Tensorflow:在conv2D_transpose的output_shape中使用None [英] Tensorflow: use None in output_shape in conv2D_transpose

问题描述

我想在网络中使用conv2d_tranpose(或反卷积)而不是向上采样. 这需要将output_shape传递给函数调用.没问题，我可以计算出来.但是我想对Batch_size使用None来保持设置的灵活性. 有可能吗?

I would like to use a conv2d_tranpose (or deconvolution) instead of upsampling in my network. This requires passing an output_shape to the function call. That is not a problem, I can calculate it. But I would like to use None for the batch_size to keep the setup flexible. Is that possible ?

这是代码行:

tf.nn.conv2d_transpose(hd_conv1, Wd_conv1, [batch_size, 14,14,64], strides=[1,2,2,1], padding="SAME")

batch_size只是我在脚本顶部设置的变量.这段代码运行良好，但是如果我使用None而不是batch_size:

batch_size is simply a variable that I set at the top of my script. This code runs fine, but if I use None instead of batch_size:

TypeError:预期的二进制或Unicode字符串，没有任何内容

TypeError: Expected binary or unicode string, got None

如果我只是省略第一个维度:

If I just leave out the first dimension:

ValueError:output_shape必须具有形状(4，)，得到的形状(3，)

ValueError: output_shape must have shape (4,), got (3,)

我认为奇怪的是，有很多方法可以处理batch_size.有些操作只是忽略它，例如普通的conv2d，但是在这里我需要明确地指定它. 无论如何，我想知道为什么我必须自己计算出output_shape.在给定输入，步幅，填充的情况下，应该易于计算. 关于output_shape的推断，有一个github 问题，可悲的是似乎没有任何后续行动.

I think it is strange that there are different ways to deal with the batch_size. Some operations simply ignore it, such as the normal conv2d, but here I need to specify it explicitly. In any case I wondered why I would have to calculate the output_shape myself, at all. With given input, strides, padding, that should be easy to calculate. There is a github issue regarding inference of output_shape, sadly there doesn't seem to be any follow-up.

我做对了吗-在output_shape中传递一个显式的batch_size? 有没有办法省略显式的batch_size?

Am I doing this right - passing in an explicit batch_size in the output_shape ? Is there a way to omit the explicit batch_size ?

推荐答案

请不要使用None，而应使用如下所示的符号表示形式.

Instead of using None, please use symbolic representation like below.

batch_size = tf.shape(something_or_other)[0]  
deconv_shape = tf.pack([batch_size, 40, 40, 32])  
conv2d_transpose(..., output_shape=deconv_shape, ...)  

请注意不要使用tf.get_shape(). tf.get_shape()和tf.shape()略有不同.

Be careful not to use tf.get_shape(). tf.get_shape() and tf.shape() are slightly different.

另请参阅tensorflow网站中有关批量大小可变的建议.

Also see recomendations in tensorflow site about variable batch size.

https://www.tensorflow.org/programmers_guide/faq

这篇关于Tensorflow:在conv2D_transpose的output_shape中使用None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！