VISSA - Basic example

A VISSA example showing the feature importance for a synthetic regression task.

The example uses a synthetic dataset with 10 standard normally distributed features. The target values only depend on two features: #0 and #5. If the VISSA method is tasked with selecting two features, it identifies the two important features as shown below.

import matplotlib.pyplot as plt
import numpy as np

from auswahl import VISSA

np.random.seed(1337)
X = np.random.randn(100, 10)
y = 5 * X[:, 0] - 2 * X[:, 5]

vissa = VISSA(n_features_to_select=2, n_submodels=100)
vissa.fit(X, y)

colors = np.full(X.shape[1], fill_value='C00')
colors[vissa.get_support()] = 'C01'

plt.bar(x=np.arange(X.shape[1]), height=abs(vissa.frequency_), color=colors)

plt.xlabel('Feature')
plt.ylabel('Selection Frequency')

plt.show()