AGENT SKILL

rectified_ode_reflow

Simulate rectified-flow Euler transport and compute straightness, reflow, and convex transport diagnostics.

License 许可
Apache-2.0
最近更新
2026-09-03
Stars
★ 351
Forks
27

01 前言 Front Matter

name
rectified_ode_reflow
description
Simulate rectified-flow Euler transport and compute straightness, reflow, and convex transport diagnostics.

02 资源文件 2

03 SKILL.md 正文

Rectified ODE Reflow

Use this skill when a learned rectified-flow velocity field must be integrated or audited for straightness. Do not use it to fit the velocity field.

Inputs

  • Initial samples z0.
  • Affine velocity parameters compatible with rectified_velocity_regression.
  • Euler step count.

Outputs

  • Final transported samples.
  • Path length and direct endpoint distance.
  • Straightness ratio and squared transport cost.

Workflow

  1. Integrate z <- z + dt * v(z,t) using Euler steps.
  2. Accumulate segment lengths for each sample.
  3. Compare accumulated path length with direct endpoint distance.
  4. Report one-step and multi-step diagnostics for reduced recovery.

Validation

Run python tests/test_ode.py or the Distiller validator with tests enabled.

Limitations

The script checks the ODE/reflow mechanism on numeric vectors. Full image quality metrics require a separate heavy runtime that is outside this bounded reduced recovery.

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