Issue with PyTorch dependency

Hi, I’ve just recently been introduced with lambda stack, ever since I got access to a newly installed gpu server.

when using PyTorch, wherever I try to call backward() while training a model on a GPU I get the following error message after my program crashes.

Could not load library libcudnn_cnn_train.so.8. Error: libcudnn_ops_train.so.8: cannot open shared object file: No such file or directory
Please make sure libcudnn_cnn_train.so.8 is in your library path!
Aborted (core dumped)

could someone please help me understand what is going on?

Can you find the library anywhere on your system?

I eventually asked the professor in charge of the server to update, and restart the server which resolved my issue. before that he suggest I post my issue here first.

1 Like