mindspore.ops.unfold

View Source On Gitee
mindspore.ops.unfold(input, kernel_size, dilation=1, padding=0, stride=1)[source]

Extracts sliding local blocks from a batched input tensor.

Consider a batched input tensor of shape \((N, C, *)\), where \(N\) is the batch dimension, \(C\) is the channel dimension, and \(*\) represent arbitrary spatial dimensions. This operation flattens each sliding Kernel_size- sized block within the spatial dimensions of input x into a column (i.e., last dimension) of a 3-D output tensor of shape \((N, C \times \prod(\text{kernel_size}), L)\), where \(C \times \prod(\text{kernel_size})\) is the total number of values within each block (a block has \(\prod(\text{kernel_size})\) spatial locations each containing a C-channeled vector), and \(L\) is the total number of such blocks:

\[L = \prod_d \left\lfloor\frac{\text{spatial_size}[d] + 2 \times \text{pads}[d] % - \text{dilations}[d] \times (\text{kernel_size}[d] - 1) - 1}{\text{strides}[d]} + 1\right\rfloor,\]

where \(\text{spatial_size}\) is formed by the spatial dimensions of input x (\(*\) above), and \(d\) is over all spatial dimensions.

Therefore, indexing output at the last dimension (column dimension) gives all values within a certain block.

The dilation, padding and stride arguments specify how the sliding blocks are retrieved.

Warning

  • The output is a 3-dimensional Tensor whose shape is \((N, C \times \prod(\text{kernel_size}), L)\) .

  • This is an experimental API that is subject to change or deletion.

Parameters
  • input (Tensor) – 4-D Tensor, supported dtypes: float16, float32, float64, complex64 and complex128.

  • kernel_size (Union[int, tuple[int], list[int]]) – The size of the kernel, should be two int for height and width. If type is int, it means that height equal with width. Must be specified.

  • dilation (Union[int, tuple[int], list[int]], optional) – The dilation of the window, should be two int for height and width. If type is int, it means that height equal with width. Default: 1 .

  • padding (Union[int, tuple[int], list[int]], optional) –

    The pad of the window, that must be a tuple/list of one or two int for height and width. Default: 0 .

    • If one int, pad_height = pad_width.

    • If two int, pad_height = padding[0], pad_width = padding[1].

  • stride (Union[int, tuple[int], list[int]], optional) – The stride of the window, should be two int for height and width. If type is int, it means that height equal with width. Default: 1 .

Returns

A Tensor, with same type as input . And its shape is as described above.

Raises
  • TypeError – If any data type of kernel_size, stride, dilation, padding is not int, tuple or list.

  • ValueError – If kernel_size, dilation, stride value is not greater than zero or elements number more than 2.

  • ValueError – If padding value is less than zero.

Supported Platforms:

Ascend GPU CPU

Examples

>>> import mindspore
>>> import numpy as np
>>> from mindspore import Tensor, ops
>>> x = Tensor(np.random.rand(4, 4, 32, 32), mindspore.float64)
>>> output = ops.unfold(x, kernel_size=3, dilation=1, stride=1)
>>> print(output.shape)
(4, 36, 900)