Enum Task 

pub enum Task {
    Detect,
    Segment,
    Pose,
    Classify,
    Obb,
    Semantic,
}

YOLO model task types.

Each task type corresponds to a different computer vision problem that YOLO models can solve. The task type determines the expected model outputs and post-processing steps.

Variants

Detect

Object detection. Predicts bounding boxes and class labels for objects in an image.

Segment

Instance segmentation. Predicts bounding boxes, class labels, and pixel-level masks for objects.

Pose

Pose estimation. Predicts bounding boxes and skeletal keypoints for objects (e.g., humans).

Classify

Image classification. Predicts class probabilities for the entire image (no localization).

Obb

Oriented bounding box detection (OBB). Predicts rotated bounding boxes for objects, useful for aerial imagery etc.

Semantic

Semantic segmentation. Assigns a class label to every pixel in the image.

Implementations

impl Task

pub const fn as_str(&self) -> &'static str

Get the string representation used in ONNX model metadata (e.g. "detect", "segment").

pub const fn model_suffix(&self) -> &'static str

ONNX filename suffix for this task, used to construct {family}n{suffix}.onnx (e.g. yolo26n-seg.onnx, yolo11n-pose.onnx, yolov8n.onnx).

use ultralytics_inference::Task;
assert_eq!(Task::Detect.model_suffix(), "");
assert_eq!(Task::Segment.model_suffix(), "-seg");

pub fn default_model(&self) -> String

Default nano YOLO26 model filename for this task.

Used by the CLI to auto-pick a model when --model is omitted but --task is set. YOLO26, YOLO11, and YOLOv8 variants are all auto-downloadable.

use ultralytics_inference::Task;
assert_eq!(Task::Detect.default_model(), "yolo26n.onnx");
assert_eq!(Task::Segment.default_model(), "yolo26n-seg.onnx");

pub const fn has_boxes(&self) -> bool

Returns true when the task outputs bounding boxes: Detect, Segment, Pose, and Obb.

pub const fn has_masks(&self) -> bool

Returns true only for the Segment task, which outputs per-instance segmentation masks.

pub const fn has_keypoints(&self) -> bool

Returns true only for the Pose task, which outputs skeletal keypoints.

pub const fn has_probs(&self) -> bool

Returns true only for the Classify task, which outputs global class probabilities.

pub const fn has_obb(&self) -> bool

Returns true only for the Obb task, which outputs oriented (rotated) bounding boxes.

pub const fn has_semantic_mask(&self) -> bool

Returns true only for the Semantic task, which outputs a per-pixel class label map.

Trait Implementations

impl Clone for Task

fn clone(&self) -> Task

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Copy for Task

impl Debug for Task

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for Task

fn default() -> Task

Returns the “default value” for a type.

impl Display for Task

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Eq for Task

impl FromStr for Task

type Err = TaskParseError

The associated error which can be returned from parsing.

fn from_str(s: &str) -> Result<Self, Self::Err>

Parses a string s to return a value of this type.

impl Hash for Task

fn hash<__H: Hasher>(&self, state: &mut __H)

Feeds this value into the given Hasher.

fn hash_slice<H>(data: &[Self], state: &mut H)

where H: Hasher, Self: Sized,

Feeds a slice of this type into the given Hasher.

impl PartialEq for Task

fn eq(&self, other: &Task) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for Task