Logistic Regression 로지스틱 회귀 with python

sklearn.linear_model.LogisticRegression — scikit-learn 1.0.1 documentation

sklearn.linear_model.LogisticRegression

 

Training the Logistic Regression model on the Training set

from sklearn.linear_model import LogisticRegression
classifier = LogisticRegression(random_state=0)//인스턴스생성 
classifier.fit(X_trains, y_tains)//훈련시키는 작업

 

Predicting a new result

//한 고객의 결과를 내는 것 
predict method 사용
classifier.predict(sc.transform([[30,87000]]))//2차원배열에서 30살의 87000달러를 받는 사람의 구매여부가 궁금

result : [0]

Predicting the Test set results[템플릿]

y_pred= classifier.predict(X_test)
print(np.concatenate((y_prep.reshape(len(y_pred),1),y_test.reshape(len(y_test),1)),1))

왼쪽 열은 테스트세트에서 얻은 결과 오른쪽 열은 실제 얻은 결과

[0.0]

[0,1]

...

(*0은 No, 1은 Yes)

 

Making the Confusion Matrix

sklearn.metrics.confusion_matrix — scikit-learn 1.0.1 documentation

confusion_matrix example

sklearn.metrics.accuracy_score — scikit-learn 1.0.1 documentation

acuuracy_score example

 

from sklearn.matrics import confusion_matrix, accuracy_score
cm = confusion_matrix(y_test, y_pred)
print(cm)
accuracy_score(y_test, y_pred)