Comping with CUDA8 - CuDNN5

TimoSaemann / caffe-segnet-cudnn5

This repository was a fork of BVLC/caffe and includes the upsample, bn, dense_image_data and softmax_with_loss (with class weighting) layers of caffe-segnet (https://github.com/alexgkendall/caffe-segnet) to run SegNet with cuDNN version 5.

Other

176 stars 127 forks source link

Comping with CUDA8 - CuDNN5 #5

Closed whatnick closed 7 years ago

whatnick commented 7 years ago

Hi, I am getting this error while compiling. Is the port complete ? "int" is incompatible with parameter of type "cudnnNanPropagation_t"

Using CMake to configure build:

-- CUDA detected: 8.0
-- Found cuDNN: ver. 5.0.5 found (include: /usr/local/cuda/include, library: /usr/local/cuda/lib64/libcudnn.so)
-- Added CUDA NVCC flags for: sm_52 sm_30

TimoSaemann commented 7 years ago

Actually this error occurs when the used cuDNN version is not supported, but the repository supports definitely cuDNN version 5. Are you sure you compiled caffe-segnet-cudnn5 ?

whatnick commented 7 years ago

Checked out this repo specifically for that reason. Will get my ML machine back in a couple of days and start fresh to try again.