quark-install

Installs AMD Quark and verifies its Python package environment after PyTorch setup.

154|27|Updated Jun 19, 2024
One-click install
npx skills add https://github.com/amd/Quark --skill quark-install
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: quark-install
Source: https://github.com/amd/Quark/tree/main/.claude/skills-impl/l1-atomic/shared/quark-install
Command: npx skills add https://github.com/amd/Quark --skill quark-install

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the friction of setting up AMD Quark by guiding installation, dependency resolution, and post-install verification after PyTorch is already in place.

Core Features & Use Cases

  • Package Installation: Install the amd-quark package from PyPI, a pre-built wheel, or source depending on the environment.
  • Dependency Setup: Handle optional ONNX Runtime support, LLM PTQ extras, and compiler prerequisites needed for kernel compilation.
  • Verification and Recovery: Confirm the import, kernel build, and ONNX custom-op setup, while catching common failures such as missing compilers or a broken Python environment.

Quick Start

Ask to install Quark in your current environment and verify the import, kernel compilation, and ONNX custom ops after confirming PyTorch is already installed.

Frequently Asked Questions about quark-install

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I install AMD Quark after PyTorch is already configured?

To install AMD Quark, you can use the `amd-quark` package from PyPI, a pre-built wheel, or build from source depending on your specific environment and existing PyTorch setup.

Do I need a specific compiler for AMD Quark kernel compilation?

Yes, AMD Quark requires specific compiler prerequisites for kernel compilation. The installation process includes dependency setup to resolve missing compiler issues and verify compiled extensions.

How do I verify my AMD Quark installation and ONNX Runtime setup?

You can verify your AMD Quark installation by confirming successful package imports, checking kernel builds, and validating the ONNX custom-op setup to catch common post-install failures.

Does AMD Quark support installation across CPU, CUDA, and ROCm workflows?

Yes, AMD Quark applies to setup and post-install validation across CPU, CUDA, and ROCm workflows, requiring correct Python version handling and optional ONNX Runtime selection for each environment.

What should I do if my AMD Quark import fails due to missing dependencies?

If your AMD Quark import fails, the dependency resolution process addresses missing compilers, broken Python environments, and LLM PTQ extras, while verifying module imports and compiled extensions.