You can find more information about environmental variables and how to set them here. You must do that for the path variables you've listed, i.e.: CUDA_NVCC_EXECUTABLE Or adding them manually on the configure file. Browse other questions tagged python linux tensorflow or ask your own question. you can install docker using Install Docker Engine on Ubuntu or use the Digital Ocean guide (i used this one) How To Install and Use Docker on Ubuntu 18.04. it is possible that I made some wrong choices, but I couldn't tell which. Install CUDA 10.1 by installing nvidia-cuda-toolkit. I was not able to fix them, so I decided to uninstall it using the instruction on the nvidia web-site 1 : sudo apt-get -purge remove Ubuntu.You can do this exporting their values on your system before configuring the library, like: export CUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda Yesterday, my Ubuntu 20.04 machine realized it had CUDA conflicts. That should solve the problem of the CUDA libraries and driver installation, but if you are building your library manually on your system, perhaps you need to do some additional work setting the needed paths before building your library. There are many ways, but perhaps the simplest is to write them on your ~./bashrc Then you have to set the environmental variables on the system (this is part of the post-installation tasks detailed on the CUDA toolkit documentation). In order to install them: sudo dpkg -i cuda-repo-ubuntu1404_6.5-14_b I guess you only have installed the repo, not the packages.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |