如何使用Keras构建自定义损失函数 [英] how to build custom loss function with Keras

问题描述

我有为暹罗网络实现的丢失功能.在Keras中，如果必须构建自己的损失函数，则应仅将输入参数作为(y_true，y_pred). 但就我而言，我有y_pred1，y_pred1，y_true1(class_label)，y_true2(class_label)，y_true3(相似性标签)

I have loss function which implemented for siamese network. In Keras if you have to build your own loss function, it should only take input arguments as (y_true, y_pred). But in my case I have y_pred1, y_pred1, y_true1(class_label), y_true2(class_label), y_true3(similarity label)

所以我的解决方案是将我喜欢的东西串联起来:

So my solution is to concatenate what I have like:

def my loss ( y_true, y_pred):
    y_true1 = y_true[:, 0]
    y_true2 = y_true[:, 1]
    label = y_true[:, 2]

    y_pred1 = y_pred[:, 0]
    y_pred2 = y_pred[:, 1]

第二个问题是，我有一个参数(alpha)，它是当前时期数的函数，我也应该将其传递给损失函数.

The second problem is, I have one parameter (alpha) which is a function of current epoch number that I should pass it to the loss function also.

通常，如果必须传递其他参数，则可以使用包装函数作为建议的解决方案

In general , if you have to pass some another argument you can use the wrapper function as the solution suggested here.

但是对于我来说，这无济于事，因为我的alpha值应根据当前纪元数进行更改.它基本上是当前纪元的Sigmoied函数.

But it will not help me in my case , because my alpha should be change depending on the current number of epoch. It is basically the Sigmoied function of the current epoch.

我可以跟踪纪元号的唯一方法是在自己的生成器中，因为我有内置在tfrecords中的数据集.因此，我正在使用自己的生成器将数据提供给模型.

The only way that I can track the epoch number is inside my own generator, because I have dataset built in tfrecords. So I am using my own generator to feed the data to model.

那么任何人都知道我该怎么办?如何跟踪当前纪元编号并使用它.

So any one have any idea what should I do? How I can track the current epoch number and use it.

推荐答案

重要！您是哪种情况?

• 案例1:具有3个输出的模型
• 案例2:一个具有一个输出的模型是三个输出的串联?

您需要三个独立的损失函数，每个函数将仅看到其自己的y_true和y_pred.

You need three independent loss functions, each function will see only its own y_true and y_pred.

def loss1(yTrue,yPred):
    ...
def loss2(yTrue,yPred):
    ...
def loss3(yTrue,yPred):
    ...

model.compile(loss=[loss1,loss2,loss3],...)

案例2

在这种情况下，您将能够按照建议的方式进行操作.

Case 2

In this case, you will be able to do it the way you proposed.

def my loss ( y_true, y_pred):
    y_true1 = y_true[:, 0]
    y_true2 = y_true[:, 1]

    y_pred1 = y_pred[:, 0]
    y_pred2 = y_pred[:, 1]

使用alpha

Alpha必须是张量"，而不是普通的var:

Using alpha

Alpha must be a "tensor", not an ordinary var:

alpha = K.variable(someInitialNpArray, dtype=...)

alpha值必须更改"，而不是重新分配:

The value of alpha must be "changed", not reassigned:

K.set_value(alpha, newValues)

现在，为on_epoch_end创建一个LambdaCallback，以更改alpha的值:

Now, create a LambdaCallback for on_epoch_end in order to change the value of alpha:

def changeAlpha(epoch,logs):
    #maybe use epoch+1, because it starts with 0
    K.set_value(alpha, valuesBasedOn(epoch))

alphaChanger = LambdaCallback(on_epoch_end=changeAlpha) #or on_epoch_begin (or start?)

损失:

def loss(true,pred):
    #blablabla

    #you can use alpha here

培训:

model.fit(..... callbacks = [alphaChanger])
model.fit_generator(......, callbacks = [alphaChanger])

这篇关于如何使用Keras构建自定义损失函数的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！