The version of Tensorflow you select will determine the compatible versions of CUDA, cuDNN, compiler, toolchain and the Nvidia driver versions to install. Step 1: Checking versions of drivers and software to install for compatibility with Tensorflow 2.1 Conclusions and a Note on Keras and Tensorflow.Step 14: Checking Correct Installation of Tensorflow and GPU support.Step 13.2: Install Tensorflow 2 in the virtual environment.Step 13.1: Set up a new Pycharm project with virtual environment.Step 13: Install Tensorflow (recommended: installation in a virtual environment).Step 9: Installing CuDNN for Ubuntu 18.04.Step 8: Downloading CuDNN for Ubuntu 18.04.Step 7: Reboot the computer again (as per instructions above) and check the driver version.Step 6: Install CUDA 10.1 and CuDNN development and runtime libraries.Step 5: Check the new Nvidia GPU driver installation.
Step 4: Install NVIDIA Driver version 430.Step 3: Download NVIDIA package repositories.Step 2.5: Check any current Nvidia drivers.Step 2.4: Check that Ubuntu system has the correct kernel headers and development packages are installed.Step 2.2: Check for your upcoming CUDA installation that your version of linux is supported.Step 2.1: Check the system has a CUDA capable GPU.Step 2: Pre-CUDA installation: check existing installations.Step 1: Checking versions of drivers and software to install for compatibility with Tensorflow 2.1.Update-alternatives: using /usr/local/cuda-11. Press to keep the current choice, or type selection number: 1 There are 2 choices for the alternative cuda (providing /usr/local/cuda). usr/local/cuda-11.5 - priority 115 /usr/local/cuda-11.6 - priority 116Īnd To make the CUDA-11.5 is the active version # update-alternatives -config cuda Link currently points to /usr/local/cuda-11.6 Link best version is /usr/local/cuda-11.6 So we will install the previous CUDA Version # apt install cuda-11-5Īnd to display all CUDA Alternatives # update-alternatives -display cuda To confirm whether CUDA is working, reboot the system, then run the following command:īut what if we need to install CUDA 11.5, for example, need to run the CUDA 11.5 based cuDNN library libcudnn8 8.3.3.40-1+cuda11.5 If (you found your GPU Card is CUDA Supported) ' > ~/.bashrc Check your GPU Graphic Card for the CUDA Enabled feature, the compute capability listed here for Nvidia CUDA GPUs Graphic Cards.
0 kommentar(er)
