Example – machine learning

Linear Regression

Load libraries

import matplotlib.pyplot as plt
import numpy as np
from sklearn import datasets, linear_model
from sklearn.metrics import mean_squared_error, r2_score

Load data

# Load the diabetes dataset
diabetes_X, diabetes_y = datasets.load_diabetes(return_X_y=True)

# Use only one feature
diabetes_X = diabetes_X[:, np.newaxis, 2]

Divide datasets into training and testing sets

# Split the data into training/testing sets
diabetes_X_train = diabetes_X[:-20]
diabetes_X_test = diabetes_X[-20:]

# Split the targets into training/testing sets
diabetes_y_train = diabetes_y[:-20]
diabetes_y_test = diabetes_y[-20:]

Generate a model

# Generate a model 
regr = linear_model.LinearRegression()

Machine learning (training): seek for the optimal parameters

# Train the model using the training sets
regr.fit(diabetes_X_train, diabetes_y_train)

Make predictions

# Make predictions using the testing set
diabetes_y_pred = regr.predict(diabetes_X_test)

Analyze models

# The coefficients
print("Coefficients: \n", regr.coef_)

# The mean squared error
print("Mean squared error: %.2f" % mean_squared_error(diabetes_y_test, diabetes_y_pred))

# The coefficient of determination: 1 is perfect prediction
print("Coefficient of determination: %.2f" % r2_score(diabetes_y_test, diabetes_y_pred))

Visualize the results

# Plot outputs
plt.scatter(diabetes_X_test, diabetes_y_test, color="black")
plt.plot(diabetes_X_test, diabetes_y_pred, color="blue", linewidth=3)
plt.xticks(())
plt.yticks(())
plt.show()