# fastsam Python API: `depthai_nodes.node.parsers.fastsam` ## Classes ### FastSAMParser Parser class for parsing the output of the FastSAM model. > **Note** > Emits `dai.SegmentationMask` messages. dai.SegmentationMask message containing the resulting segmentation masks given the prompt. #### Methods ##### init ```python def __init__(conf_threshold: float = 0.5, n_classes: int = 1, iou_threshold: float = 0.5, mask_conf: float = 0.5, prompt: str = 'everything', points: tuple[int, int] | None = None, point_label: int | None = None, bbox: tuple[int, int, int, int] | None = None, yolo_outputs: list[str] = None, mask_outputs: list[str] = None, protos_output: str = 'protos_output'): ``` Initializes the parser node. Parameters * `conf_threshold` (`float`): The confidence threshold for the detections * `n_classes` (`int`): The number of classes in the model * `iou_threshold` (`float`): The intersection over union threshold * `mask_conf` (`float`): The mask confidence threshold * `prompt` (`str`): The prompt type * `points` (`tuple[int, int] | None`): The points * `point_label` (`int | None`): The point label * `bbox` (`tuple[int, int, int, int] | None`): The bounding box * `yolo_outputs` (`list[str]`): The YOLO outputs * `mask_outputs` (`list[str]`): The mask outputs * `protos_output` (`str`): The protos output ##### build ```python def build(head_config: dict[str, Any]) -> FastSAMParser: ``` Configures the parser. Parameters * `head_config` (`dict[str, Any]`): The head configuration for the parser. Returns * `FastSAMParser`: The parser object with the head configuration set. ##### compute ```python def compute(outputs_values: list[np.ndarray], masks_outputs_values: list[np.ndarray], protos_output: np.ndarray, protos_len: int, *, conf_threshold: float, n_classes: int, iou_threshold: float, mask_conf: float, prompt: str, points: tuple[int, int] | None, point_label: int | None, bbox: tuple[int, int, int, int] | None) -> tuple[np.ndarray, int]: ``` Compute parser results from extracted tensors without sending messages. > **Note** > Uses [depthai_nodes.node.parsers.utils.fastsam.compute_fastsam_mask](https://docs.luxonis.com/software-v3/ai-inference/inference/depthai-nodes/depthai-nodes-api-reference/node/parsers/utils/fastsam.md); see that helper for tensor layout and validation details. Parameters * `outputs_values` (`list[np.ndarray]`): Detection tensors ordered by output head. * `masks_outputs_values` (`list[np.ndarray]`): Mask coefficient tensors ordered to match the detection heads. * `protos_output` (`np.ndarray`): Batched prototype tensor with shape `(1, channels, height, width)`. * `protos_len` (`int`): Number of prototype channels used by each mask coefficient vector. * `conf_threshold` (`float`): Minimum detection confidence used to filter candidates. * `n_classes` (`int`): Number of object classes encoded in the detection tensors. * `iou_threshold` (`float`): Intersection-over-union threshold for non-maximum suppression. * `mask_conf` (`float`): Probability threshold used to binarize mask logits. * `prompt` (`str`): Mask selection mode: `"everything"`, `"bbox"`, or `"point"`. * `points` (`tuple[int, int] | None`): Prompt point in image pixel coordinates for point selection. * `point_label` (`int | None`): Point-prompt label used to include or exclude matching masks. * `bbox` (`tuple[int, int, int, int] | None`): Bounding-box prompt in image pixel coordinates. Returns * `tuple[np.ndarray, int]`: A pair of the merged instance mask and the number of selected masks. No selected masks produces a background mask with value -1 and count 0. ##### emit ```python def emit(output: dai.NNData, results_masks: np.ndarray, mask_count: int): ``` Create a `dai.SegmentationMask` 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. * `results_masks` (`np.ndarray`): Merged instance mask returned by `compute()`. * `mask_count` (`int`): Number of selected masks, used for logging. ##### extract ```python def extract(output: dai.NNData) -> tuple[list[np.ndarray], list[np.ndarray], np.ndarray, int]: ``` 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 * `tuple[list[np.ndarray], list[np.ndarray], np.ndarray, int]`: Detection heads in lexical layer-name order, mask coefficient heads, prototype tensor, and prototype channel count; arrays are dequantized float32 in NCHW order. ##### 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. ##### setBoundingBox ```python def setBoundingBox(bbox: tuple[int, int, int, int]): ``` Sets the bounding box. Parameters * `bbox` (`tuple[int, int, int, int]`): The bounding box ##### setConfidenceThreshold ```python def setConfidenceThreshold(threshold: float): ``` Sets the confidence score threshold. Parameters * `threshold` (`float`): Confidence score threshold. ##### setIouThreshold ```python def setIouThreshold(iou_threshold: float): ``` Sets the intersection over union threshold. Parameters * `iou_threshold` (`float`): The intersection over union threshold. ##### setMaskConfidence ```python def setMaskConfidence(mask_conf: float): ``` Sets the mask confidence threshold. Parameters * `mask_conf` (`float`): The mask confidence threshold. ##### setMaskOutputs ```python def setMaskOutputs(mask_outputs: list[str]): ``` Sets the mask outputs. Parameters * `mask_outputs` (`list[str]`): The mask outputs ##### 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. ##### setPointLabel ```python def setPointLabel(point_label: int): ``` Sets the point label. Parameters * `point_label` (`int`): The point label ##### setPoints ```python def setPoints(points: tuple[int, int]): ``` Sets the points. Parameters * `points` (`tuple[int, int]`): The points ##### setPrompt ```python def setPrompt(prompt: str): ``` Sets the prompt type. Parameters * `prompt` (`str`): The prompt type ##### setProtosOutput ```python def setProtosOutput(protos_output: str): ``` Sets the protos output. Parameters * `protos_output` (`str`): The protos output ##### setYoloOutputs ```python def setYoloOutputs(yolo_outputs: list[str]): ``` Sets the YOLO outputs. Parameters * `yolo_outputs` (`list[str]`): The YOLO outputs #### Attributes ##### bbox Bounding box. ##### conf_threshold Confidence score threshold for detected objects. ##### iou_threshold Non-maximum suppression threshold. ##### mask_conf Mask confidence threshold. ##### mask_outputs Names of the mask outputs. ##### n_classes Number of classes in the model. ##### point_label Point label. ##### points Points. ##### prompt Prompt type. ##### protos_output Name of the protos output. ##### yolo_outputs Names of the YOLO outputs.