keras/TensorFlow 报错could not create cudnn handle: CUDNN_STATUS_NOT_INITIALIZED解决方法

keras/TensorFlow 报错如下:

could not create cudnn handle: CUDNN_STATUS_NOT_INITIALIZED

error retrieving driver version: Unimplemented: kernel reported driver version not implemented on Windows

could not destroy cudnn handle: CUDNN_STATUS_BAD_PARAM

Check failed: stream->parent()->GetConvolveAlgorithms( conv_parameters.ShouldIncludeWinogradNonfusedAlgo(), &algorithms)

解决办法:

换一个版本的cudnn。我原来是cuda9.0+cudnn7.0.5,程序跑得好好的,突然有一次跑的时候就出现上面的错误,怎么弄都无法解决,最后我把cudnn7.0.5换成cudnn7.1.4,问题解决

你可能感兴趣的:(keras/TensorFlow 报错could not create cudnn handle: CUDNN_STATUS_NOT_INITIALIZED解决方法)