# Issue with PyTorch dependency

## post by sindri on Mar 24, 2021

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?

1.5k views

## post by rhuddleston on May 11, 2021

Can you find the library anywhere on your system?

## post by sindri on May 11, 2021

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.

### Related topics

| Topic | Replies | Views | Activity |
| --- | --- | --- | --- |
| [Could not load library libcudnn_adv_train.so.8 error on lambda workstation](https://deeptalk.lambda.ai/t/could-not-load-library-libcudnn-adv-train-so-8-error-on-lambda-workstation/2698) | [5](https://deeptalk.lambda.ai/t/could-not-load-library-libcudnn-adv-train-so-8-error-on-lambda-workstation/2698/1) | 7.1k | [Sep 2022](https://deeptalk.lambda.ai/t/could-not-load-library-libcudnn-adv-train-so-8-error-on-lambda-workstation/2698/6) |
| [Fresh 20.04.1 LTS + Lambda Stack - libcudnn_adv_train.so.8](https://deeptalk.lambda.ai/t/fresh-20-04-1-lts-lambda-stack-libcudnn-adv-train-so-8/1974) | [3](https://deeptalk.lambda.ai/t/fresh-20-04-1-lts-lambda-stack-libcudnn-adv-train-so-8/1974/1) | 2.8k | [Apr 2021](https://deeptalk.lambda.ai/t/fresh-20-04-1-lts-lambda-stack-libcudnn-adv-train-so-8/1974/4) |
| [Lambda Stack - CUDNN 8 Upgrade Query](https://deeptalk.lambda.ai/t/lambda-stack-cudnn-8-upgrade-query/1966) | [8](https://deeptalk.lambda.ai/t/lambda-stack-cudnn-8-upgrade-query/1966/1) | 4.7k | [Apr 2021](https://deeptalk.lambda.ai/t/lambda-stack-cudnn-8-upgrade-query/1966/9) |
| [Updated Lambda Stack and now have a PyTorch CuDNN version mismatch Error \[resolved\]](https://deeptalk.lambda.ai/t/updated-lambda-stack-and-now-have-a-pytorch-cudnn-version-mismatch-error-resolved/629) | [5](https://deeptalk.lambda.ai/t/updated-lambda-stack-and-now-have-a-pytorch-cudnn-version-mismatch-error-resolved/629/1) | 3.9k | [Dec 2018](https://deeptalk.lambda.ai/t/updated-lambda-stack-and-now-have-a-pytorch-cudnn-version-mismatch-error-resolved/629/7) |
| [VRAM-hungry LSTM monster](https://deeptalk.lambda.ai/t/vram-hungry-lstm-monster/3035) | [6](https://deeptalk.lambda.ai/t/vram-hungry-lstm-monster/3035/1) | 1.8k | [Jun 2022](https://deeptalk.lambda.ai/t/vram-hungry-lstm-monster/3035/7) |