AGENT SKILL

deepaugment_transform_proxy

Create deterministic semantic-preserving texture and style perturbations for bounded DeepAugment-style recovery tests.

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

01 前言 Front Matter

name
deepaugment_transform_proxy
description
Create deterministic semantic-preserving texture and style perturbations for bounded DeepAugment-style recovery tests.

02 资源文件 2

03 SKILL.md 正文

DeepAugment Transform Proxy

Use this skill when full DeepAugment image-to-image networks are unavailable but a bounded recovery must still exercise the paper mechanism: label-preserving changes to texture-like/local-statistic features. Do not use this skill as evidence of full ImageNet-scale DeepAugment training.

Inputs

  • Examples with numeric features and a label.
  • A deterministic seed and perturbation strength.

Outputs

  • Augmented examples with unchanged labels.
  • A transformation log naming the applied perturbation families.

Workflow

  1. Keep semantic coordinates stable enough for the class label to remain valid.
  2. Perturb nuisance or texture-like coordinates with contrast, noise, and mixture operations.
  3. Emit a log for every generated example.
  4. Verify transformed examples differ from clean inputs while preserving labels.

Validation

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

Limitations

The proxy operates on tiny numeric features, not real images or perturbed neural image translators. It is acceptable only for declared soft-mode 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