j tweaks for CUDA 11
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* bindings update, again?
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update versions of JavaCPP Presets for FFmpeg, OpenBLAS, and NumPy
* update API to match CUDA 8
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update version of JavaCPP Presets for CPython
* C++ updated for cuDNN 8.0
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* 128-bit alignment for workspaces
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* change seed in 1 test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Fix dependecy duplication in python4j-parent pom
* Fix group id for in python4j-numpy
* few tests tweaked
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Remove macosx-x86_64-gpu from nd4j-tests-tensorflow
* few minor tweaks for IndexReduce
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one test removed
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* RL4J: Add SyncTrainer and AgentLearnerBuilder for a few algorithms (#504)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* * Update versions of JavaCPP Presets for OpenCV, FFmpeg, and MKL
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Fix L2NormalizeVertex and eclipse#9054 (#513)
* update
* Fix L2NormalizeVertex
Fix eclipse#9054
* RL4J: Add async training and advantage actor-critic (#507)
* Added async training & Advantage Actor Critic
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Fix compiler error
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Renamed ActorCriticPolicy back to ACPolicy
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
* Python GIL overhaul (#517)
* Development updates (#9053)
* RL4J: Add generic update rule (#502)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Shyrma reduce (#481)
* - start working on improving of cpu legacy code for reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving legacy loops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - still working on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - testing speed run of new reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - working on improvement of default loop for reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - update signatures of stuff which calls reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - make corrections in cuda reduce kernels
Signed-off-by: Yurii <iuriish@yahoo.com>
* - change loop for default case in broadcast legacy ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment some shape stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment unnecessary prints in RNGtests
Signed-off-by: Yurii <iuriish@yahoo.com>
* - finish to resolve conflicts after master has been merged
Signed-off-by: Yurii <iuriish@yahoo.com>
* - get rid of some compilation mistakes of cuda stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor changes
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further search for bug causing crash on java test
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add scalar case in reduce_ ... exec stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor corrections in NAtiveOps.cu
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add switch to scalar case execReduceXD functions
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in ConstantShapeHelper::createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
* - correct cuda mirrorPad
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in cuda createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
* Add support for CUDA 11.0 (#492)
* Add support for CUDA 11.0
* libnd4j tweaks for CUDA 11
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* bindings update, again?
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update versions of JavaCPP Presets for FFmpeg, OpenBLAS, and NumPy
* update API to match CUDA 8
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update version of JavaCPP Presets for CPython
* C++ updated for cuDNN 8.0
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* 128-bit alignment for workspaces
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* change seed in 1 test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Fix dependecy duplication in python4j-parent pom
* Fix group id for in python4j-numpy
* few tests tweaked
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Remove macosx-x86_64-gpu from nd4j-tests-tensorflow
* few minor tweaks for IndexReduce
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one test removed
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* RL4J: Add SyncTrainer and AgentLearnerBuilder for a few algorithms (#504)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* Removed dead code (#9057)
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* performance improvement (#9055)
* performance improvement
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* revert some changes
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Development updates (#9064)
* Update versions of JavaCPP Presets for OpenCV, FFmpeg, and MKL
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Cherry pick rl4j changes from most recent KonduitAI/deeplearning4j PR
* Update cherry pick again from last master revision.
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
Co-authored-by: dariuszzbyrad <dariusz.zbyrad@gmail.com>
* Ag pythongiloverhaul (#518)
* Development updates (#9053)
* RL4J: Add generic update rule (#502)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Shyrma reduce (#481)
* - start working on improving of cpu legacy code for reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving legacy loops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - still working on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - testing speed run of new reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - working on improvement of default loop for reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - update signatures of stuff which calls reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - make corrections in cuda reduce kernels
Signed-off-by: Yurii <iuriish@yahoo.com>
* - change loop for default case in broadcast legacy ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment some shape stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment unnecessary prints in RNGtests
Signed-off-by: Yurii <iuriish@yahoo.com>
* - finish to resolve conflicts after master has been merged
Signed-off-by: Yurii <iuriish@yahoo.com>
* - get rid of some compilation mistakes of cuda stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor changes
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further search for bug causing crash on java test
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add scalar case in reduce_ ... exec stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor corrections in NAtiveOps.cu
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add switch to scalar case execReduceXD functions
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in ConstantShapeHelper::createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
* - correct cuda mirrorPad
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in cuda createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
* Add support for CUDA 11.0 (#492)
* Add support for CUDA 11.0
* libnd4j tweaks for CUDA 11
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* bindings update, again?
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update versions of JavaCPP Presets for FFmpeg, OpenBLAS, and NumPy
* update API to match CUDA 8
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update version of JavaCPP Presets for CPython
* C++ updated for cuDNN 8.0
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* 128-bit alignment for workspaces
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* change seed in 1 test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Fix dependecy duplication in python4j-parent pom
* Fix group id for in python4j-numpy
* few tests tweaked
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Remove macosx-x86_64-gpu from nd4j-tests-tensorflow
* few minor tweaks for IndexReduce
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one test removed
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* RL4J: Add SyncTrainer and AgentLearnerBuilder for a few algorithms (#504)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* Removed dead code (#9057)
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* performance improvement (#9055)
* performance improvement
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* revert some changes
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Development updates (#9064)
* Update versions of JavaCPP Presets for OpenCV, FFmpeg, and MKL
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Cherry pick rl4j changes from most recent KonduitAI/deeplearning4j PR
* Update cherry pick again from last master revision.
* Re update python4j
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
Co-authored-by: dariuszzbyrad <dariusz.zbyrad@gmail.com>
* Bump formatter-maven-plugin from 2.0.0 to 2.12.1 (#505)
Bumps [formatter-maven-plugin](https://github.com/revelc/formatter-maven-plugin) from 2.0.0 to 2.12.1.
- [Release notes](https://github.com/revelc/formatter-maven-plugin/releases)
- [Changelog](https://github.com/revelc/formatter-maven-plugin/blob/formatter-maven-plugin-2.12.1/CHANGELOG.md)
- [Commits](https://github.com/revelc/formatter-maven-plugin/compare/formatter-maven-plugin-2.0.0...formatter-maven-plugin-2.12.1)
Signed-off-by: dependabot-preview[bot] <support@dependabot.com>
Co-authored-by: dependabot-preview[bot] <27856297+dependabot-preview[bot]@users.noreply.github.com>
* Ag fix9060 (#519)
* Development updates (#9053)
* RL4J: Add generic update rule (#502)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Shyrma reduce (#481)
* - start working on improving of cpu legacy code for reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving legacy loops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - still working on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - testing speed run of new reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - working on improvement of default loop for reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - update signatures of stuff which calls reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - make corrections in cuda reduce kernels
Signed-off-by: Yurii <iuriish@yahoo.com>
* - change loop for default case in broadcast legacy ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment some shape stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment unnecessary prints in RNGtests
Signed-off-by: Yurii <iuriish@yahoo.com>
* - finish to resolve conflicts after master has been merged
Signed-off-by: Yurii <iuriish@yahoo.com>
* - get rid of some compilation mistakes of cuda stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor changes
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further search for bug causing crash on java test
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add scalar case in reduce_ ... exec stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor corrections in NAtiveOps.cu
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add switch to scalar case execReduceXD functions
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in ConstantShapeHelper::createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
* - correct cuda mirrorPad
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in cuda createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
* Add support for CUDA 11.0 (#492)
* Add support for CUDA 11.0
* libnd4j tweaks for CUDA 11
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* bindings update, again?
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update versions of JavaCPP Presets for FFmpeg, OpenBLAS, and NumPy
* update API to match CUDA 8
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update version of JavaCPP Presets for CPython
* C++ updated for cuDNN 8.0
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* 128-bit alignment for workspaces
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* change seed in 1 test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Fix dependecy duplication in python4j-parent pom
* Fix group id for in python4j-numpy
* few tests tweaked
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Remove macosx-x86_64-gpu from nd4j-tests-tensorflow
* few minor tweaks for IndexReduce
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one test removed
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* RL4J: Add SyncTrainer and AgentLearnerBuilder for a few algorithms (#504)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* Removed dead code (#9057)
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* performance improvement (#9055)
* performance improvement
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* revert some changes
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Development updates (#9064)
* Update versions of JavaCPP Presets for OpenCV, FFmpeg, and MKL
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Added support for the archunit (#9062)
* Added support for the archunit
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Updated pom files
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Datavec code cleaup (#9071)
* removed unnecessary semicolons
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Use standard charset object
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Removed unused imports
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* WIP: Fix Conv1d causal case
* Add inital tests
* Update Conv1d tests to be a bit more robust
* Remove redundant test
* Reset from master
* Remove cuda definition (left over)
* Update rl4j again
* Update pom.xml
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
Co-authored-by: dariuszzbyrad <dariusz.zbyrad@gmail.com>
* Fixes 9061 (#521)
* Get rid of edge case in validation
* Added support for the archunit (#9062)
* Added support for the archunit
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Updated pom files
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Using embedded copying of an array instead of manual (#9073)
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Datavec bulk operation (#9075)
* Bulk operation can be used instead of iteration inspection
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Redundant 'Collection.addAll()' call inspection
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
* Removed infinitely loop (#9076)
Signed-off-by: Dariusz Zbyrad <dariusz.zbyrad@gmail.com>
Co-authored-by: dariuszzbyrad <dariusz.zbyrad@gmail.com>
* RL4J: Add async training and advantage actor-critic (#507)
* Added async training & Advantage Actor Critic
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Fix compiler error
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Renamed ActorCriticPolicy back to ACPolicy
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
(cherry picked from commit 72f5c18c830f62df2c04fbf8dc7b1353cc2d3182)
* RL4J: Add async training and advantage actor-critic (#507)
* Added async training & Advantage Actor Critic
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Fix compiler error
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Renamed ActorCriticPolicy back to ACPolicy
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
(cherry picked from commit 72f5c18c830f62df2c04fbf8dc7b1353cc2d3182)
* Revert rl4j to 72f5c18c830f62df2c04fbf8dc7b1353cc2d3182
* Delete jnind4jaurora.cpp
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
Co-authored-by: dariuszzbyrad <dariusz.zbyrad@gmail.com>
Co-authored-by: dependabot-preview[bot] <27856297+dependabot-preview[bot]@users.noreply.github.com>
* RL4J: Add partial support for RNN (#514)
* Added partial recurrent support
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Made sure the RNN always see the observation in EpsGreedy
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Converted all line endings of rl4j-core to LF (#530)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* NDJ4: Bundle configuration files required by AOT compilation with GraalVM (#529)
* NDJ4: Bundle configuration files required by AOT compilation with GraalVM
* Update dependencies to just released JavaCPP and JavaCV 1.5.4
* Ag fixtests 831 (#523)
* Update UnderSamplingPreProcessorTest.java
* Development updates (#9053)
* RL4J: Add generic update rule (#502)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
* Shyrma reduce (#481)
* - start working on improving of cpu legacy code for reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving legacy loops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - still working on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further work on improving reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - testing speed run of new reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - working on improvement of default loop for reduce op
Signed-off-by: Yurii <iuriish@yahoo.com>
* - update signatures of stuff which calls reduce ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - make corrections in cuda reduce kernels
Signed-off-by: Yurii <iuriish@yahoo.com>
* - change loop for default case in broadcast legacy ops
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment some shape stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - comment unnecessary prints in RNGtests
Signed-off-by: Yurii <iuriish@yahoo.com>
* - finish to resolve conflicts after master has been merged
Signed-off-by: Yurii <iuriish@yahoo.com>
* - get rid of some compilation mistakes of cuda stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor changes
Signed-off-by: Yurii <iuriish@yahoo.com>
* - further search for bug causing crash on java test
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add scalar case in reduce_ ... exec stuff
Signed-off-by: Yurii <iuriish@yahoo.com>
* - minor corrections in NAtiveOps.cu
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add switch to scalar case execReduceXD functions
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in ConstantShapeHelper::createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
* - correct cuda mirrorPad
Signed-off-by: Yurii <iuriish@yahoo.com>
* - add support for vectors old shape in cuda createShapeInfoWithNoUnitiesForReduce
Signed-off-by: Yurii <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
* Add support for CUDA 11.0 (#492)
* Add support for CUDA 11.0
* libnd4j tweaks for CUDA 11
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* bindings update, again?
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update versions of JavaCPP Presets for FFmpeg, OpenBLAS, and NumPy
* update API to match CUDA 8
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* * Update version of JavaCPP Presets for CPython
* C++ updated for cuDNN 8.0
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one more test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* 128-bit alignment for workspaces
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* change seed in 1 test
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Fix dependecy duplication in python4j-parent pom
* Fix group id for in python4j-numpy
* few tests tweaked
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* Remove macosx-x86_64-gpu from nd4j-tests-tensorflow
* few minor tweaks for IndexReduce
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
* one test removed
Signed-off-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: raver119@gmail.com <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* RL4J: Add SyncTrainer and AgentLearnerBuilder for a few algorithms (#504)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* Development updates (#9064)
* Update versions of JavaCPP Presets for OpenCV, FFmpeg, and MKL
Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
* Add proper annotation
* Fix classcast exception for recurrent model import case
* Update keras import to allow for proper handling of changing NCHW -> NHWC mid later
* Add output to test to ensure proper activation
* Fixes computation graphs to allow dimension ordering to change mid graph
* Add NHWC support for keras import.
* Update tests to pass /ignore out of date ones
* Add multi RNNDataformat support
* Update tests to make more pass.
Updates some tests to be correct, double checked existing models and updated reasons they may or may not fail.
* Add back old default values to ensure legacy serialization works. Replace null value default with sentinel value for default value overridden.
* Update layers to preserve changed values
* Exclude default value over ridden from comparison
* Fix conv1d import (no permute weights anymore)
* Update KerasConvolution1D.java
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
* GPU compute capability (#532)
* - GPU cpu capability flags
- CUDA MAJOR VERSION provided by cmake
Signed-off-by: AbdelRauf <rauf@konduit.ai>
* Readme
Signed-off-by: AbdelRauf <rauf@konduit.ai>
* Readme
Signed-off-by: AbdelRauf <rauf@konduit.ai>
* RL4J: Add new network implementation to help support recurrent networks (#531)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
Co-authored-by: Alexandre Boulanger <44292157+aboulang2002@users.noreply.github.com>
Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
Co-authored-by: raver119 <raver119@gmail.com>
Co-authored-by: Samuel Audet <samuel.audet@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
Co-authored-by: dariuszzbyrad <dariusz.zbyrad@gmail.com>
Co-authored-by: dependabot-preview[bot] <27856297+dependabot-preview[bot]@users.noreply.github.com>
Co-authored-by: Abdelrauf <qwr@live.ru>
2020-09-23 12:11:29 +02:00
|
|
|
CUDA_COMPUTE="-DCOMPUTE=\"$COMPUTE\""
|
2019-06-06 14:21:15 +02:00
|
|
|
|
|
|
|
if [ "$CHIP" == "cuda" ] && [ -n "$CHIP_VERSION" ]; then
|
|
|
|
case $OS in
|
|
|
|
linux*)
|
|
|
|
export CUDA_PATH="/usr/local/cuda-$CHIP_VERSION/"
|
2021-03-05 02:59:02 +01:00
|
|
|
# Cross compilation for jetson nano
|
|
|
|
if [ "$ARCH" != "x86-64" ]; then
|
|
|
|
if [ "$ARCH" == "armv8-a" ]; then
|
|
|
|
export EXTRA_CUDA_FLAGS="-DCUDA_TARGET_CPU_ARCH=AARCH64"
|
|
|
|
else
|
|
|
|
export EXTRA_CUDA_FLAGS="-DCUDA_TARGET_CPU_ARCH=ARM"
|
|
|
|
fi
|
|
|
|
else
|
|
|
|
export EXTRA_CUDA_FLAGS=""
|
|
|
|
fi
|
2019-06-06 14:21:15 +02:00
|
|
|
;;
|
|
|
|
macosx*)
|
|
|
|
export CUDA_PATH="/Developer/NVIDIA/CUDA-$CHIP_VERSION/"
|
|
|
|
;;
|
|
|
|
windows*)
|
|
|
|
export CUDA_PATH="C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v$CHIP_VERSION/"
|
|
|
|
;;
|
|
|
|
esac
|
|
|
|
fi
|
|
|
|
|
2020-03-19 06:53:21 +01:00
|
|
|
|
2019-06-06 14:21:15 +02:00
|
|
|
[[ -z ${OPENBLAS_PATH:-} ]] && OPENBLAS_PATH=""
|
2020-03-19 06:53:21 +01:00
|
|
|
OPENBLAS_PATH="${OPENBLAS_PATH//\\//}"
|
2019-06-06 14:21:15 +02:00
|
|
|
|
|
|
|
if [[ -n "${BUILD_PATH:-}" ]]; then
|
|
|
|
PREVIFS="$IFS"
|
|
|
|
IFS="$BUILD_PATH_SEPARATOR"
|
|
|
|
for P in $BUILD_PATH; do
|
|
|
|
if [[ -f "$P/include/openblas_config.h" ]]; then
|
|
|
|
OPENBLAS_PATH="$P"
|
|
|
|
fi
|
|
|
|
done
|
|
|
|
IFS="$PREVIFS"
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [[ ! -f "$OPENBLAS_PATH/include/openblas_config.h" ]]; then
|
|
|
|
echo "Could not find OpenBLAS, please make sure to run the build with Maven or set the OPENBLAS_PATH variable"
|
|
|
|
OPENBLAS_PATH=""
|
|
|
|
fi
|
|
|
|
|
|
|
|
# replace any backslash with a slash
|
|
|
|
OPENBLAS_PATH="${OPENBLAS_PATH//\\//}"
|
|
|
|
|
|
|
|
mkbuilddir() {
|
|
|
|
if [ "$CLEAN" == "true" ]; then
|
|
|
|
echo "Removing blasbuild"
|
|
|
|
rm -Rf blasbuild
|
|
|
|
fi
|
|
|
|
mkdir -p "blasbuild/$CHIP"
|
|
|
|
cd "blasbuild/$CHIP"
|
|
|
|
}
|
|
|
|
|
2020-01-20 19:32:46 +01:00
|
|
|
HELPERS=""
|
2019-09-11 20:50:28 +02:00
|
|
|
if [ "$HELPER" == "" ]; then
|
|
|
|
echo "!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!"
|
|
|
|
echo "!! !!"
|
|
|
|
echo "!! !!"
|
|
|
|
echo "!! !!"
|
|
|
|
echo "!! !!"
|
|
|
|
echo "!! WARNING! !!"
|
|
|
|
echo "!! No helper packages configured! !!"
|
|
|
|
echo "!! You can specify helper by using -h key. I.e. <-h mkldnn> !!"
|
|
|
|
echo "!! !!"
|
|
|
|
echo "!! !!"
|
|
|
|
echo "!! !!"
|
|
|
|
echo "!! !!"
|
|
|
|
echo "!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!"
|
2020-01-20 19:32:46 +01:00
|
|
|
else
|
|
|
|
# if helpers were defined, we'll propagate them to CMake
|
|
|
|
IFS=','
|
|
|
|
read -ra HLP <<< "$HELPER"
|
|
|
|
for i in "${HLP[@]}"; do
|
|
|
|
HELPERS="${HELPERS} -DHELPERS_$i=true"
|
|
|
|
done
|
|
|
|
IFS=' '
|
2019-09-11 20:50:28 +02:00
|
|
|
fi
|
2019-06-06 14:21:15 +02:00
|
|
|
|
2021-03-05 02:59:02 +01:00
|
|
|
LINKER_FLAGS=""
|
|
|
|
if [ "$EXTRA_LINK_FLAGS" != "" ]; then
|
|
|
|
LINKER_FLAGS="-DCMAKE_CXX_LINK_FLAGS=$EXTRA_LINK_FLAGS -DCMAKE_EXE_LINKER_FLAGS=$EXTRA_LINK_FLAGS -DCMAKE_CUDA_FLAGS=$EXTRA_LINK_FLAGS"
|
|
|
|
fi
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-06 14:21:15 +02:00
|
|
|
echo PACKAGING = "${PACKAGING}"
|
|
|
|
echo BUILD = "${BUILD}"
|
|
|
|
echo CHIP = "${CHIP}"
|
|
|
|
echo ARCH = "${ARCH}"
|
|
|
|
echo CHIP_EXTENSION = "${CHIP_EXTENSION}"
|
|
|
|
echo CHIP_VERSION = "${CHIP_VERSION}"
|
|
|
|
echo GPU_COMPUTE_CAPABILITY = "${COMPUTE}"
|
|
|
|
echo EXPERIMENTAL = ${EXPERIMENTAL}
|
|
|
|
echo LIBRARY TYPE = "${LIBTYPE}"
|
|
|
|
echo OPERATIONS = "${OPERATIONS_ARG}"
|
|
|
|
echo MINIFIER = "${MINIFIER_ARG}"
|
|
|
|
echo TESTS = "${TESTS_ARG}"
|
|
|
|
echo NAME = "${NAME_ARG}"
|
|
|
|
echo OPENBLAS_PATH = "$OPENBLAS_PATH"
|
2020-01-28 17:00:12 +01:00
|
|
|
echo CHECK_VECTORIZATION = "$CHECK_VECTORIZATION"
|
2020-01-20 19:32:46 +01:00
|
|
|
echo HELPERS = "$HELPERS"
|
2021-03-05 02:59:02 +01:00
|
|
|
echo EXTRA_LINK_FLAGS = "$EXTRA_LINK_FLAGS"
|
|
|
|
echo EXTRA_CUDA_FLAGS = "$EXTRA_CUDA_FLAGS"
|
|
|
|
echo EXTRA_SYSROOT = "$EXTRA_SYSROOT"
|
2019-06-06 14:21:15 +02:00
|
|
|
mkbuilddir
|
|
|
|
pwd
|
2021-03-05 02:59:02 +01:00
|
|
|
eval "$CMAKE_COMMAND" "$EXTRA_SYSROOT" "$LINKER_FLAGS" "$EXTRA_CUDA_FLAGS" "$BLAS_ARG" "$ARCH_ARG" "$NAME_ARG" -DSD_CHECK_VECTORIZATION="${CHECK_VECTORIZATION}" "$HELPERS" "$SHARED_LIBS_ARG" "$MINIFIER_ARG" "$OPERATIONS_ARG" "$BUILD_TYPE" "$PACKAGING_ARG" "$EXPERIMENTAL_ARG" "$TESTS_ARG" "$CUDA_COMPUTE" -DOPENBLAS_PATH="$OPENBLAS_PATH" -DDEV=FALSE -DCMAKE_NEED_RESPONSE=YES -DMKL_MULTI_THREADED=TRUE ../..
|
2020-01-28 17:00:12 +01:00
|
|
|
|
2019-06-06 14:21:15 +02:00
|
|
|
if [ "$PARALLEL" == "true" ]; then
|
|
|
|
MAKE_ARGUMENTS="$MAKE_ARGUMENTS -j $MAKEJ"
|
|
|
|
fi
|
|
|
|
if [ "$VERBOSE" == "true" ]; then
|
2019-11-13 15:15:18 +01:00
|
|
|
MAKE_ARGUMENTS="$MAKE_ARGUMENTS $VERBOSE_ARG"
|
2019-06-06 14:21:15 +02:00
|
|
|
fi
|
2020-01-28 17:00:12 +01:00
|
|
|
|
|
|
|
if [ "$CHECK_VECTORIZATION" == "ON" ]; then
|
|
|
|
|
|
|
|
if [ "$MAKE_COMMAND" == "make" ]; then
|
|
|
|
MAKE_ARGUMENTS="$MAKE_ARGUMENTS --output-sync=target"
|
|
|
|
fi
|
2020-03-19 06:53:21 +01:00
|
|
|
|
2020-03-20 04:14:03 +01:00
|
|
|
exec 3>&1
|
2020-03-19 06:53:21 +01:00
|
|
|
eval "$MAKE_COMMAND" "$MAKE_ARGUMENTS" 2>&1 >&3 3>&- | python3 ../../auto_vectorization/auto_vect.py && cd ../../..
|
2020-01-28 17:00:12 +01:00
|
|
|
exec 3>&-
|
|
|
|
else
|
2020-03-19 06:53:21 +01:00
|
|
|
eval "$MAKE_COMMAND" "$MAKE_ARGUMENTS" && cd ../../..
|
2020-03-20 04:14:03 +01:00
|
|
|
fi
|