Create Polynomial Regression.py

Courses/Machine Learning with Python, A Practical Introduction/(Module 2) Regression/Polynomial Regression.py

+# This code is an exercise done at module 2 as part of the course (ML0101EN) Machine Learning with Python_A Practical Introduction by IBM.
+# The intention of the exercise is to teach how to transform the features to polinomial features and solve a Linear Regression problem with polinomial features.
+
+import matplotlib.pyplot as plt
+import pandas as pd
+import numpy as np
+from sklearn import linear_model
+from sklearn.preprocessing import PolynomialFeatures
+from sklearn.metrics import r2_score
+
+url = 'https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBMDeveloperSkillsNetwork-ML0101EN-SkillsNetwork/labs/Module%202/data/FuelConsumptionCo2.csv'
+df = pd.read_csv(url)
+new_df = df[['ENGINESIZE', 'CYLINDERS', 'FUELCONSUMPTION_COMB', 'CO2EMISSIONS']]
+
+msk = np.random.rand(len(new_df)) < 0.8
+train = new_df[msk]
+test = new_df[~msk]
+
+
+poly = PolynomialFeatures(degree=2)
+regr = linear_model.LinearRegression()
+train_x = np.asanyarray(train[['ENGINESIZE']])
+train_x_poly = poly.fit_transform(train_x)
+test_x = np.asanyarray(test[['ENGINESIZE']])
+test_x_poly = poly.fit_transform(test_x)
+train_y = np.asanyarray(train[['CO2EMISSIONS']])
+test_y = np.asanyarray(test[['CO2EMISSIONS']])
+regr.fit(train_x_poly, train_y)
+print("Coefficients: ", regr.coef_)
+print("Intercept: ", regr.intercept_)
+test_y_predict = regr.predict(test_x_poly)
+print(f"Mean absolut error: {np.mean(np.absolute(test_y_predict - test_y)) :.2f}")
+print(f"Residual sum of squares: {np.mean((test_y_predict - test_y)**2) :.2f}")
+print(f"R2-score: {r2_score(test_y_predict, test_y) :.2f}")
+
+poly = PolynomialFeatures(degree=3)
+regr = linear_model.LinearRegression()
+train_x = np.asanyarray(train[['ENGINESIZE']])
+train_x_poly = poly.fit_transform(train_x)
+test_x = np.asanyarray(test[['ENGINESIZE']])
+test_x_poly = poly.fit_transform(test_x)
+train_y = np.asanyarray(train[['CO2EMISSIONS']])
+test_y = np.asanyarray(test[['CO2EMISSIONS']])
+regr.fit(train_x_poly, train_y)
+print("Coeficients: ", regr.coef_)
+print("Intercept: ", regr.intercept_)
+test_y_predict = regr.predict(test_x_poly)
+print(f"Mean Absolute Error: {np.mean(np.absolute(test_y_predict - test_y)):.2f}")
+print(f'Residual sum of squares: {np.mean((test_y_predict - test_y)**2):.2f}')
+print(f"R2-score: {r2_score(test_y_predict, test_y):.2f}")