mostlyright.spine

mr.spine() — bring your own target frame into the leakage-guarded join.

spine(df, *, entity, decision_time, y, label_available_time=None) maps a user-built frame into the canonical SpineContract shape (station, decision_time, label_available_time, y_*) from an explicit column mapping: the researcher names which of their columns is the entity, which is the point-in-time decision_time cutoff, and which are the y targets.

The mapping is never guessed, which is what keeps point-in-time integrity. A frame missing the mapped decision_time (or entity, or a y column) raises ContractError built by the shared _errmsg helper, so a wrong column mapping — a label passed as decision_time, for instance — cannot slip through silently. The user’s entity column is renamed to the canonical station key that align() and SpineContract consume. When the user does not map a label_available_time it defaults to decision_time, which is conservative and non-leaking: the label is treated as known no earlier than the decision.

The label-axis helpers (align_byo_labels / _coerce_date_column) are reused here, so a Feast-shaped entity_df whose decision_time lives in a datetime index is handled the same way as a user-supplied label frame.

Functions

Function	Description
spine(df, *, entity, decision_time, y[, …])	Map a user-built frame into the canonical spine shape.

mostlyright.spine.spine(df, , entity, decision_time, y, label_available_time=None)

Map a user-built frame into the canonical spine shape.

• Parameters:
  • df (DataFrame) – The user-built target frame.
  • entity (str) – The column naming the entity (renamed to the canonical station key).
  • decision_time (str) – The column holding the per-row point-in-time cutoff. If it is absent as a column but lives in a datetime / date-named index, the index is lifted into a column first (reusing _coerce_date_column).
  • y (str | list[str] | tuple[str, ...]) – The target column name(s). A spine with no target is meaningless.
  • label_available_time (str | None) – Optional column naming when the label became known — its own publication time; defaults to decision_time when unmapped, which is conservative and non-leaking.
• Return type: DataFrame
• Returns: A SpineContract-valid frame (station, decision_time, label_available_time, *y).
• Raises: ContractError – a mapped column (entity / decision_time / a y) is absent from the frame; the mapping is never guessed.