machine_learn/linear/train_feature.py

#!/usr/bin/python
# -*- coding: UTF-8 -*-
import sys
reload(sys) 
sys.setdefaultencoding('utf-8') 
import numpy as np
from sklearn.linear_model import LinearRegression
from sklearn import metrics

'''
随机特征
'''
def extend_feature_repeat(x):
	return [x[0],x[0]]


'''
重复特征
'''
def extend_feature_random(x):
	import random
	return [x[0],random.uniform(-10,10)]


def read_data(path, fun):
	with open(path) as f :
		lines=f.readlines()
	lines=[eval(line.strip()) for line in lines]
	X,y=zip(*lines)
	X=[fun(x) for x in X]
	X=np.array(X)
	y=np.array(y)
	return X,y


def repeat():
	X_train,y_train=read_data("train_data", extend_feature_repeat)
	X_test,y_test=read_data("test_data", extend_feature_repeat)
	model = LinearRegression()
	model.fit(X_train, y_train)
	print model.coef_
	print model.intercept_

	y_pred_train = model.predict(X_train)
	train_mse=metrics.mean_squared_error(y_train, y_pred_train)
	print "重复特征"
	print "MSE:", train_mse
	y_pred_test = model.predict(X_test)
	test_mse=metrics.mean_squared_error(y_test, y_pred_test)
	print "MSE:",test_mse
	print "推广mse差", test_mse-train_mse


def random():
	X_train, y_train = read_data("train_data", extend_feature_random)
	X_test, y_test = read_data("test_data", extend_feature_random)
	model = LinearRegression()
	model.fit(X_train, y_train)
	print model.coef_
	print model.intercept_

	y_pred_train = model.predict(X_train)
	train_mse = metrics.mean_squared_error(y_train, y_pred_train)
	print "+随机特征"
	print "MSE:", train_mse
	y_pred_test = model.predict(X_test)
	test_mse = metrics.mean_squared_error(y_test, y_pred_test)
	print "MSE:", test_mse
	print "推广mse差", test_mse - train_mse


if __name__ == '__main__':
	repeat()
	print('-------------------')
	random()