AGENT SKILL

continuous-control

Guides minimalRL DDPG, continuous PPO, and SAC workflows for Pendulum-style continuous-action control.

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

01 前言 Front Matter

name
continuous-control
license
MIT
metadata
{ "disco-role": "operating" }
description
Guides minimalRL DDPG, continuous PPO, and SAC workflows for Pendulum-style continuous-action control.
disable-model-invocation
true

02 资源文件 4

03 SKILL.md 正文

Continuous-Control Workflows

Use this sub-skill when a task asks about minimalRL's continuous-action Pendulum algorithms: DDPG, PPO-Continuous, or SAC.

Route by task

Do not use this sub-skill for

Quick operating workflow

  1. Confirm the environment has continuous actions. The minimalRL continuous scripts assume Pendulum-like observations of length 3 and a one-dimensional action in [-2, 2].

  2. Choose the algorithm family:

    • DDPG for deterministic actor plus critic, replay, OU exploration, and soft target updates.
    • Continuous PPO for on-policy Gaussian policy with clipped objective.
    • SAC for stochastic tanh-Gaussian policy, twin critics, entropy temperature update, and replay.
  3. Read references/api-reference.md before changing tensor shapes or action scaling.

  4. Run the smoke helper:

    python sub-skills/continuous-control/scripts/smoke_continuous_control.py --algorithm all
    
  5. For action range, log_prob, state/action concatenation, replay warm-up, or target update failures, read references/troubleshooting.md.

Bundled runtime files

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