keras访问预先训练模型的访问层参数以进行冻结

I saved an LSTM with multiple layers. Now, I want to load it and just fine-tune the last LSTM layer. How can I target this layer and change its parameters?

我保存了一个多层的LSTM。现在，我想加载它并调整最后一个LSTM层。如何定位此图层并更改其参数？

Example of a simple model trained and saved:

训练和保存的简单模型示例：

model = Sequential()
# first layer  #neurons 
model.add(LSTM(100, return_sequences=True, input_shape=(X.shape[1], 
X.shape[2])))
model.add(LSTM(50, return_sequences=True))
model.add(LSTM(25))
model.add(Dense(1))
model.compile(loss='mae', optimizer='adam')

I can load and retrain it but I can't find a way to target specific layer and freeze all the other layers.

我可以加载和重新训练它，但我找不到一种方法来定位特定的图层并冻结所有其他图层。

2 个解决方案

#1

If you have previously built and saved the model and now want to load it and fine-tune only the last LSTM layer, then you need to set the other layers' trainable property to False. First, find the name of the layer (or index of the layer by counting from zero starting from the top) by using model.summary() method. For example this is the output produced for one of my models:

如果您之前已经构建并保存了模型，现在想要加载它并仅微调最后一个LSTM图层，那么您需要将其他图层的可训练属性设置为False。首先，使用model.summary（）方法找到图层的名称（或从顶部开始从零开始计算图层的索引）。例如，这是我的一个模型的输出：

_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
input_10 (InputLayer)        (None, 400, 16)           0         
_________________________________________________________________
conv1d_2 (Conv1D)            (None, 400, 32)           4128      
_________________________________________________________________
lstm_2 (LSTM)                (None, 32)                8320      
_________________________________________________________________
dense_2 (Dense)              (None, 1)                 33        
=================================================================
Total params: 12,481
Trainable params: 12,481
Non-trainable params: 0
_________________________________________________________________

Then set the trainable parameters of all the layers except the LSTM layer to False.

然后将除LSTM图层以外的所有图层的可训练参数设置为False。

Approach 1:

方法1：

for layer in model.layers:
    if layer.name != `lstm_2`
        layer.trainable = False

Approach 2:

方法2：

for layer in model.layers:
    layer.trainable = False

model.layers[2].trainable = True  # set lstm to be trainable

# to make sure 2 is the index of the layer
print(model.layers[2].name)    # prints 'lstm_2'

Don't forget to compile the model again to apply these changes.

不要忘记再次编译模型以应用这些更改。

#2

An easy solution would be to name each layer, i.e.

一个简单的解决方案是命名每一层，即

model.add(LSTM(50, return_sequences=True, name='2nd_lstm'))

Then, upon loading the model you can iterate over the layers and freeze the ones matching a name condition:

然后，在加载模型时，您可以迭代图层并冻结符合名称条件的图层：

for layer in model.layers:
    if layer.name == '2nd_lstm':
        layer.trainable = False

Then you need to recompile your model for the changes to take effect, and afterwards you may resume training as usual.

然后，您需要重新编译模型才能使更改生效，之后您可以像往常一样恢复培训。