sklearn中线性回归模型中各种模块的导入路径及简单使用

模块导入总结

##############sklearn中的模块###############################################
###001KNN
from sklearn.neighbors import KNeighborsClassifier
clf = KNeighborsClassifier(n_neighbors=5)
clf.fit(X_train, y_train)
y_predict = clf.score(X_test, y_test)
y_predict
####002数据集
from sklearn import datasets		# 导入数据集

###003 线性回归
from sklearn.linear_model import LinearRegression
lin_reg = LinearRegression()
lin_reg.fit(X_train_std, y_train)
lin_reg.score(X_test_std, y_test)



###############sklearn.model_selection#####################################
###001--数据集分割
from sklearn.model_selection import train_test_split

###002网格搜索---------使用了网格搜索
from sklearn.model_selection import GridSearchCV	#网格搜索
clf = KNeighborsClassifier()
param_grid = [{
    'n_neighbors':[i for i in range(0,10)],
    'weights':['uniform','distance'] 
}
]
gs_clf = GridSearchCV(clf, param_grid=param_grid)
best_clf = gridsearch.best_estimator_					#最优模型
best_clf.score(X_test, y_test)




###############sklearn.metrics#################################################
###001准确率预测
from sklearn.metrics import accuracy_score
accuracy_score(y_predict, y_test)		# 准确值预测：r2

###002 均方误差
from sklearn.metrics import mean_squared_error
mean_squared_error(y_test, y_predict)

###003 绝对值误差
from sklearn.metrics import mean_absolute_error
mean_absolute_error(y_test, y_predict)

###004r2误差
from sklearn.metrics import r2_score
r2_score(y_test, y_predict)



##########
#######sklearn.preprocessing##################################################
###001数据归一化--实例化，fit，transform
from sklearn.preprocessing import StandardScaler
standscaler = StandardScaler()
standscaler.fit(X_train)
std_x_train = standscaler.transform(X_train)
std_x_test = standscaler.transform(X_test)

###########################sklearn.preprocessing###########################
###001 多项式回归模型--实例化，fit--transform
from sklearn.preprocessing import PolynomialFeatures
poly = PolynomialFeatures(degree = 2)
poly.fit(X)
X2 = poly.transform(X)

from sklearn.pipeline import Pipeline		#pipline使用
from sklearn.preprocessing import StandardScaler
pipe_reg = Pipeline([
    ('poly', PolynomialFeatures(degree=2)),
    ('scaler',StandardScaler()),
    ('lin_reg', LinearRegression())
])

pipe_reg.fit(X, y)
y_predict = pipe_reg.predict(X)

############################岭回归##########################################
from sklearn.linear_model import Ridge
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error
from sklearn.preprocessing import PolynomialFeatures
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler

def pipeRegression(degree, alpha):
    return Pipeline([
    ("poly", PolynomialFeatures(degree = degree)),
    ("scaler", StandardScaler()),
    ("lin_reg", Ridge(alpha = alpha))
]) 
ridge1 = pipeRegression(50, 0.000001)
ridge1.fit(X_train, y_train)

y1_predict = ridge1.predict(X_test)
mean_squared_error(y_test, y1_predict)

KNN算法

import numpy as np
import matplotlib.pyplot as plt
from math import sqrt
from collections import Counter

#获取数据

X = [[3.40, 2.8],
     [3.1, 1.8],
     [1.5, 3.4],
     [3.6, 4.7],
     [2.3, 2.9],
     [7.4,4.5],
     [5.7, 3.5],
     [9.2, 2.5],
     [7.9, 3.4]    
]

y = [0, 0, 0, 0, 0, 1, 1, 1, 1]

x_train = np.array(X)
y_train = np.array(y)

x = np.array([5.1, 3.4])

from sklearn.neighbors import KNeighborsClassifier
cls = KNeighborsClassifier(n_neighbors=3)
cls.fit(x_train, y_train)
cls.predict(x.reshape(1, -1))