Logistic Regression: Python

# import libraries
from sklearn.linear_model import LogisticRegression
import mglearn
import matplotlib.pyplot as plt

from sklearn.model_selection import train_test_split

# c is strengh of regularization

# loading data and checking performance

from sklearn.datasets import load_breast_cancer

cancer = load_breast_cancer()

X_train, X_test, y_train, y_test = train_test_split(

cancer.data, cancer.target, stratify=cancer.target, random_state=42)

logisticregression = LogisticRegression().fit(X_train, y_train)

print("training set score: %f" % logisticregression.score(X_train, y_train))

print("test set score: %f" % logisticregression.score(X_test, y_test))

training set score: 0.943662 test set score: 0.958042

logisticregression100 = LogisticRegression(C=100).fit(X_train, y_train) print("training set score: %f" % logisticregression100.score(X_train, y_train)) print("test set score: %f" % logisticregression100.score(X_test, y_test))

training set score: 0.957746 test set score: 0.951049

logisticregression001 = LogisticRegression(C=0.01).fit(X_train, y_train) print("training set score: %f" % logisticregression001.score(X_train, y_train)) print("test set score: %f" % logisticregression001.score(X_test, y_test))

training set score: 0.934272 test set score: 0.930070

for C in [0.001, 1, 100]:

    lr_l1 = LogisticRegression(C=C, penalty="l2").fit(X_train, y_train)

    print("training accuracy of L1 logreg with C=%f: %f"% (C, lr_l1.score(X_train, y_train)))

    print("test accuracy of L1 logreg with C=%f: %f" % (C, lr_l1.score(X_test, y_test)))

training accuracy of L1 logreg with C=0.001000: 0.953052 test accuracy of L1 logreg with C=0.001000: 0.944056 training accuracy of L1 logreg with C=1.000000: 0.943662 test accuracy of L1 logreg with C=1.000000: 0.958042 training accuracy of L1 logreg with C=100.000000: 0.957746 test accuracy of L1 logreg with C=100.000000: 0.951049