AGENT SKILL

genetic-algorithms

Use scikit-opt GA-family APIs for Gray-coded, elitist, and real-coded optimization.

License 许可
MIT
最近更新
2026-09-03
Stars
★ 351
Forks
27

01 前言 Front Matter

name
genetic-algorithms
license
MIT
metadata
{ "disco-role": "operating" }
description
Use scikit-opt GA-family APIs for Gray-coded, elitist, and real-coded optimization.
disable-model-invocation
true

02 资源文件 5

03 SKILL.md 正文

genetic-algorithms

Use this sub-skill for bounded optimization with GA, EGA, and RCGA, including integer or mixed-precision search, continuation with repeated run(max_iter) calls, history extraction, and safe register()-based operator swaps.

Read these bundled files

  • references/api-reference.md — exact constructor signatures, parameter meanings, result/history fields, and the GA_TSP handoff note. Read this when you need precise argument names or return values.
  • references/workflows.md — which GA-family class to choose, how to continue a run, how to set integer or mixed precision, and when to route problem workflows elsewhere. Read this before planning a solution.
  • references/custom-operators.md — SkoBase.register, operator signatures, built-in ranking/selection/crossover/mutation helpers, and a safe custom selection pattern. Read this when replacing defaults.
  • references/troubleshooting.md — fixes for even-population assertions, objective shape issues, bound/precision mismatches, penalty surprises, custom operator errors, and stochastic convergence. Read this when a run fails or behaves oddly.
  • scripts/smoke_genetic_algorithms.py — tiny deterministic import-and-run smoke for GA, EGA, RCGA, and register() behavior. Run this after changes or before handoff.

Route elsewhere when needed

  • Route DE, PSO, SA, and AFSA continuous optimizers to continuous-optimizers.
  • Route full TSP or routing workflows to routing-and-combinatorial; treat GA_TSP only as a handoff marker here.
  • Route run-mode, vectorization, cache, multiprocessing, and GA.to(device) details to objective-functions-and-speedups.

This sub-skill stays small on purpose: the bundled references carry the exact API and workflow detail, while the runtime tree remains self-contained.

04 同仓库更多 Skills

VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-triton-opt

Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block/grid/warp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-solve

END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-skill

PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under ./pass_dir/, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-pattern-fusion

Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()/replacement_args()/replacement_func() files, when a pass "failed to match", or when deciding how to fuse multiple kernels/ops.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-orchestrate

Round/batch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep/revert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-feedback

Fast, deterministic feedback for PassNet work: pre-flight pattern/match verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).

★ 351