mindspore.ops.col2im

mindspore.ops.col2im(input_x, output_size, kernel_size, dilation, padding_value, stride)[source]

Combines an array of sliding local blocks into a large containing tensor.

Parameters

  • input_x (Tensor) – 4D tensor with data type float16 or float.

  • output_size (Tensor) – 1D tensor with 2 elements of data type int.

  • 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]]) – The size of the dilation, should be two int for height and width. If type is int, it means that height equal with width. Default: 1.

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

  • stride (Union[int, tuple[int], list[int]]) – The size of the stride, should be two int for height and width. If type is int, it means that height equal with width. Default: 0.

Returns

A 4D Tensor, with same type as ‘input_x’.

Raises

  • TypeError – If kernel_size, dilation, padding_value, stride data type is not in Union[int, tuple[int], list[int]].

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

  • ValueError – If padding_value value is less than zero or elements number more than 2.

  • ValueError – If input_x.shape[2] != kernel_size[0] * kernel_size[1].

  • ValueError – If input_x.shape[3] does not match the calculated number of sliding blocks.

Supported Platforms:

GPU

Examples

>>> x = Tensor(input_data=np.random.rand(16, 16, 4, 25), dtype=mstype.float32)
>>> output_size = Tensor(input_data=[8, 8], dtype=mstype.int32)
>>> output = ops.col2im(x, output_size, [2, 2], [2, 2], [2, 2], [2, 2])
>>> print(output.shape)
(16, 16, 8, 8)