如何在此代码段中使用 pickle？

Question

我有一个简单的代码片段来训练模型，但是当我使用 pickle 保存模型以备将来使用时，它给了我一个错误信息：

cannot pickle thread.LOCK objects

我使用了不止一种格式的泡菜，但它给了我同样的错误。

import pickle

model = keras.Sequential([
    keras.layers.Dense(SHAPE, input_shape=(SHAPE,)),
    keras.layers.Dense(300, activation='sigmoid'),
    keras.layers.Dense(10, activation='softmax')
])


#******************    COMPILING THE MODE        *****************
LEARNING_RATE = 0.0005
model.compile(optimizer=keras.optimizers.Adam(lr=LEARNING_RATE),
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy']              
             )

# ***********      TRAINING THE MODEL   **********
EPOCHS = 20
BATCH_SIZE=50

history_original_data = model.fit(X_original_train_images, y_original_train_labels, epochs=EPOCHS, batch_size=BATCH_SIZE) 
hist_original=history_original_data.history


### PICKLE TO SAVE THE MODEL TO BE USED WITHOU PRO-TRAINING IT
pickname ="SequentialNeuroNetwork.pkl"
PickleSeq = open(pickname, 'wb')
pickle.dump(model, PickleSeq)
PickleSeq.close()

我原以为上面的代码片段运行会很顺利，但它让我付出了代价。

Answer 1

您使用的是哪个版本的keras？我几乎可以肯定旧版本不支持 pickle。

或者，建议使用 model.save() 将您的模型保存在 keras 中。正如keras页面常见问题解答中所述：

You can use model.save(filepath) to save a Keras model into a single HDF5 file which will contain:

the architecture of the model, allowing to re-create the model

the weights of the model

the training configuration (loss, optimizer)

the state of the optimizer, allowing to resume training exactly where you left off.

You can then use keras.models.load_model(filepath) to reinstantiate your model. load_model will also take care of compiling the model using the saved training configuration (unless the model was never compiled in the first place).

来源：https://keras.io/getting-started/faq/#how-can-i-save-a-keras-model

如何在此代码段中使用 pickle？

How do I use pickle in this code snippet?

python

machine-learning

pickle

data-science

tensorflow