BalanceOutput

class dataeval.metrics.bias.BalanceOutput(balance: ndarray[Any, dtype[float64]], factors: ndarray[Any, dtype[float64]], classwise: ndarray[Any, dtype[float64]], factor_names: list[str], class_list: ndarray[Any, dtype[Any]])

Output class for balance() bias metric

balance

Estimate of mutual information between metadata factors and class label

Type:

NDArray[np.float64]

factors

Estimate of inter/intra-factor mutual information

Type:

NDArray[np.float64]

classwise

Estimate of mutual information between metadata factors and individual class labels

Type:

NDArray[np.float64]

factor_names

Names of each metadata factor

Type:

list[str]

class_list

Array of the class labels present in the dataset

Type:

NDArray

plot(row_labels: list[Any] | ndarray[Any, dtype[Any]] | None = None, col_labels: list[Any] | ndarray[Any, dtype[Any]] | None = None, plot_classwise: bool = False) → Figure

Plot a heatmap of balance information

Parameters:

• row_labels (ArrayLike or None, default None) – List/Array containing the labels for rows in the histogram

• col_labels (ArrayLike or None, default None) – List/Array containing the labels for columns in the histogram

• plot_classwise (bool, default False) – Whether to plot per-class balance instead of global balance