机器学习（六）使用sklearn库的model存储

# -*- coding: utf-8 -*-
"""
Created on Tue Oct 25 21:39:04 2016

@author: Administrator
model说明，normalization
"""

from sklearn import datasets
from sklearn.cross_validation import train_test_split
from sklearn.linear_model import LinearRegression


load_data=datasets.load_boston()
data_x=load_data.data
data_y=load_data.target

x_train,x_test,y_train,y_test=train_test_split(data_x,data_y,test_size=0.1)

model=LinearRegression()
model.fit(x_train,y_train)

"""
两种存储model的形式

"""

"""
method 1: pickle
"""
import pickle

#save
with open('clf.pickle','wb') as f:
    pickle.dump(model,f)

#restore
with open('clf.pickle','rb') as f:
    model2=pickle.load(f)
print model2.predict(data_x[:10,:])
print data_y[:10]

print ' +----------------------------------------+'
"""
method 2: sklearn 外部库joblib
"""
from sklearn.externals import joblib
#save
joblib.dump(model,'clf2.pkl')
#restore
model3=joblib.load('clf2.pkl')
print model3.predict(data_x[10:20,:])
print data_y[10:20]