PyCaret 4.0 exception hierarchy.
Every error an engine operation raises should be (a subclass of) PyCaretError
so that a UI / agent caller can catch the engine's errors without also catching
unrelated upstream library errors.
PyCaret 4.0 exception hierarchy.
Every error an engine operation raises should be (a subclass of) PyCaretError
so that a UI / agent caller can catch the engine's errors without also catching
unrelated upstream library errors.
Root of the PyCaret engine exception hierarchy.
Raised when an experiment is configured with inconsistent parameters.
Inherits from ValueError as well so existing sklearn-style try/except
blocks continue to work.
Raised when an operation is invoked on an unfitted Experiment.
Inherits from RuntimeError to match sklearn's NotFittedError behaviour
(which is itself a ValueError — we stop short of that because it's
semantically closer to a runtime-state mismatch).
Raised when create_model(id) is called with an id not in the registry.
Raised when a metric name is referenced that isn't in the registry.