search-workflow

Official

Automate operator search for optimized kernels.

Authormindspore-ai
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Automates generation of high-performance operator implementations by running adaptive_search or evolve workflows. It targets AKG-based workflows and supports background, silent-mode execution with progress polling for monitoring tasks.

Core Features & Use Cases

  • Run adaptive_search or evolve workflows to generate high-performance operator implementations.
  • Execute in the background with silent mode and progress polling for monitoring.
  • Collect results and produce generated_code.py and summary.json for quick integration.
  • Use Case: tuning a new operator by exploring multiple design variants and selecting the best performing impl.

Quick Start

Use the search-workflow with your task file and required framework/backend/arch/dsl, specify an output path, and let it orchestrate the workflow and generate code.

Dependency Matrix

Required Modules

None required

Components

scripts

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: search-workflow
Download link: https://github.com/mindspore-ai/akg/archive/main.zip#search-workflow

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.