trustyai.explainers.CounterfactualResult

class trustyai.explainers.CounterfactualResult(result: CounterfactualResult)

Wraps Counterfactual results. This object is returned by the CounterfactualExplainer, and provides a variety of methods to visualize and interact with the results of the counterfactual explanation.

__init__(result: CounterfactualResult) None

Constructor method. This is called internally, and shouldn’t ever need to be used manually.

Methods

__init__(result)

Constructor method.

as_dataframe()

Return the counterfactual result as a dataframe

as_html()

Return the counterfactual result as a Pandas Styler object.

plot([block, call_show])

Plot the counterfactual result.

Attributes

proposed_features_array

Return the proposed feature values found from the counterfactual explanation as a Numpy array.

proposed_features_dataframe

Return the proposed feature values found from the counterfactual explanation as a Pandas DataFrame.

as_dataframe() DataFrame

Return the counterfactual result as a dataframe

Returns:
pandas.DataFrame

DataFrame containing the results of the counterfactual explanation, containing the following columns:

  • Features: The names of each input feature.

  • Proposed: The found values of the features.

  • Original: The original feature values.

  • Constrained: Whether this feature was constrained (held fixed) during the search.

  • Difference: The difference between the proposed and original values.

as_html() Styler

Return the counterfactual result as a Pandas Styler object.

Returns:
pandas.Styler

Styler containing the results of the counterfactual explanation, in the same schema as in as_dataframe(). Currently, no default styles are applied in this particular function, making it equivalent to self.as_dataframe().style.

plot(block=True, call_show=True) None

Plot the counterfactual result.

property proposed_features_array

Return the proposed feature values found from the counterfactual explanation as a Numpy array.

property proposed_features_dataframe

Return the proposed feature values found from the counterfactual explanation as a Pandas DataFrame.