Tensorflow Keras自定义损失函数访问张量通道

问题描述：

我有一个形状为(64、64、2)的2通道numpy数组，作为我的CNN的输入. 我要构建 https://www.tensorflow.org/guide中所述的自定义损失函数/keras/train_and_evaluate :

I have a 2-channel numpy array of shape (64, 64, 2) as input to my CNN. I want to build a customized loss function as described in https://www.tensorflow.org/guide/keras/train_and_evaluate :

def basic_loss_function(y_true, y_pred):
    return tf.math.reduce_mean(tf.abs(y_true - y_pred))

model.compile(optimizer=keras.optimizers.Adam(),
              loss=basic_loss_function)

model.fit(x_train, y_train, batch_size=64, epochs=3)

但是我想要比这个基本的东西更复杂的东西.我需要做一个逆DFT(ifft2d)，并且我的y_pred和y_true期望分别是形状(64,64,2)，其中2个通道是fft2的实部和虚部.如何正确访问y_pred和y_true通道(我猜这是某种keras/张量层?)以 RealPart + 1j * ImagPart 的形式重建复数(用numpy表示) y_pred [:，:，0]和y_pred [:，:，1])吗?

But I want to something more complicated that this basic one. What I need is to do an inverse DFT (ifft2d) and my y_pred and y_true are expected to be each of shape (64,64,2), with the 2 channels being the real and imaginary parts of a fft2. How can I access correctly the y_pred and y_true channels (which are some kind of keras/tensor layer I guess?) to rebuild a complex number in the form RealPart+1j*ImagPart (in numpy it would be y_pred[:,:,0] and y_pred[:,:,1] ) ?

->总之，有人知道y_pred和y_true是哪种对象以及如何访问其通道/元素吗? (这不容易调试，因为它需要在已编译的CNN中运行，因此请事先了解一下)

--> In summary, does someone know exactly what kind of object is y_pred and y_true and how to access their channels/elements? (This is not easy to debug since would need to be run in a compiled CNN, so better know it beforehand)

答

y_true和y_pred是形状为(batchsize, ...[output shape]...)的张量.输入的形状为(64,64,2)，但我不确定输出的外观，如果输出确实为(64,64,2)，则y_pred或y_true的形状为(64,64,64,2).

y_true and y_pred are Tensors of shape (batchsize, ...[output shape]...). Your input has shape (64,64,2) but I'm not sure what your output looks like, if your output is indeed (64,64,2) then y_pred or y_true have shape (64,64,64,2) given your batchsize=64.

使用张量进行处理非常类似于numpy的语法，因此您可以使用带有张量的切片符号，例如y_true[:,:,:,0](请注意添加的批处理维度).

Dealing with Tensors is very much like numpy's syntax so You can use slice notation with tensors e.g., y_true[:,:,:,0] (note the added batch dimension).

Tensorflow具有用于计算DFT，FFT等的功能.参见 tf.signal 和

Tensorflow has functions for computing DFT,FFT,..etc. See tf.signal and tf.signal.rfft2d

如果损失函数涉及输入操作，而不仅仅是输出y_true和y_pred，则可以按以下方式使用model.add_loss代替model.compile(loss= basic_loss_function)

If your loss function involves operations on the input, not just the outputs y_true and y_pred, then you can use model.add_loss instead of model.compile(loss= basic_loss_function) as follows

x = Input(shape=(64,64,2))
y_true = Input(shape=...))
# your CNN layers
y_pred = Dense(128)(net)

model = Model(input=[x, y_true], output=output)
model.add_loss(basic_loss_function(x, y_true, y_pred))

请注意，标签(又名y_true)现在是模型的输入.

note that the labels (aka y_true) is now an input to the model.

Tensorflow Keras自定义损失函数访问张量通道

相关推荐