Main Menu

1. Installation of CUDA

URL of the Manual

    sudo apt update && sudo apt upgrade -y
    
    sudo vi /etc/modprobe.d/blacklist-nvidia-nouveau.conf

    #paste the below provided lines
    blacklist nouveau
    options nouveau modeset=0
    #save and exit

    sudo update-initramfs -u

    sudo lshw -C display
    # The above command must show the GPU devices mounted in a virtual machine

Click here to download the CUDA library

    wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb
    sudo dpkg -i cuda-keyring_1.1-1_all.deb
    sudo apt-get update
    sudo apt-get -y install cuda-toolkit-12-3

    sudo apt-get install -y cuda-drivers
    sudo apt-get -y install cuda

    # check the cuda version at /usr/local/cuda-<version>
      E.g., /usr/local/cuda-12.3
    echo 'export PATH=/usr/local/cuda-12.3/bin${PATH:+:${PATH}}' >> ~/.bashrc
    echo 'export LD_LIBRARY_PATH=/usr/local/cuda-12.3/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}' >> ~/.bashrc

    source ~/.bashrc

    sudo reboot  #Restart the server.

    nvidia-smi
    #above command shows the GPUs that exists in the machine  

2. Installation of CUDNN

Download the latest version of CUDNN from NVidia website

    sudo dpkg -i cudnn-local-repo-ubuntu2204-8.9.7.29_1.0-1_amd64.deb

Check the last line displayed after executing above command and execute it.

    sudo cp /var/cudnn-local-repo-ubuntu2204-8.9.7.29/cudnn-local-08A7D361-keyring.gpg /usr/share/keyrings/

    sudo apt-get update
    sudo apt-get install libcudnn8

3. Installation of TensorFlow, Keras, and JAX

    conda activate jupyterHub

    conda update -n base -c defaults conda
    or
    conda update --all

    conda install -c conda-forge tensorflow
    or
    conda install tensorflow=2.12.*=gpu_*

    
    conda install keras


    #The below package is not advised as it may create errors in tensorflow  
    pip install --upgrade "jax[cpu]"     #CPU only

4. Installation of Scikit (sklearn)

    conda install -c conda-forge scikit-learn

5. Installation of Dask, Pytorch and Prophet

    conda install -c rapidsai -c conda-forge -c nvidia dask-cuda cuda-version=12.3

    conda install pytorch::pytorch
    or
    conda install conda-forge::pytorch-gpu

    conda install -c conda-forge prophet

6. Installation of GDAL

    conda install -c conda-forge gdal

7. Installation of PAMI and other packages

    pip install pami

    pip install autots

    pip install auto-ts

    pip install darts

    pip install etna[all]

    pip install greykite

    pip install kats