mindspore.ops.unsorted_segment_min

mindspore.ops.unsorted_segment_min(x, segment_ids, num_segments)

Compute the minimum of the input tensor along segments.

The following figure shows the calculation process of unsorted_segment_min:

$${\text{ output }}_i={\text{min}}_{j\dots }\text{ data }[j\dots ]$$

where $min$ over tuples $j...$ such that $segment\mathrm{\_}ids[j...]==i$.

Note

• If the segment_id i is absent in the segment_ids, then output[i] will be filled with the maximum value of the x's type.

• The segment_ids must be non-negative tensor.

Parameters

• x (Tensor) – The input tensor.

• segment_ids (Tensor) – Indicate the segment to which each element belongs.

• num_segments (Union[int, Tensor], optional) – Number of segments, it can be an int or 0-D tensor.

Returns

Tensor

Supported Platforms:

Ascend GPU CPU

Examples

>>> import mindspore
>>> x = mindspore.tensor([[1, 2, 3], [4, 5, 6], [4, 2, 1]])
>>> segment_ids = mindspore.tensor([0, 1, 1])
>>> num_segments = 2
>>> mindspore.ops.unsorted_segment_min(x, segment_ids, num_segments)
Tensor(shape=[2, 3], dtype=Int64, value=
[[1, 2, 3],
 [4, 2, 1]])