fid_activation_statistics
Compute FID-style activation mean/covariance summaries for generated or real feature matrices.
- License 许可
- Apache-2.0
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- fid_activation_statistics
- description
- Compute FID-style activation mean/covariance summaries for generated or real feature matrices.
02 资源文件 2
03 SKILL.md 正文
Fid Activation Statistics
Use this skill when a recovery or implementation needs the fid_statistics module from the TTUR/FID paper without rereading the original repository. Do not use it to claim full GAN reproduction unless real image features, dataset statistics, and training evidence are available.
Inputs
- Numeric arrays or JSON files matching the module contract.
- For recovery orchestration, a module-plan target and runtime handoff.
Outputs
- Deterministic JSON-compatible statistics, distances, traces, or recovery checks.
- Diagnostics sufficient to audit reduced/proxy recovery.
Workflow
- Validate input shapes and finite numeric values.
- Execute the module-specific deterministic script in
scripts/. - Preserve diagnostics and command evidence when used in recovery.
- Treat synthetic activations or toy dynamics as proxy evidence only.
Validation
Run python tests/test_fid_activation_statistics.py from this skill directory.
Limitations
This skill is self-contained and does not require the original TTUR repository. It does not download Inception models or train full GANs.
04 同仓库更多 Skills
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.
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.
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.
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.
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.
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).