Menu
I am trying to set up an anaconda virtual environment with tensorflow-gpu. When I go to install it, it says that it will also be installing cudatoolkit and cudnn, both of which I already have installed and configured on my machine. Will this mess with my existing configuration? Will this all be local inside of the virtual environment?
Note that the versions that it is asking to install are NOT the same as those already on my machine.
I currently have another virtual environment which runs tensorflow-gpu perfectly well, but it never asked me to install these packages.
For reference, I am running Ubuntu 18.04, Cuda 9.0, and cuDNN 7.0.3
Thanks for the help! I've spent a long time getting the packages configured correctly and don't want to have to start over.
676
Step 1: Register an nvidia developer account and download cudnn here (about 80 MB). You might need nvcc --version to get your cuda version. Step 2: Check where your cuda installation is. For most people, it will be /usr/local/cuda/ .
cuDNN is built on top of the CUDA framework which is how you use NVIDIA GPUs for general purpose computing tasks. High performance GPU acceleration is helpful for machine learning tasks because it it allows computers to speed up processes, saving you time.
To build an application, a developer has to install only the CUDA Toolkit and necessary libraries required for linking. In order to run a CUDA application, the system should have a CUDA enabled GPU and an NVIDIA display driver that is compatible with the CUDA Toolkit that was used to build the application itself.
Anaconda cudatoolkit doesnt contain all the files which are included in a full CUDA install. Rather cudatoolkit includes the libraries which are redistributable and are used in downstream packages like Tensorflow, PyTorch, etc
[1] https://github.com/ContinuumIO/anaconda-issues/issues/9676
176
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With
