depth_to_space

dragon.vm.tensorflow.nn.depth_to_space(
  input,
  block_size,
  data_format='NHWC',
  name=None
)[source]

Rearrange depth data into spatial blocks.

Examples:

n, h, w, c, bs = 1, 1, 1, 4, 2
x = tf.range(n * h * w * c).reshape((n, h, w, c))
y = tf.reshape(x, (n, h, w, bs, bs, c // (bs ** 2)))
y = tf.transpose(y, (0, 1, 3, 2, 4, 5))
y = tf.reshape(y, (n, h * bs, w * bs, c // (bs ** 2)))
z = tf.nn.depth_to_space(x, 2)  # Equivalent
Parameters:
  • input (dragon.Tensor) – The input tensor.
  • block_size (int, required) – The size of spatial block.
  • data_format ({'NCHW', 'NHWC'}, optional) – The optional data format.
  • name (str, optional) – A optional name for the operation.
Returns:

dragon.Tensor – The output tensor.