Tests a non-seeded random set of configurations with sparse data Expects ------- * All configurations should fit, predict and

Fixtures -------- mock_logger: PickleableClientLogger A mock logger to use make_sklearn_dataset : Callable[..., Tuple[np.nda

Parameters ---------- automl : AutoML The fitted automl object to test Expects ------- * The ensemble should not be

Makes sure that when using a cv strategy, we are able to fit a regressor

Expects ------- * TODO

Test that seeded SimpleClassificaitonPipeline returns good results on iris Expects ------- * The performance of configuration

Test that the SimpleClassificationPipeline default config for random forest with no preprocessing can be iteratively fit on iris. Exp

Test that SimpleClassificationPipeline default config returns good results on a multilabel version of iris. Expects -------

Makes sure that imputation works for both numerical and categorical data. This also has to be guaranteed for numpy and pandas like objects.

Expects ------- * Should delete runs so they can not be found again by the ensemble builder

Expects ------- * Should

Expects ------- * When running the builder, it's previous ensemble should not have it's runs deleted until a new ensemble is built.