体重预测[线性回归]

'''一、问题描述：根据表1所示训练数据构建线性回归模型并求取测试数据的预测精度'''

import numpy as np

import matplotlib.pyplot as plt

from sklearn.linear_model import LinearRegression

x_train=np.array([[0.86],[0.96],[1.12],[1.35],[1.55],[1.63],[1.71],[1.85]])

y_train=np.array([[12],[15],[20],[35],[48],[51],[59],[75]])

x_test=[[0.83],[1.08],[1.26],[1.51],[1.60],[1.67],[1.75],[1.90]]

y_test=[[11],[17],[27],[41],[50],[64],[66],[89]]

'''二、数据处理：

数据归一化处理'''

X_train=(x_train-np.min(x_train))/(np.max(x_train)-np.min(x_train))

Y_train=(y_train-np.min(y_train))/(np.max(y_train)-np.min(y_train))

X_test=(x_test-np.min(x_test))/(np.max(x_test)-np.min(x_test))

Y_test=(y_test-np.min(y_test))/(np.max(y_test)-np.min(y_test))

'''======================================================================'''

'''三、利用线性回归构建模型'''

'''归一化'''

L=LinearRegression(fit_intercept=True)

L.fit(X_train,Y_train)

ys=L.score(X_test,Y_test)

print(ys)

'''未归一化'''

L.fit(x_train,y_train)

yn=L.score(x_test,y_test)

print(yn)

'''====================================================================='''

#精度对比图

plt.figure(1)

plt.xlim(0, 1)

plt.ylim(0, 1)

plt.xlabel("DataClass")

plt.ylabel("TestAccuracy")

Data_list=['Normalized','Not Normalized']

xlis=[0.3,0.7]

ylis=[ys,yn]

plt.bar(xlis,ylis,width=0.3,tick_label=Data_list,color=['g','b'])

for i in range(len(xlis)):

    plt.text(xlis[i], ylis[i],format(ylis[i],'.2f'))

plt.show()

'''======================================================================'''

#身高体重预测图

print("L的截距为：",L.intercept_[0])

print("L的斜率为：",L.coef_[0][0])

b=L.intercept_[0]

k=L.coef_[0][0]

Y_fit=k*np.array(x_test)+b

plt.figure(2)

plt.xlabel("height")

plt.ylabel("weight")

plt.scatter(x_test,Y_fit,color='black',label='fit')

plt.plot(x_test,Y_fit,color='red',label='fit_line')

plt.scatter(x_test,y_test,color='y',label='test')

plt.grid(True)

plt.legend(loc='upper left')

plt.show()

'''四、实验分析：

体重数据随身高数据的增大而增大，归一化数据的测试精度比未归一化数据测试精度高'''