configuration-and-data
Helps agents create, validate, render, and troubleshoot Ludwig configs, schemas, feature declarations, datasets, and preprocessing inputs.
- License 许可
- Apache-2.0
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- configuration-and-data
- license
- Apache 2.0
- metadata
- { "disco-role": "operating" }
- description
- Helps agents create, validate, render, and troubleshoot Ludwig configs, schemas, feature declarations, datasets, and preprocessing inputs.
- disable-model-invocation
- true
02 资源文件 5
- 参考文档references/configuration.md1.9 KB
- 参考文档references/data-formats.md1.3 KB
- 参考文档references/troubleshooting.md1.1 KB
- 脚本scripts/make_tiny_dataset.py1.5 KB
- 脚本scripts/validate_ludwig_config.py2.4 KB
03 SKILL.md 正文
Configuration and Data
Use this sub-skill when the task is about Ludwig YAML/config dictionaries, feature declarations, schema export, preprocessing, synthetic data, data formats, config generation, or dataset-column troubleshooting.
Start here
- Identify the target workflow: config authoring, config validation/rendering, data preparation, synthetic fixture creation, or LLM-backed config generation.
- Read configuration.md for config sections and feature/model-type choices.
- Read data-formats.md for dataset layouts and split-column guidance.
- For a local fixture, run:
python scripts/make_tiny_dataset.py --output-dir /tmp/ludwig-tiny
python scripts/validate_ludwig_config.py /tmp/ludwig-tiny/config.yaml
- If validation fails, use troubleshooting.md before moving to training.
Commands and APIs
- CLI:
ludwig generate_config,ludwig init_config,ludwig render_config,ludwig export_schema,ludwig preprocess,ludwig synthesize_dataset. - Python:
ludwig.config_generation.generate_config,ludwig.config_generation.get_ludwig_schema_context,ludwig.automl.init_config,LudwigModel.preprocess. - Config sections:
model_type,input_features,output_features,combiner,trainer,backend,preprocessing,hyperopt,adapter,quantization,prompt.
Route onward
- After config/data validation, route training to training-and-experiments.
- Route HPO/AutoML search-space work to automl-and-hyperopt.
- Route prediction-data fixture work to prediction-evaluation-and-inspection.
- Route serving payload construction to serving-export-and-deployment.
Safety notes
generate_configuses external LLM providers unless mocked or run only with--help; verify API keys before real calls.- Dataset-zoo and Kaggle/Hugging Face flows can use network and cache. Prefer local CSV/Parquet fixtures for smoke checks.
preprocesswrites processed data/metadata; choose an explicit scratch output path when applicable.
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).