# tensorplay.functional.segment_reduce

Source: https://www.tensorplay.cn/docs/generated/tensorplay.functional.segment_reduce.html

```python
tensorplay.functional.segment_reduce(data: Tensor, reduce: str, *, lengths: Tensor | None = None, indices: Tensor | None = None, offsets: Tensor | None = None, axis: _int = 0, unsafe: _bool = False, initial: Number | _complex | None = None) → Tensor
```

Perform a segment reduction operation on the input tensor along the specified axis.

Parameters:

- data ([Tensor](/docs/generated/tensorplay.Tensor.html#tensorplay.Tensor)) – The input tensor on which the segment reduction operation will be performed.

- reduce ([str](https://docs.python.org/3/builtins/stdtypes.html#str)) – The type of reduction operation. Supported values are sum, mean, max, min, prod.

Keyword Arguments:

- lengths ([Tensor](/docs/generated/tensorplay.Tensor.html#tensorplay.Tensor), optional) – Length of each segment. Default: None.

- offsets ([Tensor](/docs/generated/tensorplay.Tensor.html#tensorplay.Tensor), optional) – Offset of each segment. Default: None.

- axis ([int](https://docs.python.org/3/builtins/functions.html#int), optional) – The axis perform reduction. Default: 0.

- unsafe ([bool](https://docs.python.org/3/builtins/functions.html#bool), optional) – Skip validation If True. Default: False.

- initial (Number, optional) – The initial value for the reduction operation. Default: None.

Example:

```
>>> data = tensorplay.tensor([[1, 2, 3, 4],[5, 6, 7, 8],[9, 10, 11, 12]], dtype=tensorplay.float32, device='cuda')
>>> lengths = tensorplay.tensor([2, 1], device='cuda')
>>> tensorplay.segment_reduce(data, 'max', lengths=lengths)
tensor([[ 5.,  6.,  7.,  8.],
        [ 9., 10., 11., 12.]], device='cuda:0')
```