How is the GPU support for PyTorch on CentOS

To enable PyTorch GPU acceleration on CentOS system, you need to install CUDA, cuDNN and GPU versions of PyTorch. The following steps will guide you through the process:

CUDA and cuDNN installation

1. Confirm CUDA version compatibility: Use the nvidia-smi command to view the CUDA version supported by your NVIDIA graphics card. For example, your MX450 graphics card may support CUDA version 11.1 or later.

2. Download and install CUDA Toolkit: Visit the official website of NVIDIA CUDA Toolkit and download and install the corresponding version according to the highest CUDA version supported by your graphics card.

3. Install the cuDNN library: Go to the NVIDIA cuDNN official website , download the cuDNN library compatible with your CUDA version, and follow the official guide to complete the installation.

PyTorch GPU version installation

1. Install PyTorch GPU version with pip: Use the pip command to install a compatible PyTorch GPU version according to your CUDA version. For example, for CUDA 11.1, you can refer to the commands provided by the official website of PyTorch to install it, and make sure to select a version that matches your CUDA and cuDNN versions.

Verify GPU support

1. Check CUDA availability: Use the following Python code to verify that CUDA is installed correctly and available:

    import torch
   print(torch.cuda.is_available()) # Output True means CUDA is available

2. Get GPU information: Run the following code to get the number of GPUs, the currently used GPU device number, and the GPU name:

    print(torch.cuda.device_count()) # Output the number of GPUs print(torch.cuda.current_device()) # Output the current GPU device number print(torch.cuda.get_device_name(0)) # Output the first GPU device name

If the above steps are completed successfully, you can use PyTorch's GPU acceleration feature on your CentOS system. If you have any questions, please refer to the official PyTorch documentation or relevant community forum for help.

The above is the detailed content of How is the GPU support for PyTorch on CentOS. For more information, please follow other related articles on the PHP Chinese website!