Keras上对model进行保存和提取save&reload

本代码为Python实现Keras搭建神经网络训练回归模型

# Regressor save & reload example

import numpy as np
# for reproducibility
np.random.seed(1337)
from keras.models import Sequential
from keras.layers import Dense
from keras.models import load_model

# create some data
X = np.linspace(-1, 1, 200)
# randomize the data
np.random.shuffle(X)
Y = 0.5*X + 2 + np.random.normal(0, 0.05, (200, ))
# first 160 data points
X_train, Y_train = X[:160], Y[:160]
# last 40 data points
X_test, Y_test = X[160:], Y[160:]
model = Sequential()
model.add(Dense(output_dim = 1, input_dim = 1))
model.compile(loss = 'mse', optimizer = 'sgd')
for step in range(301):
    cost = model.train_on_batch(X_train, Y_train)

# 思路就是先创建一个model，做一次预测，
# 然后保存模型和参数，跟着把模型删除掉，
# 提取保存的模型文件，再做一次预测，对比输出结果
# save
print('test before save:', model.predict(X_test[0:2]))
# h5是保存格式，需要安装有HDF5，pip install h5py
model.save('my_model.h5')
# deletes the existing model
# 保存完模型之后，删掉它，后面可以来比较是否成功的保存
del model

# load
# 导入保存好的模型，再执行一遍预测
model = load_model('my_model.h5')
print('test after load:', model.predict(X_test[0:2]))

"""
# 另外还有其他保存模型并调用的方式
# 第一种是只保存权重而不保存模型的结构
# save and load weights
model.save_weights('my_model_weights.h5')
model.load_weights('my_model_weights.h5')

# 第二种是用 model.to_json 保存完结构之后，然后再去加载这个json_string
# save and load fresh network without trained weights
from keras.models import model_from_json
json_string = model.to_json()
model = model_from_json(json_string)
"""

运行结果：

Using TensorFlow backend.

test before save: [[1.8832371]
 [2.1850314]]
test after load: [[1.8832371]
 [2.1850314]]