Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

What are the cuda, cudnn and caffe version used #1

Open
jaiabhayk opened this issue Mar 27, 2017 · 1 comment
Open

What are the cuda, cudnn and caffe version used #1

jaiabhayk opened this issue Mar 27, 2017 · 1 comment

Comments

@jaiabhayk
Copy link

jaiabhayk commented Mar 27, 2017

Hi,
Please can you update the cuda, cudnn and caffe version used ?
I am getting different kind of errors while trying to compile with different cudnn versions.

cd caffe-maste-jps
make clean all

It would be great if you can tell me the source files changed for customized caffe (that way I can use the latest cudnn version which wiil be faster) ?

@MehdiNoroozi
Copy link
Owner

MehdiNoroozi commented Jul 30, 2017

Hi,

Sorry for late reply. I saw your message just before. We have trained a new version using stride 4 in transfer learning and got much better results. Will release in Aug.

I have changed several files. The faster way would be to compile my branch with the new cudnn. You need to replace the following files with those in a version of caffe that works in your machine.

include/caffe/util/cudnn.hpp
include/caffe/layers/cudnn_*
src/caffe/util/cudnn.cpp
src/caffe/layers/cudnn_*

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants