导入相关包

import numpy as np import pandas as pd from sklearn.model_selection import KFold from sklearn.linear_model import Lasso from xgboost import XGBClassifier from sklearn.metrics import accuracy_score import matplotlib.pyplot as plt

读取乳腺癌数据集

data = pd.read_csv('./breast_cancer.csv')

将输入特征和标签分离

X = data.iloc[:, 0:-1].values y = data.iloc[:, -1].values

五折交叉划分数据集

kf = KFold(n_splits=5, shuffle=True, random_state=0)

初始化特征筛选器

lasso = Lasso(alpha=0.001)

保存筛选出的重要特征

important_features = []

保存特征性能曲线

x_axis = [] y_axis = []

对每一折数据集进行训练

for i, (train_index, test_index) in enumerate(kf.split(X)): # 选取训练集 X_train = X[train_index] y_train = y[train_index] # 选取测试集 X_test = X[test_index] y_test = y[test_index] # 特征筛选 lasso.fit(X_train, y_train) # 获取筛选出的重要特征及其特征性能 important_features_temp = [X_train[i] for i in range(len(lasso.coef_)) if lasso.coef_[i] != 0] x_axis.append(i) y_axis.append(lasso.score(X_train, y_train)) # 将筛选出的重要特征合并到总的重要特征列表中 important_features = list(set(important_features_temp + important_features))

用筛选出的重要特征训练Xgboost模型

model = XGBClassifier() model.fit(X_train[important_features], y_train)

计算模型的准确率

acc = accuracy_score(y_test, model.predict(X_test[important_features]))

画出特征性能曲线

plt.plot(x_axis, y_axis) plt.title('Feature Performance Curve') plt.show()

print('Accuracy', acc)