tensorflow training gpu vs cpu

基于mnist手写数字训练集的效率比较

硬件配置:
interl i7-6700hq 2.60GHz * 8core
nvidia GeForce 940MX显卡

软件配置:
gpu版本:python3.7.0, tensorflow-gpu 1.13.1, cuda 10.0, cudnn 7.5.1
cpu版本:python3.5.4, tensorflow 1.13.1

测试对比:
下图所示:cpu8核跑满, gpu利用率84% : gpu training速度 2倍+ 于 cpu速度
tensorflow training gpu vs cpu
 
参考:
cuda下载:
https://developer.nvidia.com/cuda-toolkit-archive
cudnn下载:
https://developer.nvidia.com/rdp/cudnn-download
gpu查看:
c:\Program Files\NVIDIA Corporation\NVSMI>nvidia-smi.exe


相关推荐