如何打印Keras张量的值? [英] How can I print the values of Keras tensors?

问题描述

我正在实现自己的Keras损失功能.如何访问张量值?

I am implementing own Keras loss function. How can I access tensor values?

我尝试过的

def loss_fn(y_true, y_pred):
    print y_true

它打印

Tensor("target:0", shape=(?, ?), dtype=float32)

是否有Keras函数可以访问y_true值?

Is there any Keras function to access y_true values?

推荐答案

Keras的后端具有print_tensor，使您可以执行此操作.您可以通过以下方式使用它:

Keras' backend has print_tensor which enables you to do this. You can use it this way:

import keras.backend as K

def loss_fn(y_true, y_pred):
    y_true = K.print_tensor(y_true, message='y_true = ')
    y_pred = K.print_tensor(y_pred, message='y_pred = ')
    ...

该函数返回相同的张量.当评估该张量时，它将打印其内容，并以message开头. 从 Keras文档:

The function returns an identical tensor. When that tensor is evaluated, it will print its content, preceded by message. From the Keras docs:

请注意，print_tensor返回的新张量与x相同，应在以下代码中使用.否则，评估过程中将不考虑打印操作.

Note that print_tensor returns a new tensor identical to x which should be used in the following code. Otherwise the print operation is not taken into account during evaluation.

因此，请确保随后使用张量.

So, make sure to use the tensor afterwards.

这篇关于如何打印Keras张量的值?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！