如何在 Keras 中创建一个自定义损失函数来评估每个时期后的预测？

Question

我正在使用 Keras 开发一个神经网络，可以将英语句子翻译成一种自定义语言。为此，我想创建一个自定义损失函数，它对每个句子进行预测并评估它是否符合自定义语言的语法规则，如果不符合标准损失函数的价值。

如何在每个 epoch 之后而不是在编译期间评估张量？

下面是我的自定义损失函数。由于在模型编译期间还没有批次，y_pred 的形状为 (None, x, y) 并且无法评估以获得预测。我的想法是在编译期间分配一个标准损失函数，当批次到达时计算自定义损失。不幸的是，从未达到自定义损失。

def custom_loss(tokenizer, punishment_rate):

    def compile_loss(y_true, y_pred):
        shape = K.int_shape(y_pred)
        #standard loss function
        loss = K.sparse_categorical_crossentropy(y_true, y_pred)

        if shape[0] is not None:
            #THIS is never reached and that's the problem
            prediction = logits_to_text(K.eval(y_pred), tokenizer)

            #test if prediction complies to grammar rules
            compileable = compiles(prediction) ^ 1

            compile_error = compileable * punishment_rate
            loss =  K.sparse_categorical_crossentropy(y_true, y_pred, axis=-1) * (1 + compile_error)

        return loss

    return compile_loss

是否有任何解决方法可以仅在填充批次时评估张量？或者，在模型编译后通过回调更改损失函数，而不必重新编译模型？

Answer 1

根据keras source，您可以使用损失函数包装器来创建自定义损失函数类，然后将其无缝传递给您的模型。

举个例子：

#Import the wrapper
from keras.losses import LossFunctionWrapper

#Create your class extending the wrapper
class MyLossFunction(LossFunctionWrapper):

#Implement the constructor - here you can give extended arguments to it.
   def __init__(self, 
                 tokenizer, 
                 punishment_rate,
                 reduction=losses_utils.Reduction.SUM_OVER_BATCH_SIZE,
                 name='my_custom_text_function'):
        super(MyLossFunction, self).__init__(
            my_function, 
            name=name,
            reduction=reduction,
            tokenizer = tokenizer, 
            punishment_rate= punishment_rate)

#Now you have to define your function "my_function":
#Please, notice that ALL loss functions that follow keras model needs two arguments: 
#y_true (correct result) and y_pred (the result obtained in the network).
def my_function(y_true, y_pred, tokenizer, punishment_rate):
   shape = K.int_shape(y_pred)
   if shape[0] is not None:
       prediction = logits_to_text(K.eval(y_pred), tokenizer)

       #test if prediction complies to grammar rules
       compileable = compiles(prediction) ^ 1

       compile_error = compileable * punishment_rate
       return K.sparse_categorical_crossentropy(y_true, y_pred, axis=-1) * (1 + compile_error)
   return K.sparse_categorical_crossentropy(y_true, y_pred)

然后你可以实例化它并在你的编译器中使用：

custom_loss= MyLossFunction(tokenizer = ..., punishment_rate = ...)
classifier.compile(optimizer=optimizer, 
                      loss=custom_loss, 
                      metrics= ['binary_accuracy'])