转置卷积（反卷积）算术

1 回答

• 1

  是的，你几乎是对的 .

  一个小的修正是 tf.nn.conv3d_transpose 期望 NCDHW 或 NDHWC 输入格式（你的似乎是 NHWDC ），滤镜形状应该是 [depth, height, width, output_channels, in_channels] . 这会影响 filter 和 stride 中的维度顺序：

  # Original format: NHWDC.
  original = tf.placeholder(dtype=tf.float32, shape=[None, 16, 16, 4, 192])
  print original.shape
  
  # Convert to NDHWC format.
  input = tf.reshape(original, shape=[-1, 4, 16, 16, 192])
  print input.shape
  
  # input shape:  [batch, depth, height, width, in_channels].
  # filter shape: [depth, height, width, output_channels, in_channels].
  # output shape: [batch, depth, height, width, output_channels].
  filter = tf.get_variable('filter', shape=[4, 2, 2, 192, 192], dtype=tf.float32)
  conv = tf.nn.conv3d_transpose(input,
                                filter=filter,
                                output_shape=[-1, 7, 32, 32, 192],
                                strides=[1, 1, 2, 2, 1],
                                padding='SAME')
  print conv.shape
  
  final = tf.reshape(conv, shape=[-1, 32, 32, 7, 192])
  print final.shape
  

  哪个输出：

  (?, 16, 16, 4, 192)
  (?, 4, 16, 16, 192)
  (?, 7, 32, 32, 192)
  (?, 32, 32, 7, 192)
  

  回复于 2024-04-25T16:48:19+08:00