For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Mojo function

non_maximum_suppression_shape

def non_maximum_suppression_shape[dtype: DType](boxes: ManagedTensorSlice[IOSpec[_, _].Input, static_spec=boxes.static_spec], scores: ManagedTensorSlice[IOSpec[_, _].Input, static_spec=scores.static_spec], max_output_boxes_per_class: Int64, iou_threshold: Float32, score_threshold: Float32) -> IndexList[Int(2)]

Computes the output shape for the mo.non_maximum_suppression graph op.

Parameters:

• ​dtype (DType): Element type of the boxes and scores tensors.

Args:

• ​boxes (ManagedTensorSlice[IOSpec[_, _].Input, static_spec=boxes.static_spec]): Rank-3 tensor of bounding boxes with shape (batch, num_boxes, 4) where each box is [y1, x1, y2, x2].
• ​scores (ManagedTensorSlice[IOSpec[_, _].Input, static_spec=scores.static_spec]): Rank-3 tensor of scores with shape (batch, num_classes, num_boxes).
• ​max_output_boxes_per_class (Int64): Maximum number of boxes to select per class.
• ​iou_threshold (Float32): Intersection-over-union threshold for suppression; boxes with IoU above this value are suppressed.
• ​score_threshold (Float32): Minimum score for a box to be considered; boxes with score below this value are filtered out.

Returns:

IndexList[Int(2)]: Two-element IndexList of shape (num_selected_boxes, 3).