data

YOLOConfig

extendsModelConfig

YOLOConfig(num_classes: int = 80, in_channels: int = 3, split_size: int = 7, num_boxes: int = 2, lambda_coord: float = 5.0, lambda_noobj: float = 0.5, score_thresh: float = 0.5, nms_thresh: float = 0.5, tiny: bool = False)

source edit

Configuration for YOLOv1 (Redmon et al., CVPR 2016).

Args: num_classes: Number of object classes C (COCO default: 80). in_channels: Input image channels. split_size: Grid cells per side S (default 7 → 7×7 grid). num_boxes: Bounding boxes predicted per cell B (default 2). lambda_coord: Up-weighting factor for box coordinate loss. lambda_noobj: Down-weighting factor for no-object confidence loss. score_thresh: Minimum class-confidence score at inference. nms_thresh: IoU threshold for NMS. tiny: Use the tiny Darknet backbone instead of the full one.

Examples

>>> from lucid.models.vision.yolo import YOLOConfig
>>> cfg = YOLOConfig()
>>> cfg.num_classes
80
>>> cfg.model_type
'yolo'

Used by 1

lucid.models.vision.yolo

Constructors

dunder

init

→None

__init__(num_classes: int = 80, in_channels: int = 3, split_size: int = 7, num_boxes: int = 2, lambda_coord: float = 5.0, lambda_noobj: float = 0.5, score_thresh: float = 0.5, nms_thresh: float = 0.5, tiny: bool = False)

source edit

data

YOLOConfig

extendsModelConfig

YOLOConfig(num_classes: int = 80, in_channels: int = 3, split_size: int = 7, num_boxes: int = 2, lambda_coord: float = 5.0, lambda_noobj: float = 0.5, score_thresh: float = 0.5, nms_thresh: float = 0.5, tiny: bool = False)

source edit

Configuration for YOLOv1 (Redmon et al., CVPR 2016).

Examples

>>> from lucid.models.vision.yolo import YOLOConfig
>>> cfg = YOLOConfig()
>>> cfg.num_classes
80
>>> cfg.model_type
'yolo'

Used by 1

lucid.models.vision.yolo

Constructors

dunder

init

→None

__init__(num_classes: int = 80, in_channels: int = 3, split_size: int = 7, num_boxes: int = 2, lambda_coord: float = 5.0, lambda_noobj: float = 0.5, score_thresh: float = 0.5, nms_thresh: float = 0.5, tiny: bool = False)

source edit