lightGBM gpu环境配置

推荐先看一手官方的Installation Guide.我用的是ubuntu 16.04,一些要求如下图:

主要是OpenCL以及libboost两个环境的要求.

(1) OpenCL的安装.我这里之前安装了NVIDA-395的驱动以及CUDA 9.0,只需要在编译的时候给出对应的path就可以了(这部分后面编译的时候会有解释).可以用clinfo查看OpenCL的信息.

Lenovo-Rescuer-15ISK:~$ clinfo
Number of platforms
  Platform Name                                   NVIDIA CUDA
  Platform Vendor                                 NVIDIA Corporation
  Platform Version                                OpenCL 1.2 CUDA 9.1.
  Platform Profile                                FULL_PROFILE
  Platform Extensions                             cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64 cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_copy_opts cl_nv_create_buffer
  Platform Extensions function suffix             NV

  Platform Name                                   NVIDIA CUDA
Number of devices
  Device Name                                     GeForce GTX 960M
  Device Vendor                                   NVIDIA Corporation
  Device Vendor ID                                0x10de
  Device Version                                  OpenCL 1.2 CUDA
  Driver Version                                  390.87
  Device OpenCL C Version                         OpenCL C 1.2 

(2)libboost安装.

> 从https://dl.bintray.com/boostorg/release/1.69.0/source/boost_1_69_0.tar.gz下载对应的压缩包

> 解压缩 tar -xzvf boost_1_69_0.tar.gz

> 切换到对应的文件目录下 sudo ./bootstrap.sh

> 根据提示 sudo ./b2 install 这样头文件就被默认安装在/usr/local/include头文件下，库文件就被默认安装在/usr/local/lib下

上述两个依赖搞定之后,安装官方给的安装方式(linux)如下:

git clone --recursive https://github.com/microsoft/LightGBM ; cd LightGBM
mkdir build ; cd build
cmake -DUSE_GPU= ..
# if you have installed NVIDIA CUDA to a customized location, you should specify paths to OpenCL headers and library like the following:
# cmake -DUSE_GPU= -DOpenCL_LIBRARY=/usr/local/cuda/lib64/libOpenCL.so -DOpenCL_INCLUDE_DIR=/usr/local/cuda/include/ ..
make -j4

注意,如果已经安装过cuda的要指出OpenCL所在的位置,之后正常编译就好了.