meridian.model.eda.eda_spec.AggregationConfig

A configuration for customizing variable aggregation functions.

meridian.model.eda.eda_spec.AggregationConfig(
    *,
    control_variables: AggregationMap = dataclasses.field(default_factory=dict),
    non_media_treatments: AggregationMap = dataclasses.field(default_factory=dict)
)

The aggregation function can be called in the form f(x, axis=axis, **kwargs) to return the result of reducing an np.ndarray over an integer valued axis. It's recommended to explicitly define the aggregation functions instead of using lambdas.

Attributes
control_variables	A dictionary mapping control variable names to aggregation functions. Defaults to np.sum if a variable is not specified.
non_media_treatments	A dictionary mapping non-media variable names to aggregation functions. Defaults to np.sum if a variable is not specified.

Methods

__eq__

__eq__(
    other
)

Return self==value.