ReliabilityModelPlot.trust_map

static ReliabilityModelPlot.trust_map(df_rel, figsize=(5.5, 5), title=None, ax=None)

Score vs. OOD-score scatter colored by reliable — the two trust axes at a glance.

A point high on the x-axis (confident) but above the in-domain boundary (out-of-domain) is the untrustworthy, extrapolated prediction.

Parameters:

• df_rel (pd.DataFrame) – Output of ReliabilityModel.predict() (needs score, ood_score, reliable).

• figsize (tuple, default=(5.5, 5)) – Figure size (used only when ax is None).

• title (str, optional) – Axes title.

• ax (matplotlib.axes.Axes, optional) – Axes to draw on; a new figure is created if None.

Returns:

• fig (matplotlib.figure.Figure) – The created (or parent) figure.

• ax (matplotlib.axes.Axes) – The axes drawn on.

Examples

ReliabilityModelPlot().trust_map() plots prediction score vs. OOD score, colored by the reliable flag — a confident point above the in-domain boundary is an untrustworthy extrapolation.

import aaanalysis as aa
import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import make_classification
aa.options["verbose"] = False
aa.plot_settings()
X, labels = make_classification(n_samples=160, n_features=10, n_informative=6, random_state=42)
X_train, labels_train = X[:120], labels[:120]
# include one clearly out-of-distribution sample so the trust status is visible
X_new = np.vstack([X[120:], (X_train[labels_train == 1][0] + 20.0)[None, :]])
rm = aa.ReliabilityModel(random_state=42).fit(X=X_train, labels=labels_train)
df_rel = rm.predict(X=X_new)
df_eval = rm.eval(X=X[120:], labels=labels[120:])

aa.ReliabilityModelPlot().trust_map(df_rel=df_rel, figsize=(5.5, 5),
                                    title="Trust map", ax=None)
plt.tight_layout()
plt.show()