CUDNN Error On Runtime

AI & Data Science Deep Learning (Training & Inference) cuDNN
1

2020-02-24 13:20:14.131414: E tensorflow/stream_executor/cuda/cuda_dnn.cc:378] Loaded runtime CuDNN library: 7105 (compatibility version 7100) but source was compiled with 7005 (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.
2020-02-24 13:20:14.133105: F tensorflow/core/kernels/conv_ops.cc:717] Check failed: stream->parent()->GetConvolveAlgorithms( conv_parameters.ShouldIncludeWinogradNonfusedAlgo(), &algorithms)
Aborted

We are using jetson TX2

cat /usr/local/cuda/version.txt
CUDA Version 9.0.252

cat /usr/include/cudnn.h | grep CUDNN_MAJOR -A 2
#define CUDNN_MAJOR 7
#define CUDNN_MINOR 1
#define CUDNN_PATCHLEVEL 5

Please provide a solution for issue.

2

Hi,

As per the error, it seems that the code is complied on different version.
“Loaded runtime CuDNN library: 7105 (compatibility version 7100) but source was compiled with 7005 (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.”

Either you have to rebuild the code on latest cuDNN version or downgrade it to the version using which code is originally complied.

Thanks