Crash: Could not create cuDNN handle when convnets are used(环境tensorflow1.2-cuda8.0-cudnn5.1.10)

本机环境

今天遇到一个安装时候的问题，所以在这里进行一下记录，也为其他人提供些方便。
首先，我使用的Ubuntu16.04，显卡GeForce GTX 1080，环境tensorflow1.2-cuda8.0-cudnn5.1.10。
在运行程序的时候出现一下错误：

2018-04-05 11:09:14.862073: E tensorflow/stream_executor/cuda/cuda_dnn.cc:359] could not create cudnn handle: CUDNN_STATUS_INTERNAL_ERROR
2018-04-05 11:09:14.862113: E tensorflow/stream_executor/cuda/cuda_dnn.cc:326] could not destroy cudnn handle: CUDNN_STATUS_BAD_PARAM
2018-04-05 11:09:14.862120: F tensorflow/core/kernels/conv_ops.cc:671] Check failed: stream->parent()->GetConvolveAlgorithms(&algorithms) 
已放弃 (核心已转储)

解决方案

问题在github上面得到了解答
https://github.com/tensorflow/tensorflow/issues/6698
主要解决方法是

#Just for those who are driven mad by this:

#I occasionally got a CUBLAS error as well. So I did this:

cd /usr/local/cuda/samples/7_CUDALibraries/simpleCUBLAS
make
./simpleCUBLAS

#and discovered that I could not initialise CUBLAS

#So next I did this (based on advice)

sudo rm -f ~/.nv

#And it worked. Cheers.....thats 4 days wasted. Hope this saves someone else

简单一些说，当你遇到这个问题，你可以通过

sudo rm -f ~/.nv

这个操作进行解决问题，简单粗暴。希望大家可以在这方面节省些时间，顺利科研。