将最后一层（输出层）的权重从经过训练的网络加载到新模型

Question

是否可以使用 set_weights 和 get_weights 方案将权重从经过训练的网络加载到我的新模型的最后一层？关键是，我将每一层的权重保存为一个 mat 文件（训练后），以便在 Matlab 中进行一些计算，我只想将最后一层的修改权重加载到我的新模型和其他层的最后一层获得与训练模型相同的权重。这有点棘手，因为保存的格式是 mat。weights1 = lstm_model1.layers[0].get_weights()[0]biases1 = lstm_model1.layers[0].get_weights()[1]weights2 = lstm_model1.layers[2].get_weights()[0]biases2 = lstm_model1.layers[2].get_weights()[1]weights3 = lstm_model1.layers[4].get_weights()[0]biases3 = lstm_model1.layers[4].get_weights()[1]# Save the weights and biases for adaptation algorithm savemat("weights1.mat", mdict={'weights1': weights1})  savemat("biases1.mat", mdict={'biases1': biases1})      savemat("weights2.mat", mdict={'weights2': weights2})   savemat("biases2.mat", mdict={'biases2': biases2})      savemat("weights3.mat", mdict={'weights3': weights3}) savemat("biases3.mat", mdict={'biases3': biases3})  我如何才能将其他层的旧权重加载到新模型（没有最后一层）并将最后一层的修改权重加载到新模型的最后一层？

Answer 1

如果将其另存为 .h5 文件格式，则可以正常工作。但是，我不确定 .mat：

简单来说，您只需调用get_weights所需的层，类似地，set_weights调用另一个模型的相应层：

last_layer_weights = old_model.layers[-1].get_weights()

new_model.layers[-1].set_weights(last_layer_weights)

如需更完整的代码示例，请点击此处：

# Create an arbitrary model with some weights, for example

model = Sequential(layers = [

    Dense(70, input_shape = (100,)),

    Dense(60),

    Dense(50),

    Dense(5)])

# Save the weights of the model

model.save_weights(“model.h5”)

# Later, load in the model (we only really need the layer in question)

old_model = Sequential(layers = [

    Dense(70, input_shape = (100,)),

    Dense(60),

    Dense(50),

    Dense(5)])

old_model.load_weights(“model.h5”)

# Create a new model with slightly different architecture (except for the layer in question, at least)

new_model = Sequential(layers = [

    Dense(80, input_shape = (100,)),

    Dense(60),

    Dense(50),

    Dense(5)])

# Set the weights of the final layer of the new model to the weights of the final layer of the old model, but leaving other layers unchanged.

new_model.layers[-1].set_weights(old_model.layers[-1].get_weights())

# Assert that the weights of the final layer is the same, but other are not.

print (np.all(new_model.layers[-1].get_weights()[0] == old_model.layers[-1].get_weights()[0]))

>> True

print (np.all(new_model.layers[-2].get_weights()[0] == old_model.layers[-2].get_weights()[0]))

>> False