机器学习---线性回归器

import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
from sklearn.datasets import load_iris


data,label_ = load_iris(return_X_y=True)
fram = np.zeros(150*5).reshape([150,5])
fram[:,:4] = data
fram[:,4] = label_



number_train = int(0.8*data.shape[0])
number_test = data.shape[0]-number_train

X_train = data[:number_train,1].reshape(-1,1)
y_train = data[:number_train,2].reshape(-1,1)

X_test = data[number_train:,1].reshape(-1,1)
y_test = data[number_train:,2].reshape(-1,1)

from sklearn import linear_model
linear_regressior = linear_model.LinearRegression()
linear_regressior.fit(X_train,y_train)

y_predict = linear_regressior.predict(X_train)

plt.figure(dpi=100,figsize=(10,8))
plt.scatter(X_train,y_train,c="r")
plt.plot(X_train,y_predict,c="b",linewidth = 4)
plt.show()
y_test_predict = linear_regressior.predict(X_test)

import sklearn.metrics as sm
print("均方误差：",round(sm.mean_squared_error(y_test,y_test_predict,),2))
print("解释方差分：",round(sm.explained_variance_score(y_test,y_test_predict,),2))

import pickle
output_file = "./saved_model.pkl"
with open(output_file,"wb") as f:
    pickle.dump(linear_regressior,f)