AGENT SKILL

installation-and-configuration

Routes tf-faster-rcnn install, build, CUDA, TensorFlow, Cython NMS, and config-preset questions, including safe environment checks and common failure recovery.

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

01 前言 Front Matter

name
installation-and-configuration
license
MIT
metadata
{ "disco-role": "operating" }
description
Routes tf-faster-rcnn install, build, CUDA, TensorFlow, Cython NMS, and config-preset questions, including safe environment checks and common failure recovery.
disable-model-invocation
true

02 资源文件 4

03 SKILL.md 正文

Installation and Configuration

Use this sub-skill when a tf-faster-rcnn checkout will not install, build, or import cleanly, or when you need to verify CUDA/TensorFlow/protobuf compatibility, Cython NMS extension readiness, or config preset behavior.

Covered tasks

  • set up a legacy Python environment for source inspection or runtime prep
  • understand the repo's TensorFlow 1.x / protobuf compatibility window
  • build or diagnose lib/setup.py and lib/Makefile
  • choose between GPU NMS, CPU NMS, and pure Python inspection
  • read and override lib/model/config.py plus experiments/cfgs/*.yml
  • run a safe smoke check without downloading data or training

Route elsewhere

Read first

Safe check

Run the bundled inspector from any directory:

  • python scripts/check_environment.py --repo-root <tf-faster-rcnn-root>

The inspector only reads local files and imports. It does not download datasets, fetch checkpoints, or start training.

Remember

  • cfg.USE_GPU_NMS defaults to True.
  • model.nms_wrapper imports both compiled NMS modules eagerly, so cfg.USE_GPU_NMS=False does not by itself make a missing build safe.
  • If the inspector reports missing nvcc, CUDAHOME, or model.nms_wrapper, start with the install/build and troubleshooting references before trying demo or training routes.

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