#way1  利用pickle.dump()將訓練好的分類器序列化（轉為二進位制），利用 pickle.loads()反序列化；
>>> from sklearn import svm
>>> from sklearn import datasets
>>> clf = svm.SVC(gamma='scale')
>>> 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='ovr', degree=3, gamma='scale', kernel='rbf',
    max_iter=-1, probability=False, random_state=None, shrinking=True,
    tol=0.001, verbose=False)

>>> import pickle
>>> s = pickle.dumps(clf)  #儲存模型
>>>
 
 clf2 = pickle.loads(s)  #下載模型
>>> clf2.predict(X[0:1])
array([0])
>>> y[0]
0

#  In the specific case of scikit-learn, it may be better to use joblib’s replacement of pickle (joblib.dump & joblib.load), which is more efficient on objects that carry large numpy arrays internally as
 
 is often the case for fitted scikit-learn estimators, but can only pickle to the disk and not to a string

#way2  joblib.dump()  joblib.load()
>>> from sklearn.externals import joblib
>>> joblib.dump(clf, 'filename.joblib')   #儲存模型
>>> clf = joblib.load('filename.joblib')   #下載模型

在使用上述函式下載模型時，存在一些security 和 maintainability問題。具體參見：官方文件