Plotting
The plotting module is the rendering backend used by all GPClarity classes.
Each of the three functions is normally invoked via the class method (e.g.
profiler.plot(), tracker.plot_evolution(), influence.plot_influence()),
which set sensible defaults and forward to these functions. Call them directly
only when you need full control over figure/axes layout.
Note
Prefer calling profiler.plot(X_test), tracker.plot_evolution(), and
influence.plot_influence(X_train) on the class objects rather than
importing and calling these functions directly.
import matplotlib.pyplot as plt
from gpclarity.plotting import plot_uncertainty_profile
fig, ax = plt.subplots()
plot_uncertainty_profile(
model, X_test, X_train=X_train, y_train=y_train,
confidence_level=2.0, show_regions=True, ax=ax
)
plt.tight_layout()
plt.show()