openvino.runtime.opset16.segment_max#

openvino.runtime.opset16.segment_max(data: Node | int | float | ndarray, segment_ids: Node | int | float | ndarray, num_segments: Node | int | float | ndarray | None = None, fill_mode: Literal['ZERO', 'LOWEST'] | None = None, name: str | None = None) → Node#

The SegmentMax operation finds the maximum value in each specified segment of the input tensor.

Parameters:

  • data – ND tensor of type T, the numerical data on which SegmentMax operation will be performed.

  • segment_ids – 1D Tensor of sorted non-negative numbers, representing the segments.

  • num_segments – An optional scalar value representing the segments count. If not provided, it is inferred from segment_ids.

  • fill_mode – Responsible for the value assigned to segments which are empty. Can be “ZERO” or “LOWEST”.

  • name – Optional name for the node.

Returns:

The new node performing SegmentMax operation.