arista.processing.orchestrator

Batch driver that fills stimulus_responses and adaptation_fits.

Dispatches by stimulus protocol family:

  • thermal_step with a target_sequence → compute_stimulus_responses()

  • thermal_adapt (HotAdapt / ColdAdapt) → fit_adaptation()

  • mechanical → no aggregates (Bending)

  • thermal_adapt without target_sequence → adaptation only

  • thermal_step without target_sequence → none (Laurin pilots

    step / step_neg / long / long_neg lack a defined sequence yet)

Idempotent: re-running over a DB that already has aggregates skips those recordings unless reprocess=True is passed (in which case the existing rows for the recording are deleted and recomputed).

Functions

process_all(conn, *[, progress, reprocess])

Compute aggregates for every recording missing them.

Classes

ProcessingStats([inserted_responses, ...])

Tally for one process_all run.
class arista.processing.orchestrator.ProcessingStats(inserted_responses=0, inserted_adaptations=0, skipped_already_done=0, failed_adaptations=0)[source]

Bases: object

Tally for one process_all run.

Parameters:

  • inserted_responses (int)

  • inserted_adaptations (int)

  • skipped_already_done (int)

  • failed_adaptations (int)

as_dict()[source]
Return type:

dict[str, int]

failed_adaptations: int = 0
inserted_adaptations: int = 0
inserted_responses: int = 0
skipped_already_done: int = 0
arista.processing.orchestrator.process_all(conn, *, progress=None, reprocess=False)[source]

Compute aggregates for every recording missing them.

Parameters:

  • conn (Connection) – Open SQLite connection to a populated arista.db.

  • progress (bool | None) – tqdm visibility (None = auto-detect TTY, True = force on, False = force off).

  • reprocess (bool) – If True drop existing aggregates before recomputing. Useful after fixing the response formula.

Returns:

ProcessingStats with counts of inserted / already-done / failed-adaptation recordings.

Return type:

ProcessingStats