18
правок
Изменения
→Пример кода для Scikit-learn
== Пример кода для Scikit-learn ==
'''import''' matplotlib.pyplot '''as''' plt
'''from''' sklearn '''import''' datasets, linear_model
<font color = green># generate dataset</font>
X, y = datasets.make_regression(n_samples=1_000, n_features=1, noise=8, shuffle=True)
<font color = green># test and train data sizes</font>
train_size = 700
test_size = 300
<font color = green># split the data into training/testing sets</font>
X_train = X[:-train_size]
X_test = X[-test_size:]
<font color = green># split the targets into training/testing sets</font>
y_train = y[:-train_size]
y_test = y[-test_size:]
<font color = green># create linear regression object</font>
regr = linear_model.LinearRegression()
<font color = green># train the model using the training sets</font>
regr.fit(X_train, y_train)
<font color = green># make predictions using the testing set</font>
y_pred = regr.predict(X_test)
<font color = green># plot outputs</font>
plt.scatter(X_test, y_test, color='red', s=5)
plt.plot(X_test, y_pred, color='blue', linewidth=2)
plt.xticks(())
plt.yticks(())
plt.show()
Возможный результат исполнения программы:
[[Файл: Linear_regression_example.png]]
