Keras模型导入异常

ValueError

代码部分如下

# 训练模型
early_stopping = EarlyStopping(monitor='val_loss',patience=100)          # 早停,500轮
checkpoint = ModelCheckpoint("./best_model2.hdf5",
                              monitor='val_loss',
                              save_weights_only=True,
                              verbose=1,
                              save_best_only=True, 
                              period=1)   # 保存最佳模型参数
model.fit(train_data_x, train_data_y, epochs=10000, batch_size=128, validation_split=0.2, callbacks=[early_stopping,checkpoint])
# 保存模型
model.save('./model_5w_1day.h5') 

model = load_model('best_model2.hdf5')
scores= model.evaluate(test_data_x, test_data_y, verbose=0)
print("%s: %f" % (model.metrics_names[1], scores[1]))       # 模型性能：1是loss，2是mse，3是acc

然后报错，提示不能加载只读模型

ValueError                                Traceback (most recent call last)
<ipython-input-31-04beb895ff41> in <module>()
      1 # 评估模型
----> 2 model = load_model('best_model2.hdf5')
      3 scores= model.evaluate(test_data_x, test_data_y, verbose=0)
      4 print("%s: %f" % (model.metrics_names[1], scores[1]))       # 模型性能：0是loss，1是mse，2是acc
      5 pred_y = model.predict(test_data_x)

D:\Anaconda3\lib\site-packages\keras\engine\saving.py in load_model(filepath, custom_objects, compile)
    417     f = h5dict(filepath, 'r')
    418     try:
--> 419         model = _deserialize_model(f, custom_objects, compile)
    420     finally:
    421         if opened_new_file:

D:\Anaconda3\lib\site-packages\keras\engine\saving.py in _deserialize_model(f, custom_objects, compile)
    219         return obj
    220 
--> 221     model_config = f['model_config']
    222     if model_config is None:
    223         raise ValueError('No model found in config.')

D:\Anaconda3\lib\site-packages\keras\utils\io_utils.py in __getitem__(self, attr)
    300             else:
    301                 if self.read_only:
--> 302                     raise ValueError('Cannot create group in read only mode.')
    303                 val = H5Dict(self.data.create_group(attr))
    304         return val

ValueError: Cannot create group in read only mode.

最后找到原因：checkpoint保存模型时只保存了参数，而不是完整模型

checkpoint = ModelCheckpoint("./best_model2.hdf5",
                              monitor='val_loss',
                              save_weights_only=True,   # 只保存了模型参数，并不是完整模型
                              verbose=1,
                              save_best_only=True, 
                              period=1)   # 保存最佳模型参数

修改成如下代码就正常了

checkpoint = ModelCheckpoint("./best_model2.hdf5",
                              monitor='val_loss',
                              save_weights_only=False,
                              verbose=1,
                              save_best_only=True, 
                              period=1)   # 保存最佳模型参数