python如何利用joblib保存训练模型

from sklearn.externals import joblib
#保存模型
def Save_Model(self, model, filepath):
    joblib.dump(model, filename=filepath)
def Decision_Tree_classifier(self,x_train,y_train,max_depth=None,min_samples_split=2,min_samples_leaf=1):
    Decision_Tree=tree.DecisionTreeClassifier(max_depth=max_depth,min_samples_split=min_samples_split,min_samples_leaf=min_samples_leaf)
    Decision_Tree.fit(x_train,y_train)
    self.save_model(Decision_Tree,os.path.join(c_config.UPLOAD_FOLODER,'model','Decision_Tree.m'))
    return Decision_Tree

def Load_Model(self, filepath):
    model = joblib.load(filepath)
    return model

import pickle
# train the model
model = ...
# save the model
with open('my_model.pkl', 'wb') as f:
    pickle.dump(model, f)

import pickle
# load the saved model
with open('my_model.pkl', 'rb') as f:
    model = pickle.load(f)
# predict using the loaded model
model.predict(X)

import pickle
# load the saved model
with open('my_model.pkl', 'rb') as f:
    model = pickle.load(f)
# continue training the model
model.fit(X_train, y_train)
# save the updated model
with open('my_updated_model.pkl', 'wb') as f:
    pickle.dump(model, f)

import joblib
# train the model
model = ...
# save the model
joblib.dump(model, 'my_model.joblib')

import joblib
# load the saved model
model = joblib.load('my_model.joblib')
# predict using the loaded model
model.predict(X)

import joblib
# load the saved model
model = joblib.load('my_model.joblib')
# continue training the model
model.fit(X_train, y_train)
# save the updated model
joblib.dump(model, 'my_updated_model.joblib')

# train the model
model = ...
# save the model weights to HDF5 file
model.save_weights('my_model_weights.h5')

import tensorflow as tf
# define the model architecture
model = ...
# load the saved model weights
model.load_weights('my_model_weights.h5')
# predict using the loaded model weights
model.predict(X)

import tensorflow as tf
# define the model architecture
model = ...
# load the saved model weights
model.load_weights('my_model_weights.h5')
# continue training the model
model.fit(X_train, y_train)
# save the updated model weights to HDF5 file
model.save_weights('my_updated_model_weights.h5')

import onnx
# train the model
model = ...
# convert the model to ONNX format
onnx_model = onnx.convert(model)
# save the model
onnx.save_model(onnx_model, 'my_model.onnx')

import onnxruntime
# load the saved model
onnx_session = onnxruntime.InferenceSession('my_model.onnx')
# predict using the loaded model
input_name = onnx_session.get_inputs()[0].name
output_name = onnx_session.get_outputs()[0].name
result = onnx_session.run([output_name], {input_name: X})