# BNNS.LossFunction.YoloParameters A structure that contains the parameters for You Only Look Once (YOLO) loss computation. ``` struct YoloParameters ``` ## Topics ### Instance Properties [`let anchorBoxCount: Int`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/anchorBoxCount) The number of anchor boxes in each cell. [`let anchorBoxSize: Int`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/anchorBoxSize) The size of the anchor box. [`let anchorsData: UnsafeMutablePointer`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/anchorsData) Maximum IOU for treating as no object. [`let classificationScale: Float`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/classificationScale) The value that specifies the classification scaling factor. [`let gridColumnCount: Int`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/gridColumnCount) The number of columns in the grid. [`let gridRowsCount: Int`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/gridRowsCount) The number of rows in the grid. [`let huberDelta: Float`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/huberDelta) A value that’s interpreted as width-height loss. [`let noObjectMaximumIoU: Float`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/noObjectMaximumIoU) The value that specifies intersection over union (IOU) that’s the maximum the function treats as not an object. [`let noObjectScale: Float`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/noObjectScale) The value that specifies the no-object confidence scaling factor. [`let objectMinimumIoU: Float`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/objectMinimumIoU) The value that specifies intersection over union (IOU) that’s the minimum the function treats as an object. [`let objectScale: Float`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/objectScale) The value that specifies the object confidence loss-scaling factor. [`let rescore: Bool`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/rescore) A Boolean value that determines whether to rescore confidence according to prediction verus ground truth Intersection Over Union (IOU). [`let whScale: Float`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/whScale) A Boolean value that determines whether to rescore confidence according to prediction verus ground truth Intersection Over Union (IOU). [`let xyScale: Float`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/xyScale) The value that specifies the x, y loss-scaling factor. ### Initializers [`init(huberDelta: Float, gridColumnCount: Int, gridRowsCount: Int, anchorBoxCount: Int, anchorBoxSize: Int, rescore: Bool, xyScale: Float, whScale: Float, objectScale: Float, noObjectScale: Float, classificationScale: Float, objectMinimumIoU: Float, noObjectMaximumIoU: Float, anchorsData: UnsafeMutablePointer)`](/documentation/Accelerate/BNNS/LossFunction/YoloParameters/init(huberDelta:gridColumnCount:gridRowsCount:anchorBoxCount:anchorBoxSize:rescore:xyScale:whScale:objectScale:noObjectScale:classificationScale:objectMinimumIoU:noObjectMaximumIoU:anchorsData:)) --- Copyright © 2026 Apple Inc. All rights reserved. | [Terms of Use](https://www.apple.com/legal/internet-services/terms/site.html) | [Privacy Policy](https://www.apple.com/privacy/privacy-policy)