Training a logistic regression model with scikit-learn

1. Since scikit-learn implements a highly optimized version of logistic regression that also supports multiclass settings off-the-shelf, we will skip the implementation and use the sklearn.linear_model.LogisticRegression class as well as the familiar fit method to train the model:

from sklearn.linear_model import LogisticRegression

lr = LogisticRegression(C=1000.0, random_state=0)
lr.fit(X_train_std, y_train)

2. Showing

plot_decision_regions(X_combined_std, y_combined, classifier=lr, test_idx=range(105, 150))
plt.xlabel('petal length [standardized]')
plt.ylabel('petal width [standardized]')
plt.legend(loc='upper left')
plt.show()

Reference: 《Python Machine Learning》