It is unclear how a user should configure LD_LIBRARY_PATH after pip install tensorflow[and-cuda] about tensorflow HOT 6 OPEN

I am having an incredibly frustrating experience. I followed the exact steps outlined here (https://www.tensorflow.org/install/pip), and despite my best efforts and hours of searching I keep getting "Cannot dlopen some GPU libraries" .. "Skipping registering GPU devices...".

I keep reading something about the LD_LIBRARY_PATH however I cannot find an obvious fix especially as I am using a python venv! I miss the days when tf-gpu worked on windows. This is pushing me to pytorch because I am exhausted trying different commands and path configurations to no avail. Why is it so difficult to use my GPU with Tensorflow?

from tensorflow.

@jackhart We had a long discussion about it and figured it out: #63362

It's a path issue. Supposedly, there's a pull request where this gets fixed either in 2.16.2 or 2.17.

from tensorflow.

@jackhart it seems practically impossible for someone owning a PC with CUDA-enabled GPU to perform deep learning experiments with TensorFlow version 2.16.1 and utilize his GPU locally without manually performing some extra steps not included (until today) in the official TensorFlow documentation of the standard installation procedure of TensorFlow for Linux users with GPUs at least as a temporal fix! That's why I submitted the pull request in good faith and for the shake of all users as TensorFlow is "An Open Source Machine Learning Framework for Everyone".

Hope that the next patch version of TensorFlow will fix the bug as soon as possible!

tensorflow/docs#2299

from tensorflow.

I am having an incredibly frustrating experience. I followed the exact steps outlined here (https://www.tensorflow.org/install/pip), and despite my best efforts and hours of searching I keep getting "Cannot dlopen some GPU libraries" .. "Skipping registering GPU devices...".

I keep reading something about the LD_LIBRARY_PATH however I cannot find an obvious fix especially as I am using a python venv! I miss the days when tf-gpu worked on windows. This is pushing me to pytorch because I am exhausted trying different commands and path configurations to no avail. Why is it so difficult to use my GPU with Tensorflow?

@Edfrost321 a revised document addressing the issue of TensorFlow installation for Linux users with GPUs is pending review. Here is the link: https://github.com/sgkouzias/docs/blob/patch-1/site%2Fen%2Finstall%2Fpip.md

I hope it helps!

from tensorflow.

Hi @jackhart ,
Sorry for the delay, discussion is going on belongs to this issue already #63362 , #2299. Once we get update, let you know.

Thank you!

from tensorflow.

@sgkouzias

That's a huge help, thank you!

from tensorflow.