TargetAggregationConfig

class datarobotx.TargetAggregationConfig(max_unaggregated_class_values=None, min_class_support=None, aggregation_class_name=None, excluded_from_aggregation=None)

Target aggregation configuration.

Parameters that default to ‘None’ (or are omitted by the user) are overridden to server-side defaults at runtime. Consult the DataRobot REST API and GUI documentation for additional information on each parameter.

Parameters:

• max_unaggregated_class_values (int) – The maximum number of unique labels before aggregation kicks in. Should be at least len(excludedFromAggregation) + 1 for multiclass and at least len(excludedFromAggregation) for multilabel.

• min_class_support (int) – Minimum number of instances necessary for each target value in the dataset. All values with fewer instances than this value will be aggregated

• aggregation_class_name (str) – The name of the class that will be assigned to all rows with aggregated classes. Should not match any excluded_from_aggregation or we will have 2 classes with the same name and no way to distinguish between them. This option is only available formulticlass projects. By default ‘DR_RARE_TARGET_VALUES’ is used.

• excluded_from_aggregation (list of str) – List of target values that should be guaranteed to kept as is, regardless of other settings.

See also

DRConfig

Configuration object for DataRobot project and autopilot settings, also includes detailed examples of usage

Attributes:

aggregation_class_name	The name of the class that will be assigned to all rows with aggregated classes.
excluded_from_aggregation	List of target values that should be guaranteed to kept as is, regardless of other settings.
max_unaggregated_class_values	The maximum number of unique labels before aggregation kicks in.
min_class_support	Minimum number of instances necessary for each target value in the dataset.

Inherited methods:

keys()	rtype: Collection[str]
to_dict()	Return configuration as a dict.

property aggregation_class_name: str

The name of the class that will be assigned to all rows with aggregated classes. Should not match any excluded_from_aggregation or we will have 2 classes with the same name and no way to distinguish between them. This option is only available formulticlass projects. By default ‘DR_RARE_TARGET_VALUES’ is used.

Notes

aggregation_class_name : str

property excluded_from_aggregation: List[str]

List of target values that should be guaranteed to kept as is, regardless of other settings.

Notes

excluded_from_aggregation : list of str

property max_unaggregated_class_values: int

The maximum number of unique labels before aggregation kicks in. Should be at least len(excludedFromAggregation) + 1 for multiclass and at least len(excludedFromAggregation) for multilabel.

Notes

max_unaggregated_class_values : int

property min_class_support: int

Minimum number of instances necessary for each target value in the dataset. All values with fewer instances than this value will be aggregated.

Notes

min_class_support : int

to_dict()

Return configuration as a dict.

Return type:

Dict[str, Any]