Linear_classifiers

<<<<<<< HEAD

OpenAI_farmbot_Alphagarden

=======

Installing the correct versions of CUDA, PyTorch

For Dell G15 + RTX 3060 GPU, a driver version of Driver Version: 495.29.05 & CUDA Version: 11.5 works with Pytorch 1.10.2.

For a local installation of PyTorch, use Stable 1.10.2 with Linux, Pip, Python, CUDA11.3 chosen at https://pytorch.org/get-started/locally/ Use the following command:

pip3 install torch==1.10.2+cu113 torchvision==0.11.3+cu113 torchaudio==0.10.2+cu113 -f https://download.pytorch.org/whl/cu113/torch_stable.html

This pytorch is for CUDA 11.3 but works for CUDA 11.5 as well.

DO not install CUDA 11.6

DO NOT install Driver version 510.39.01 with CUDA version 11.6 according to the following link: https://developer.nvidia.com/cuda-downloads?target_os=Linux&target_arch=x86_64&Distribution=Ubuntu&target_version=18.04&target_type=deb_local

Reboot your computer after making any changes for them to take effect.

For a local installation of PyTorch, use Stable 1.10.2 with Linux, Pip, Python, CUDA11.3 chosen at https://pytorch.org/get-started/locally/ Use the following command:

pip3 install torch==1.10.2+cu113 torchvision==0.11.3+cu113 torchaudio==0.10.2+cu113 -f https://download.pytorch.org/whl/cu113/torch_stable.html

Surprisngly, Pytorch 1.10.2 works with CUDA version 11.6 as well. For RTX 3060 on Dell G15, the driver version 510.39.01 works with CUDA Version: 11.6. But this driver verion is very UNSTABLE. Computer hangs, crashes and fails to boot half of the time.

51f813872a24a242207d0a7b3bff6dcea0f8e87b

OpenAI_farmbot_Alphagarden

OpenAI_farmbot_Alphagarden