数据集介绍:
字段 | 字段说明 |
Survived | 是否存活(0代表否,1代表是) |
Pclass | 社会阶级(1代表上层阶级,2代表中层阶级,3代表底层阶级) |
Sex | 船上乘客的性别 |
Age | 船上乘客的年龄 |
代码部分:
import pandas as pd
from sklearn.preprocessing import LabelEncoder
#读取数据
data = pd.read_csv('泰坦尼克.csv')
#数据预处理
#对Pclass进行OntHot编码
data = pd.get_dummies(data,columns = ['Pclass'])
#将Sex列转换成0、1
le = LabelEncoder()
data['Sex'] = le.fit_transform(data['Sex']) #使用fit_transform()方法进行训练并转换
# 划分训练集测试集,测试集比例设置为0.2
from sklearn import model_selection
x = data.drop(['Survived'], axis=1)
y = data['Survived']
#函数最终将返回四个变量,分别为`x`的训练集和测试集,以及`y`的训练集和测试集。
x_train,x_test,y_train,y_test = model_selection.train_test_split(x, y,
random_state=10, #随机种子
test_size=0.2, #测试集比例
stratify=y) #保持比例
#调用模型
from sklearn.ensemble import RandomForestClassifier as RFC
rfc = RFC(n_estimators=100, random_state=10)
rfc.fit(x_train, y_train)
rfc_pred = rfc.predict(x_test)
#模型评估
from sklearn.metrics import classification_report,roc_auc_score,plot_roc_curve
from sklearn.metrics import average_precision_score,plot_precision_recall_curve
result = classification_report(y_test, rfc_pred)
rfc_prob = rfc.predict_proba(x_test)[:,1]
#输出AUC的值
auc_score = roc_auc_score(y_true=y_test,y_score=rfc_prob)
#输出AP值
ap_score = average_precision_score(y_true=y_test,y_score=rfc_prob)
#画出PR曲线
plot_precision_recall_curve(estimator=rfc,X=x_test,y=y_test,pos_label=1)
#画出ROC曲线
plot_roc_curve(estimator=rfc,X=x_test,y=y_test,pos_label=1)