skills.rest
mindspore-ai
mindspore-aiOFFICIAL
Explore all skills by this author →
WebsiteGitHub

search-workflow

Run adaptive_search or evolve workflows to generate operator implementations.

258|48|Updated Jun 22, 2020
Search Tags:#automation#workflow#monitoring#evolve#akg#adaptive_search#operator-generation
One-click install
npx skills add https://github.com/mindspore-ai/akg --skill search-workflow
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: search-workflow
Source: https://github.com/mindspore-ai/akg/tree/main/akg_agents/workspace/.opencode/skills/search-workflow
Command: npx skills add https://github.com/mindspore-ai/akg --skill search-workflow

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

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.

Frequently Asked Questions about search-workflow

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

FAQPage Schema
How do I automate AKG operator generation for high-performance kernels?▼

You can automate AKG operator generation by running adaptive_search or evolve workflows. This process executes in the background with silent mode and progress polling, outputting generated_code.py and summary.json for integration.

How do I monitor the progress of a background operator search workflow?▼

You can monitor background operator search workflows using silent mode execution with progress polling. This allows you to track the status of adaptive_search or evolve tasks running for high-performance kernel implementation generation.

What inputs do I need to run an adaptive_search workflow for operator tuning?▼

Running an adaptive_search workflow requires a task file path, framework, backend, architecture, DSL, and an output path. Optional device and workflow-specific parameters can also be specified to tune the operator implementation.

Can I use evolve workflows to explore multiple design variants for operator implementation?▼

Yes, you can use evolve workflows to explore multiple design variants for operator implementation. This approach automates the selection of the best performing implementation, generating a summary.json and generated_code.py for quick integration.

Does the operator search workflow support background execution on specific devices?▼

The operator search workflow supports background execution with progress polling. Device specifications are optional parameters, allowing the adaptive_search or evolve workflows to target specific hardware for high-performance kernel generation.

Why use adaptive_search for generating high-performance operator implementations?▼

Use adaptive_search to automate the exploration of design variants and generate optimized operator implementations. It orchestrates AKG-based workflows in the background, collecting results into generated_code.py and summary.json for immediate integration.

Related Skills

Explore More Software Engineering Skills→
Just-itJust-it
33

latency-optimizer

Analyze Triton kernels for Ascend NPU and generate optimized versions.

Community
Advanced
plurigridplurigrid
60

mlx-jax-splitmix

Generate deterministic colors across CPU and GPU with SplitMix64 PRNG.

Official
Advanced
openJiuwen-aiopenJiuwen-ai
2.5k

akg-agents

Automate AKG kernelgen workflows with branch enforcement and environment setup.

Official
Intermediate
mindspore-aimindspore-ai
258

cpu-basics

Explain foundational CPU kernel concepts and patterns for AKG operators.

Official
Basic