如何在Tensorflow中使用conv1d_transpose? [英] How to use the conv1d_transpose in Tensorflow?

问题描述

conv1d_transpose尚未在Tensorflow的稳定版本中提供，但实现可用

The conv1d_transpose is not yet in the stable version of Tensorflow, but an implementation is available on github

我想创建一个一维反卷积网络.输入的形状为[-1, 256, 16]，输出的形状应为[-1,1024,8].内核的大小为5，步幅为4.

I would like to create a 1D deconvolution network. The shape of the input is [-1, 256, 16] and the output should be [-1,1024,8]. The kernel's size is 5 and the stride is 4.

我尝试使用此功能构建一维卷积层:

I tried to build a 1D convolutional layer with this function:

    (output_depth, input_depth) = (8, 16)
    kernel_width = 7
    f_shape = [kernel_width, output_depth, input_depth]
    layer_1_filter = tf.Variable(tf.random_normal(f_shape))

    layer_1 = tf_exp.conv1d_transpose(
        x,
        layer_1_filter,
        [-1,1024,8],
        stride=4, padding="VALID"
    )

layer_1的形状为TensorShape([Dimension(None), Dimension(None), Dimension(None)])，但应为[-1,1024,8]

我怎么了?如何在Tensorflow中实现一维反卷积?

What do I wrong? How is it possible to implement 1D deconvolution in Tensorflow?

推荐答案

新的 tf.contrib.nn.conv1d_transpose 添加到 查看全文