ngraph.psroi_pooling

ngraph.psroi_pooling(input: Union[_pyngraph.Node, int, float, numpy.ndarray], coords: Union[_pyngraph.Node, int, float, numpy.ndarray], output_dim: int, group_size: int, spatial_scale: float, spatial_bins_x: int, spatial_bins_y: int, mode: str, name: Optional[str] = None) _pyngraph.Node

Return a node which produces a PSROIPooling operation.

Parameters
  • input – Input feature map {N, C, …}

  • coords – Coordinates of bounding boxes

  • output_dim – Output channel number

  • group_size – Number of groups to encode position-sensitive scores

  • spatial_scale – Ratio of input feature map over input image size

  • spatial_bins_x – Numbers of bins to divide the input feature maps over

  • spatial_bins_y – Numbers of bins to divide the input feature maps over

  • mode – Mode of pooling - “avg” or “bilinear”

Returns

PSROIPooling node