python实现多元线性回归

import matplotlib.pyplot as plt import pandas as pd import numpy as np from sklearn.linear_model import LinearRegression # 导入线性回归模块 from sklearn.preprocessing import PolynomialFeatures # 二元多项式公式：z=intercept+coefficients[0]*x*y+coefficients[1]*x+coefficients[2]*y+ # coefficients[3]*x²+coefficients[4]*x*y+coefficients[5]*y² # 推到出来的公式： # variable_X = [1, x1, x2, x3, x4, x 1 ∗ x 1 x1*x1x1∗x1, x 1 ∗ x 2 x1*x2x1∗x2, x 1 ∗ x 3 x1*x3x1∗x3, x 1 ∗ x 4 x1*x4x1∗x4, x 2 ∗ x 2 x2*x2x2∗x2, x 2 ∗ x 3 x2*x3x2∗x3, x 2 ∗ x 4 x2*x4x2∗x4, x 3 ∗ x 3 x3*x3x3∗x3, x 3 ∗ x 4 x3*x4x3∗x4, x 4 ∗ x 4 x4*x4x4∗x4] # x是表示自变量数值如：Z=ax+bY+c，x的向量第一个是x--30.19，x向量的第二个是Y--55， # x = [[30.19, 55], [22, 55.85], [17.8, 46.1], [12.4, 56.4], [4.5, 52.5], [30.19, 40.25], [22, 39.75], [17.75, 39.3], [15.38, 40.9], # [4.5, 38.5], [30.4, 35.2], [22, 34.76], [17.9, 34.1], [13.38, 35.5], [4.5, 34.5], [30.4, 25.25], [22, 25.53], [17.19, 24.7], # [14.09,25.4], [4.5, 25], ] # # Y变量是Z，因变量 # y = [66, 67.8, 57, 68.8, 62.5, 50.5, 48.2, 49.6, 48.6, 45, 44.4, 42.76, 42.6, 41.6, 41.5, 33.2, 30.6, 29.2, 29.2, 27, # ] # file =r'D:\work\py\pycure.xlsx' # data=pd.read_excel(file) # data.columns=['y','x1','x2'] # x1=sm.add_constant(data.iloc[:,1:]) #生成自变量 # y1=data['y']#生成因变量 df = pd.read_excel("D:\work\py\pycure.xlsx", usecols=[1, 2],names=None) # 读取项目名称和行业领域两列，并不要列名 取第2，3列 x = df.values.tolist() df = pd.read_excel("D:\work\py\pycure.xlsx", usecols=[0, 0],names=None) # 读取项目名称和行业领域两列，并不要列名 取第一列 df_li = df.values.tolist() y = [] for s_li in df_li: y.append(s_li[0]) # print(y) # print(x) # print(y) print("==================================================================") for index in range(1, 100): data = pd.DataFrame({'IN': x, 'OUT': y}) data_train = np.array(data['IN']).reshape(data['IN'].shape[0], 1) # print(data) data_test = data['OUT'] # print(data_train) poly_reg = PolynomialFeatures(degree=index) X_ploy = poly_reg.fit_transform(x) regr = LinearRegression() regr.fit(X_ploy, data_test) print("vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv") print("degree = ", index) print("coefficients = ", regr.coef_) print("intercept = ", regr.intercept_) print("R^2 = ", regr.score(X_ploy, data_test)) print("^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^") if (regr.score(X_ploy, data_test) >= 0.999): break print("==================================================================")