gistfile1.txt

import numpy as np
import pandas as pd
from sklearn.linear_model import LinearRegression

# Create a predictor X
X = np.random.default_rng().normal(size=100)
# Create a Noise Vector e
e = np.random.default_rng().normal(size=100)

B_0 = 0.5
B_1 = -0.3
B_2 = 3
B_3 = 0.7

Y = B_0 + B_1 * X + B_2 * (X ** 2) + B_3 * (X ** 3) + e

X = X.reshape(-1, 1)

features = np.array([X**i for i in range(1, 11)])

best_rss_set = []
best_features_set = []

while features.any():
    best_rss = None
    best_idx = None
    best_set = None
    for idx, feature in enumerate(features):
        features_set = np.hstack((best_set, feature))
        model = LinearRegression().fit(features_set,Y)
        y_pred = model.predict(features_set)
        Rss = np.sum((Y - y_pred)**2)
        # Descobrir melhor RSS e melhor Índice
        if best_rss == None or Rss < best_rss:
            best_rss = Rss
            best_idx = idx
            best_set = features_set
    # Alimentar a best_rss_set e a best_features_set
    best_rss_set.append(best_rss)
    best_features_set.append(features_set)
    # Iterando o While
    features = np.delete(features, best_idx)


# Preciso guardar o RSS e de quais features saiu esse RSS