0

我想用它tf.nn.conv2d_transpose来为 GAN 网络构建一个反卷积层。

我想创建一个函数deconv_layer。它生成一个新层,该层输出的filter_num过滤器expand_size的分辨率是输入的倍数。

我的代码是:

def deconv_layer(x, filter_num, kernel_size=5, expand_size=2):

    x_shape = x.get_shape().as_list()

    with tf.name_scope('deconv_'+str(filter_num)):

        size_in = x_shape[-1]
        size_out = filter_num

        w = tf.Variable(tf.random_normal([kernel_size, kernel_size, size_in, size_out], mean=0.0, stddev=0.125), name="W")
        b = tf.Variable(tf.random_normal([size_out], mean=0.0, stddev=0.125), name="B")

        conv = tf.nn.conv2d_transpose(x, w, output_shape=[-1, x_shape[-3]*expand_size, x_shape[-2]*expand_size, filter_num], strides=[1,expand_size,expand_size,1], padding="SAME")
        act = tf.nn.relu(tf.nn.bias_add(conv, b))

        tf.summary.histogram('weights', w)
        tf.summary.histogram('biases', b)
        tf.summary.histogram('activations', act)

    return act

错误信息:

ValueError: input channels does not match filter's input channels
At conv = tf.nn.conv2d_transpose(...)

我不确定我tf.nn.conv2d_transpose是否正确使用。我尝试基于卷积层创建它。

4

1 回答 1

1

过滤器尺寸错误。根据文档

filter:一个 4-D 张量,其类型与 value 和 shape [height, width, output_channels, in_channels] 相同。过滤器的 in_channels 维度必须与值(输入)的维度相匹配。

您需要将w尺寸更改为:

w = tf.Variable(tf.random_normal([kernel_size, kernel_size, size_out, size_in], mean=0.0, stddev=0.125), name="W")
于 2018-05-19T12:42:36.793 回答