Intel® oneAPI Deep Neural Network Library (oneDNN) is an open-source performance library for deep learning applications. The library includes basic building blocks for neural networks optimized for Intel Architecture Processors and Intel Processor Graphics. oneDNN is intended for deep learning applications and framework developers interested in improving application performance on Intel CPUs and GPUs. You can find library source code and code used by these samples at the oneDNN Github repository.
This sample guides users how to compile oneDNN applications and executes the binaries on different Intel architectures. The sample also includes a Jupyter Notebook that demonstrates how to compile the code with various oneDNN configurations in Intel® DevCloud for oneAPI environment.
| Optimized for | Description |
|---|---|
| OS | Linux* Ubuntu* 18.04; Windows |
| Hardware | Skylake with GEN9 or newer |
| Software | Intel® oneAPI Deep Neural Network Library (oneDNN) Intel® oneAPI DPC++/C++ Compiler Intel® oneAPI Threading Building Blocks (oneTBB) GNU Compiler Collection Intel® C++ Compiler |
| What you will learn | Running a simple convolutional model on Intel CPU or Intel GPU |
| Time to complete | 15 minutes |
This sample demonstrates how to compile oneDNN applications and executes the binaries on different Intel architectures by using Intel® oneAPI Base Toolkit.
With this code sample, you will learn:
- How to configure oneDNN environment for different pre-built oneDNN binaries from Intel® oneAPI Base Toolkit.
- How to compile oneDNN applications with different pre-built oneDNN binaries from Intel® oneAPI Base Toolkit.
- How to run oneDNN applications on different Intel architectures.
- How to enable oneDNN verbose log for validation.
All compiled binaries execute on the system's CPU by default and can be executed on Intel GPU
using a command line parameter gpu.
This sample uses example files under ${DNNLROOT}/examples/
from oneDNN distribution. You can find those codes in
oneDNN Github repository.
Detailed code walkthrough is available in oneDNN developer guide.
Perform the following steps:
source ${INTEL_ONEAPI_INSTALL_FOLDER}/setvars.sh
mkdir build
cd build
cmake ..
make
NOTE: The source files such as "getting_started.cpp" will be copied from
${INTEL_ONEAPI_INSTALL_FOLDER}/dnnl/latest/cpu_dpcpp_gpu_dpcpp/examples/
to build/src folder. Users can rebuild the all source codes by typing
"make" under build folder.
You can get additional information during the execution of this sample by setting
environment variable DNNL_VERBOSE=1.
Take getting_started.cpp as an example.
export DNNL_VERBOSE=1
./bin/getting-started-cpp
By default, the sample uses the Intel® oneAPI DPC++/C++ Compiler and will execute on CPUs or Intel GPUs. You can build the sample with CPU support with other compilers and threading runtimes:
- GNU C++ Compiler and GNU OpenMP runtime
source ${INTEL_ONEAPI_INSTALL_FOLDER}/setvars.sh --dnnl-configuration=cpu_gomp
mkdir build
cd build
CC=gcc CXX=g++ cmake ..
make
- Intel® C++ Compiler and Intel OpenMP runtime
source ${INTEL_ONEAPI_INSTALL_FOLDER}/setvars.sh --dnnl-configuration=cpu_iomp
mkdir build
cd build
CC=icx CXX=icpx cmake ..
make
- Intel® C++ Compiler and TBB runtime
source ${INTEL_ONEAPI_INSTALL_FOLDER}/setvars.sh --dnnl-configuration=cpu_tbb
mkdir build
cd build
CC=icx CXX=icpx cmake ..
make
Open "Intel oneAPI command prompt for Intel 64 for Visual Studio 2017" or "Intel oneAPI command prompt for Intel 64 for Visual Studio 2019" and perform the following steps:
C:\Program Files (x86)\Intel\oneAPI\setvars.bat --dnnl-configuration=cpu_vcomp --force
mkdir build
cd build
cmake -G "Visual Studio 16 2019" ..
cmake --build .
C:\Program Files (x86)\Intel\oneAPI\setvars.bat
2. Download oneAPI Level Zero headers from Github and unpack the archive.
NOTE: Intel oneAPI DPC++/C++ Compiler on Windows requires CMake v3.21 or later
mkdir build
cd build
set CC=icx
set CXX=icx
cmake .. -G Ninja -DDNNL_CPU_RUNTIME=DPCPP
-DDNNL_GPU_RUNTIME=DPCPP
-DCMAKE_PREFIX_PATH=<path to Level Zero headers>
You can get additional information during the execution of this sample by setting
environment variable DNNL_VERBOSE=1.
set DNNL_VERBOSE=1
Take getting_started.cpp as an example.
bin\Debug\getting-started-cpp.exe
If running a sample in the Intel DevCloud, remember that you must specify the compute node (CPU, GPU) and whether to run in batch or interactive mode. For more information, see the Intel® oneAPI Base Toolkit Get Started Guide (https://devcloud.intel.com/oneapi/get-started/base-toolkit/).
You can specify the target device for this sample using command-line arguments:
cpu(default) directs the application to run on the system's CPUgpudirects the sample to run on Intel GPU
Note: When executed with
gpuparameter the sample will return an error if the sample is compiled with oneDNN configuration that does not support GPU, or no Intel GPUs are found in the system.
You can get additional information during the execution of this sample by setting
environment variable DNNL_VERBOSE=1.
The include folder is located at ${DNNLROOT}\include on your development system.
Example passed on CPU.
When executed with DNNL_VERBOSE=1:
dnnl_verbose,info,cpu,runtime:DPC++
dnnl_verbose,info,cpu,isa:Intel AVX2
dnnl_verbose,info,gpu,runtime:DPC++
dnnl_verbose,info,cpu,engine,0,backend:OpenCL,name:Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz,driver_version:2020.10.7
dnnl_verbose,info,gpu,engine,0,backend:Level Zero,name:Intel(R) Gen12LP,driver_version:0.8.0
dnnl_verbose,exec,cpu,eltwise,jit:avx2,forward_inference,data_f32::blocked:acdb:f0 diff_undef::undef::f0,,alg:eltwise_relu alpha:0 beta:0,1x3x13x13,0.125
Example passed on CPU.
If an error occurs, troubleshoot the problem using the Diagnostics Utility for Intel® oneAPI Toolkits. Learn more.
Code samples are licensed under the MIT license. See License.txt for details.
Third party program Licenses can be found here: third-party-programs.txt.