一, sklearn 中提供了高效的模型持久化模块 joblib, 将模型保存至硬盘.
- from sklearn.externals import joblib
- #lr 是一个 LogisticRegression 模型
- joblib.dump(lr, 'lr.model')
- lr = joblib.load('lr.model')
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')
两种保存 Model 的模块 pickle 与 joblib.
使用 pickle 保存
首先简单建立与训练一个 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 来保存与读取训练好的 Model. (若忘记什么是 pickle, 可以回顾 13.8 pickle 保存数据视频.)
- import pickle #pickle 模块
- # 保存 Model(注: save 文件夹要预先建立, 否则会报错)
- with open('save/clf.pickle', 'wb') as f:
- pickle.dump(clf, f)
- # 读取 Model
- with open('save/clf.pickle', 'rb') as f:
- clf2 = pickle.load(f)
- #测试读取后的 Model
- print(clf2.predict(X[0:1]))
- ==========================================================================================================
使用 joblib 保存
joblib 是 sklearn 的外部模块.
- from sklearn.externals import joblib #jbolib 模块
- # 保存 Model(注: save 文件夹要预先建立, 否则会报错)
- joblib.dump(clf, 'save/clf.pkl')
- # 读取 Model
- clf3 = joblib.load('save/clf.pkl')
- # 测试读取后的 Model
- print(clf3.predict(X[0:1]))
最后可以知道 joblib 在使用上比较容易, 读取速度也相对 pickle 快.
来源: http://www.bubuko.com/infodetail-2881345.html