brainstorming
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).
Skills 둘러보기
개발 및 엔지니어링 · SKILL.md
sar_norm_adaptation_policy
sar_norm_adaptation_policyUse this skill when selecting trainable parameters for SAR-style test-time adaptation while freezing non-normalization model weights.
Independent GitHub snapshot · Automated classification · Original documentation governs
Skill 문서
GitHubname: sar_norm_adaptation_policy description: Use this skill when selecting trainable parameters for SAR-style test-time adaptation while freezing non-normalization model weights.
Norm Adaptation Policy
When To Use
Use this skill for SAR stable test-time adaptation work that needs: Configure a model or parameter set so test-time adaptation updates only normalization affine parameters. Do not use it as a generic ImageNet-C runner or as permission to read the original SAR repository during recovery.
Inputs
- model parameter metadata or small proxy parameter dict
Outputs
- trainable parameter selection and frozen parameter report
Workflow
- Read the caller-provided stream, logits, parameter metadata, or entropy trace.
- Apply the SAR-specific invariant documented in this skill: Stable wild TTA requires batch-agnostic normalization and restricted affine adaptation instead of full model updates.
- Produce structured numeric evidence for recovery logs.
- Keep proxy/reduced results explicitly labeled when the full ImageNet-C runtime is unavailable.
Validation
Run python scripts/norm_adaptation_policy.py when the script has a CLI, then run python -m pytest tests or the Distiller skill validator with --run-tests.
Limitations
This skill captures a reusable SAR mechanism. It does not contain pretrained weights, ImageNet-C data, or original repository code, and it should be combined with other SAR skills for end-to-end recovery.