’s install guide for Windows suggests an older CUDA, it seems:Ĭonda install -c conda-forge cudatoolkit=11.2 cudnn=8.1.0 Starting with TensorFlow 2.11, you will need to install TensorFlow in WSL2, or install tensorflow-cpu and, optionally, try the TensorFlow-DirectML-Plugin” The latest Tensorflow version seems to be 2.11 (but not Win10 compatible?)Īt, I see the following (scary) message: “TensorFlow 2.10 was the last TensorFlow release that supported GPU on native-Windows. The latest CUDA version is 12.1 (I currently have 11.7) The latest driver for my GPU (RTX 2070 Super, Windows 10) is 528.49 (I currently have 516.94). All of the TensorFlow versions work with DeepLabCut.Install CUDA (versions up to CUDA11 are supported, together with TF2.5).Note, DeepLabCut is up to date with the latest CUDA and tensorflow versions!.Here are some of the relevant texts from the install page (which I find confusing/inconsistent): In short, what are the current recommendations for the GPU install? I’ve carefully read the installation instructions and tips and followed links to external (stackoverflow) threads, but I feel the install docs are not consistent. I recall from earlier installations that this is tricky business, and want to get it right and avoid a mess. Hi, I’m trying to get up and running with the latest DLC (use case is single subject) and am stuck on the GPU-related aspects.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |