an MLSidecar instance (managed or remote); defaults to the
shared, process-lifetime instance from MLSidecarProvider
Score 1..N rows. The implementation owns sidecar lifecycle (start/connect); the processor only calls this method.
the predict request (artifact + frozen preprocessing + schema + rows)
ISidecarPredictor backed by the self-managing MLSidecar client. Lazily starts the sidecar on first use (managed or remote mode is decided by
MLSidecarfrom options/env), then warm-caches the connection for the run. Defaults to the process-shared MLSidecarProvider instance rather than constructing a privateMLSidecar— see that class's doc comment for why a per-instance sidecar leaked a Python child process per action run. Tests (and callers that genuinely need an isolated sidecar) can still inject their own via the constructor parameter.