Problem with cuda version of python tensorflow-gpu in anaconda

While learning deep learning, I'm going to use Anaconda and Pi-Cham to run deep learning I want to take some of them from the paper and turn them into tensorflow-gpu.

Request CUDA 10.0 ("cudart64_100.dll") in Faicham Anaconda requested CUDA 10.2, so it's not compatible right now

I've tried all the ways to download the CUDA version 10.0 or 10.2 or both

Is there a way to change the CUDA virtual environment to 10.0 in Anaconda?

cuda tensorflow python gpu

2022-09-21 14:02

1 Answers

https://blog.kovalevskyi.com/multiple-version-of-cuda-libraries-on-the-same-machine-b9502d50ae77

Please read the post above.

I haven't done it myself. I think LD_LIBRARY_PATH environment variable is referenced when using cuda library. After activating each of the two environments, you can change it to the correct version of the directory name and set it up.

The above posting is based on Ubuntu environmental standards, so it's different from setting environmental variables in the Windows environment, so I think you can just change it and set it up.

Hard Luck.

2022-09-21 14:02

If you have any answers or tips

Popular Tags

python x 4647

android x 1593

java x 1494

javascript x 1427

c x 927

c++ x 878

ruby-on-rails x 696

php x 692

python3 x 685

html x 656