Setting up an AI/ML Workstation with Intel GPU Max 1550 (Ponte Vecchio)

This guide will walk you through the process of setting up an AI/ML workstation using the Intel GPU Max 1550 (Ponte Vecchio). We'll cover the installation of Conda, necessary drivers, and key AI/ML frameworks optimized for Intel hardware.

1. System Requirements for this Guide

• A system with Intel GPU Max 1550 (Ponte Vecchio)
• Ubuntu 20.04 or later (recommended)
• Latest Intel GPU Drivers
• OneAPI: Link
• Sudo Permissions
• Optional: Intel AI Tools

2. Install Conda

1. Check if wget is installed:

   which wget

2. If wget is not installed, install it:

   sudo apt-get update
   sudo apt-get install wget

3. Download the Miniconda installer:

   wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh

4. Make the installer executable:

   chmod +x Miniconda3-latest-Linux-x86_64.sh

5. Run the installer:

   ./Miniconda3-latest-Linux-x86_64.sh

6. Follow the prompts in the installer.

7. After installation, restart your terminal or source your .bashrc file:

   source ~/.bashrc

3. Set Up Conda Environment

1. Create and activate a new conda environment:

   conda create -n intel_ml python=3.10
   conda activate intel_ml

2. Install AI/ML packages optimized for Intel hardware:

   conda install -c https://software.repos.intel.com/python/conda -c conda-forge --override-channels intel-extension-for-tensorflow=2.15=*cpu* intel-extension-for-pytorch=2.3.100 oneccl_bind_pt=2.3.0 torchvision=0.18.1 torchaudio=2.3.1 deepspeed=0.14.2

4. Verify Installation

To verify that the AI tools are properly installed, use the following commands:

• Intel® Extension for PyTorch* (GPU):

  python -c "import torch; import intel_extension_for_pytorch as ipex; print(torch.__version__); print(ipex.__version__); [print(f'[{i}]: {torch.xpu.get_device_properties(i)}') for i in range(torch.xpu.device_count())];"

• Intel® Extension for TensorFlow* (GPU):

  python -c "from tensorflow.python.client import device_lib; print(device_lib.list_local_devices())"

6. Optional: Install Additional AI/ML Tools

• Intel® Optimization for XGBoost*:

  conda install -c conda-forge xgboost

• Intel® Extension for Scikit-learn*:

  conda install scikit-learn-intelex

• Modin*:

  conda install -c conda-forge modin

• Intel® Neural Compressor:

  pip install neural-compressor

• IPEX-LLM: Link to IPEX-LLM, Intel's suite of supported software for running LLMs

7. Start Developing

You're now ready to start developing AI/ML applications using your Intel GPU Max 1550. Remember to activate your conda environment before starting your work:

conda activate intel_ml

Happy coding!