Performs a round robin iteration of several iterables Adapted from https://docs.python.org/3/library/itertools.html#recipes ..code:: python

Replace NaN and Inf (there should not be any!) :param array: :return:

Start the task

Whether this task has been started

Stop a given task

Note ---- The keyword arguments caught in `dask.distributed.Client` need to be specified here so they don't get passe

Return sum of CPU time for all so far finished tasks.

Return sum of wall clock time for all so far finished tasks.

Parameters ---------- path: Path The path to check Returns ------- bool Whet

A context manager to catch warnings and send them to the logger If no logger is passed, warnings propogate as they normally would. Parameter

Rich Caruana's ensemble selection method with bagging.

Makes sure that init params is honored at the implementation level

This method centralizes the checks for resampling strategies Parameters ---------- X: (SUPPORTED_FEAT_TYPES)

Fast version of Rich Caruana's ensemble selection method.

find all expected files in self.dir_ fills self.found_files

Get the configuration of a classification pipeline given some dataset info Parameters ---------- datamanager: AbstractDataManager

Return the configuration space for the CASH problem. This method should be called by the method get_hyperparameter_search_space of a

This is inside an extra function for testing purpose

Get the configuration of a regression pipeline given some dataset info Parameters ---------- datamanager: AbstractDataManager Abs

In case of failure during ensemble building, this method returns the single best model found by AutoML. This is a rob

iterates over all found files (self.found_files) and calls the corresponding function to validate file

Rich Caruana's ensemble selection method.

Given a configuration in config, this procedure makes sure that the given implementation, which should be a autosklearn component, ho

Convert a path to a module as seen from here

Calculate only metafeatures for which a 1HotEncoded feature matrix is necessery.

Calculate the loss with a given metric Parameters ---------- solution: np.ndarray The solutions prediction: np.ndarray

Should return a dask client

Factory to make a Dask client and a function to close it them. Parameters ---------- id: str An id to associate with this das

Find the index of the next occurence where func is True. Parameters ---------- itr : Iterable[T] The iterable to search over

Select dummy ensembles given predictions of base models and targets. Parameters ---------- base_models_predictions: np.ndarra