configuration-actions
Configure Lux display, plotting, sampling, custom recommendation actions, and widget debugging.
- License 许可
- Apache-2.0
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- configuration-actions
- license
- Apache 2.0
- metadata
- { "disco-role": "operating" }
- description
- Configure Lux display, plotting, sampling, custom recommendation actions, and widget debugging.
- disable-model-invocation
- true
02 资源文件 4
- 参考文档references/configuration.md6.8 KB
- 参考文档references/custom-actions.md4.6 KB
- 参考文档references/troubleshooting.md4.8 KB
- 脚本scripts/config_action_smoke.py4.1 KB
03 SKILL.md 正文
configuration-actions
Use this sub-skill when the task is about Lux session configuration rather than building individual visualizations or connecting to SQL tables.
Route here
- Changing
lux.configoptions for display, plotting backend, chart sizing, sampling, heatmaps, ranking, top-k selection, fallback/debug behavior, or executor selection. - Registering, inspecting, or removing custom recommendation actions with
lux.config.register_actionandlux.config.remove_action. - Diagnosing Lux widget setup with
lux.debug_info()and notebook/lab extension commands. - Forcing cached recommendations to refresh after configuration or action-manager changes.
Route elsewhere
- For constructing
Clause,Vis,VisList, or exporting a selected chart to Altair, Matplotlib, Vega-Lite, or code, usevisualization-export. - For
LuxSQLTable, SQL connections, SQL table setup, PostgreSQL service checks, orset_SQL_connection, usesql-backend. - For dataframe intent syntax, default recommendation categories, and
df.intent, usepandas-intent-recommendations. - For semantic data type overrides, temporal/geographic handling, and index/groupby caveats, use
special-data-types.
Bundled references
references/configuration.mdgives thelux.configoption map, safe assignment patterns, refresh rules, and invalid-value behavior.references/custom-actions.mdgives the custom action contract, validator semantics, removal behavior, and a self-contained recipe.references/troubleshooting.mdmaps common widget, stale recommendation, invalid config, and custom action failures to fixes.scripts/config_action_smoke.pyis an offline smoke test that imports Lux, registers a validator-gated action on a tiny dataframe, and checkslux.debug_info(return_string=True).
Operating rules
- Import
luxbefore creating Pandas dataframes that should becomeLuxDataFrameobjects. - Prefer setting configuration before the first dataframe display. If recommendations have already been generated, call
df.expire_recs()before redisplaying the dataframe. - Keep examples offline and local. Do not depend on hosted datasets or the Lux source checkout.
- Treat SQL executor configuration as a handoff to
sql-backendunless the task only needs to switch back tolux.config.set_executor_type("Pandas").
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).