Supervised Learning - 분류하기

 

모델 : decision_tree ,random_forest, svm_model, sgd_model, logistic_model

 

정답지가 있는 데이터 기반으로 학습 및 예측

예측한 결과를 답지와 비교를 통해 정확도를 파악한다.

 

1. 데이터 준비

from sklearn.datasets import load_breast_cancer
from sklearn.model_selection import train_test_split
from sklearn.metrics import classification_report
import pandas as pd

# 데이터 불러오기
bc = load_breast_cancer()
bc_data = bc.data
bc_label = bc.target

# 데이터 정리
bc_df = pd.DataFrame(data=bc_data, columns=bc.feature_names)
bc_df["label"] = bc.target

2. 모델준비 및 학습

from sklearn.model_selection import train_test_split

# trainig dataset, test dataset
X_train, X_test, y_train, y_test = train_test_split(bc_data, 
                                                    bc_label, 
                                                    test_size=0.2, 
                                                    random_state=7)

from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn import svm
from sklearn.linear_model import SGDClassifier
from sklearn.linear_model import LogisticRegression

#모델준비
decision_tree = DecisionTreeClassifier(random_state=32)
random_forest = RandomForestClassifier(random_state=32)
svm_model = svm.SVC()
sgd_model = SGDClassifier() 
logistic_model = LogisticRegression()

model_list = [decision_tree ,random_forest, svm_model, sgd_model, logistic_model]
#학습
for model in model_list : 
  model.fit(X_train, y_train)

3. 예측

y_pred_list = []
for model in model_list : 
  y_pred = model.predict(X_test)
  y_pred_list.append(y_pred)

4. 정확도

from sklearn.metrics import accuracy_score
acc_list = []
for y_pred in y_pred_list : 
  accuracy = accuracy_score(y_test, y_pred)
  acc_list.append(accuracy)

class_report_list = [];
from sklearn.metrics import classification_report
for y_pred in y_pred_list : 
  class_report_list.append(classification_report(y_test, y_pred))