一、 sklearn中提供了高效的模型持久化模組joblib，將模型儲存至硬碟。

from sklearn.externals import joblib
#lr是一個LogisticRegression模型
joblib.dump(lr, 'lr.model')
lr = joblib.load('lr.model')

連結：https://www.zhihu.com/question/27187105/answer/55895472  

二、pickle

>>> from sklearn import svm
>>> from sklearn import datasets
>>> clf = svm.SVC()
>>> iris = datasets.load_iris()
>>> X, y = iris.data, iris.target
>>> clf.fit(X, y)  
SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,
    decision_function_shape=None, degree=3, gamma='auto', kernel='rbf',
    max_iter=-1, probability=False, random_state=None, shrinking=True,
    tol=0.001, verbose=False)

>>> import pickle
>>> s = pickle.dumps(clf)
>>> clf2 = pickle.loads(s)
>>> clf2.predict(X[0:1])
array([0])
>>> y[0]
0


或者 ：
 

>>> from sklearn.externals import joblib
>>> joblib.dump(clf, 'filename.pkl') 
>>> clf = joblib.load('filename.pkl') 

兩種儲存Model的模組pickle與joblib。

使用 pickle 儲存 

首先簡單建立與訓練一個SVCModel。

from sklearn import svm
from sklearn import datasets

clf = svm.SVC()
iris = datasets.load_iris()
X, y = iris.data, iris.target
clf.fit(X,y)

==========================================================================================================
 

使用pickle來儲存與讀取訓練好的Model。 (若忘記什麼是pickle，可以回顧13.8 pickle 儲存資料視訊。)

import pickle #pickle模組

#儲存Model(注:save資料夾要預先建立，否則會報錯)
with open('save/clf.pickle', 'wb') as f:
    pickle.dump(clf, f)

#讀取Model
with open('save/clf.pickle', 'rb') as f:
    clf2 = pickle.load(f)
    #測試讀取後的Model
    print(clf2.predict(X[0:1]))

==========================================================================================================
 

使用 joblib 儲存 

joblib是sklearn的外部模組。

from sklearn.externals import joblib #jbolib模組

#儲存Model(注:save資料夾要預先建立，否則會報錯)
joblib.dump(clf, 'save/clf.pkl')

#讀取Model
clf3 = joblib.load('save/clf.pkl')

#測試讀取後的Model
print(clf3.predict(X[0:1]))

# [0]

最後可以知道joblib在使用上比較容易，讀取速度也相對pickle快。

===================================================================== 連結：https://www.zhihu.com/question/27187105/answer/97334347

https://morvanzhou.github.io/tutorials/machine-learning/sklearn/3-5-save/