# yolo Python API: `depthai_nodes.node.parsers.yolo` ## Classes ### YOLOComputeInputs Immutable configuration and tensor bundle for YOLO decoding. Created by `YOLOExtendedParser.extract()` and consumed by `compute()`. Fields correspond to arguments of [depthai_nodes.node.parsers.utils.yolo.compute_yolo_detections](https://docs.luxonis.com/software-v3/ai-inference/inference/depthai-nodes/depthai-nodes-api-reference/node/parsers/utils/yolo.md). Optional pose and mask fields are populated only for the selected model mode. Array contents remain mutable even though the dataclass fields cannot be reassigned. #### Attributes ##### anchors ##### conf_threshold ##### input_shape ##### iou_threshold ##### keypoint_edges ##### keypoint_label_names ##### kpts_outputs ##### label_names ##### layer_names ##### mask_conf ##### masks_outputs_values ##### max_det ##### n_classes ##### n_keypoints ##### outputs_values ##### protos_len ##### protos_output ##### strides ##### subtype ##### v26_mask_coeffs ##### v26_pose_kpts ##### v26_protos ### YOLOExtendedParser Parser class for parsing the output of the YOLO Instance Segmentation and Pose Estimation models. > **Note** > Emits `dai.ImgDetections` messages. dai.ImgDetections message containing bounding boxes, labels, label names, confidence scores, and keypoints or masks and protos of the detected objects. #### Methods ##### init ```python def __init__(conf_threshold: float = 0.5, n_classes: int = 1, label_names: list[str] | None = None, iou_threshold: float = 0.5, mask_conf: float = 0.5, n_keypoints: int = 17, max_det: int = 300, anchors: list[list[list[float]]] | None = None, subtype: str = '', keypoint_label_names: list[str] | None = None, keypoint_edges: list[tuple[int, int]] | None = None): ``` Initialize the parser node. Parameters * `conf_threshold` (`float`): The confidence threshold for the detections * `n_classes` (`int`): The number of classes in the model * `label_names` (`list[str] | None`): The names of the classes * `iou_threshold` (`float`): The intersection over union threshold * `mask_conf` (`float`): The mask confidence threshold * `n_keypoints` (`int`): The number of keypoints in the model * `max_det` (`int`): Maximum number of detections to retain. * `anchors` (`list[list[list[float]]] | None`): The anchors for the YOLO model * `subtype` (`str`): The version of the YOLO model * `keypoint_label_names` (`list[str] | None`): The labels for the keypoints * `keypoint_edges` (`list[tuple[int, int]] | None`): Connection pairs of the keypoints. Example: [(0,1), (1,2), (2,3), (3,0)] shows that keypoint 0 is connected to keypoint 1, keypoint 1 is connected to keypoint 2, etc. ##### build ```python def build(head_config: dict[str, Any]): ``` Configures the parser. Parameters * `head_config` (`dict[str, Any]`): The head configuration for the parser. Returns * `YOLOExtendedParser`: The parser object with the head configuration set. ##### compute ```python def compute(inputs: YOLOComputeInputs) -> dict[str, Any]: ``` Decode a prepared YOLO input bundle. > **Note** > Uses [depthai_nodes.node.parsers.utils.yolo.compute_yolo_detections](https://docs.luxonis.com/software-v3/ai-inference/inference/depthai-nodes/depthai-nodes-api-reference/node/parsers/utils/yolo.md) for decoding and validation. Parameters * `inputs` (`YOLOComputeInputs`): Tensor and configuration bundle returned by `extract()`. Returns * `dict[str, Any]`: A dictionary containing `mode` (0 detection, 1 pose, 2 segmentation), normalized `bboxes`, `scores`, `labels`, `label_names`, `keypoints`, `keypoints_scores`, `keypoint_label_names`, `keypoint_edges`, and `masks`. Boxes use center-XY/width/height. A segmentation mask contains int16 detection indexes and -1 background; other modes return `None` for masks. ##### emit ```python def emit(output: dai.NNData, payload: dict[str, Any]): ``` Create a `dai.ImgDetections` message and send it on `out`. Copies source timestamps and sequence number, and carries the source image transformation when present. Parameters * `output` (`dai.NNData`): Neural network output carrying tensors and source timestamps, sequence number, and optional image transformation. * `payload` (`dict[str, Any]`): Dictionary returned by `compute()` with the mode-specific boxes, scores, labels, keypoints, and masks. ##### extract ```python def extract(output: dai.NNData) -> YOLOComputeInputs: ``` Select and dequantize the model tensors needed for parsing. Parameters * `output` (`dai.NNData`): Neural network output carrying tensors and source timestamps, sequence number, and optional image transformation. Returns * `YOLOComputeInputs`: A `YOLOComputeInputs` bundle with dequantized tensors, resolved image geometry, strides, and current parser settings. Raises * `ValueError`: If YOLO26 has no configured input shape, or head strides cannot be resolved. ##### run ```python def run(self): ``` Read queued network outputs, parse them, and emit results while running. The pipeline invokes this processing loop. It exits when the input queue closes or the node stops. ##### setAnchors ```python def setAnchors(anchors: list[list[list[float]]]): ``` Sets the anchors for the YOLO model. Parameters * `anchors` (`list[list[list[float]]]`): The anchors for the YOLO model. ##### setConfidenceThreshold ```python def setConfidenceThreshold(threshold: float): ``` Sets the confidence score threshold for detected objects. Parameters * `threshold` (`float`): Confidence score threshold for detected objects. ##### setIouThreshold ```python def setIouThreshold(iou_threshold: float): ``` Sets the intersection over union threshold. Parameters * `iou_threshold` (`float`): The intersection over union threshold. ##### setKeypointEdges ```python def setKeypointEdges(keypoint_edges: list[tuple[int, int]]): ``` Sets the edges for the keypoints. Parameters * `keypoint_edges` (`list[tuple[int, int]]`): The edges for the keypoints. ##### setKeypointLabelNames ```python def setKeypointLabelNames(keypoint_label_names: list[str]): ``` Sets the label names for the keypoints. Parameters * `keypoint_label_names` (`list[str]`): The labels for the keypoints. ##### setLabelNames ```python def setLabelNames(label_names: list[str]): ``` Sets the names of the classes. Parameters * `label_names` (`list[str]`): The names of the classes. ##### setMaskConfidence ```python def setMaskConfidence(mask_conf: float): ``` Sets the mask confidence threshold. Parameters * `mask_conf` (`float`): The mask confidence threshold. ##### setNumClasses ```python def setNumClasses(n_classes: int): ``` Sets the number of classes in the model. Parameters * `n_classes` (`int`): The number of classes in the model. ##### setNumKeypoints ```python def setNumKeypoints(n_keypoints: int): ``` Sets the number of keypoints in the model. Parameters * `n_keypoints` (`int`): The number of keypoints in the model. ##### setOutputLayerNames ```python def setOutputLayerNames(output_layer_names: list[str]): ``` Sets the output layer names for the parser. Parameters * `output_layer_names` (`list[str]`): The output layer names for the parser. ##### setStrides ```python def setStrides(strides: list[int] | tuple[int, ...]): ``` Sets the strides for YOLO output heads. Parameters * `strides` (`list[int] | tuple[int, ...]`): Strides for YOLO output heads. ##### setSubtype ```python def setSubtype(subtype: str): ``` Sets the subtype of the YOLO model. Parameters * `subtype` (`str`): The subtype of the YOLO model. #### Attributes ##### anchors Anchors for the YOLO model (optional). ##### conf_threshold Confidence score threshold for detected objects. ##### input_shape ##### iou_threshold Intersection over union threshold. ##### keypoint_edges Pairs of keypoint indexes defining skeleton edges. For example, `[(0, 1), (1, 2)]` connects keypoint 0 to 1 and 1 to 2. ##### keypoint_label_names Labels for the keypoints. ##### label_names Names of the classes. ##### mask_conf Mask confidence threshold. ##### max_det ##### n_classes Number of classes in the model. ##### n_keypoints Number of keypoints in the model. ##### n_prototypes ##### output_layer_names ##### strides Strides for the YOLO output heads. ##### subtype Version of the YOLO model.