---

name: panel-learning description: "Use sktime for time series classification, regression, clustering, and panel estimator selection workflows." disable-model-invocation: true metadata: disco-role: operating license: BSD 3-Clause

Panel Learning

Use this sub-skill when the task is supervised or unsupervised learning over a collection of time series instances: classification, regression, clustering, panel model selection, and panel estimator troubleshooting.

Route here

• Time series classification with one class/category per series.
• Time series regression with one numeric target per series.
• Time series clustering with fit_predict or cluster labels for panel instances.
• Panel estimator selection by capability tags, optional dependency availability, and input shape constraints.

Route away

• Raw mtype conversion, file formats, and dataset I/O: data-interfaces.
• Feature extractor or transformer internals: transformations-pipelines.
• Forecasting or detection tasks: their dedicated sub-skills.

Operating protocol

1. Identify whether y is categorical, numeric, or absent.
2. Confirm X is panel-shaped: instances first for numpy3D, or a two-level (instance, time) index for pd-multiindex.
3. Check len(y) == n_instances for supervised tasks.
4. Inspect tags such as capability:multivariate, capability:unequal_length, capability:missing_values, and capability:predict_proba.
5. Start with dummy/core baselines and small estimator settings; then move to optional estimators only after dependency checks.

References and helper

• API reference for panel shapes, estimator families, signatures, and tags.
• Workflows for classification, regression, clustering, tag search, and model selection.
• Troubleshooting for shape, label, soft dependency, probability, scoring, and clustering failures.
• Run scripts/panel_learning_smoke.py for a no-network panel learning smoke.