describe_platform

Scope: read · Signature: describe_platform(kind: {presets|templates|algorithms|metrics|metric_catalog|foundation_models|foundation_engines|transformers|domains|config_schema|plan_schema|forecast_models|detectors|clustering_algorithms|explain_presets}, name: str | None = None, task: {binary_classification|multiclass_classification|regression} | None = None, context: {supervised|forecast|clustering|anomaly|recommendation|ranking|association} | None = None) -> Any

Use when planning: the config vocabulary ParleML offers, never invented.

kind ∈ {presets, templates, algorithms, metrics, metric_catalog, foundation_models, foundation_engines, transformers, domains, config_schema, plan_schema, forecast_models, detectors, clustering_algorithms, explain_presets}. plan_schema name ∈ {supervised, validation, forecast, association, cluster, anomaly, recommendation, ltr}: that plan block's JSON Schema and every setting's dotted path — offer and plan only what it lists. domains: per column domain, the features one column makes, what a companion column adds, the settings and whether this server has the data (IP geo) — read it before hand-building such features. config_schema: the supervised config's sections (name → one section). name → one entry's detail (templates, detectors, domains). task (binary_classification | multiclass_classification | regression) filters algorithms and metrics. context (supervised | forecast | clustering | anomaly | recommendation | ranking | association) is required for metric_catalog.

Powered by: ml4 config describe (ml4/src/parleml_ml4/config/describe.py) and the template catalogue.