model_evaluation.model_evaluation_plotting

model_evaluation.model_evaluation_plotting(
    pipeline,
    X_test,
    y_test,
    display_labels=None,
)

Compute classification metrics and generate confusion matrix visualizations.

Parameters

Name	Type	Description	Default
pipeline	sklearn estimator or sklearn.pipeline.Pipeline	Fitted model object that implements .predict() and .score() methods.	required
X_test	pandas.DataFrame	Test feature matrix. Must not contain NaN values.	required
y_test	array-like of shape (n_samples,)	True labels for test data. Can be pandas Series, numpy array, or list. Must not contain NaN values.	required
display_labels	array-like of shape (n_classes,)	Target class labels for confusion matrix display. If None, uses numeric labels.	None

Returns

Name	Type	Description
metrics	dict	Dictionary containing: - ‘accuracy’ : float - Classification accuracy on test data - ‘f2’ : float - F2 score (beta=2) - ‘y_pred’ : numpy.ndarray - Predicted labels
cm_table	pandas.DataFrame	Confusion matrix as a crosstab (rows=true labels, columns=predicted labels).
cm_display	sklearn.metrics.ConfusionMatrixDisplay	Confusion matrix display object for visualization.

Raises

Name	Type	Description
	TypeError	If pipeline is not fitted, or if input types are invalid.
	ValueError	If X_test and y_test have different lengths, or contain NaN values.

Examples

>>> from sklearn.datasets import load_iris
>>> from sklearn.model_selection import train_test_split
>>> from sklearn.svm import SVC
>>> import pandas as pd
>>> 
>>> X, y = load_iris(return_X_y=True)
>>> X = pd.DataFrame(X)
>>> y = pd.Series(y)
>>> X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
>>> 
>>> model = SVC().fit(X_train, y_train)
>>> metrics, cm_table, cm_display = model_evaluation_plotting(
...     model, X_test, y_test, 
...     display_labels=['setosa', 'versicolor', 'virginica']
... )
>>> 
>>> print(metrics['accuracy'])
0.9667
>>> print(cm_table)
Predicted    0   1   2
Actual               
0           10   0   0
1            0   9   1
2            0   0  10
>>> cm_display.plot()  # Shows confusion matrix visualization

Notes

• The model must be fitted before calling this function
• F2 score emphasizes recall over precision (beta=2)
• For binary classification, uses pos_label=“Y”