Question

tensorflow gpu与cudnn版本不兼容?

tensorflow gpu与cudnn版本不兼容?之前能用的，不知道怎么鼓捣坏了。。显卡:NVIDIA GeForce GTX 950Mtensorflow:1.5.0cuda:9.0cudnn:7.0.5报错信息:Loaded runtime CuDNN library: 7104 (compatibility version 7100) but source was compiled with 7003 (compatibility version 7000). If using a binary install, upgrade your CuDNN library to match. If building from sources, make sure the library loaded at runtime matches a compatible version specified during compile configuration.我conda install tensorflow-gpu会自动给我装1.13的，用不成，安装tensorflow1.6版本以上的就出bug。还有conda install tensorflow-gpu==1.5.0会出错，只能用pip安装。。

Answer 1

安装GPU版本的tensorflow就可以了pipinstall--upgrade还需要安装GPU加速包，CUDA和cuDNN的安装包了，注意版本号分别是CUDA8.0和cuDNN5.1。如果都安装成功，运行程序时会自动使用GPU计算

追问

你没有读题