ml-pipes-ultralytics

ml-pipes-ultralytics provides Ultralytics YOLO prediction, embedding, tracking, and result operations as composable ml-pipes operators. Native Ultralytics models and Results stay intact, while inference configuration and pipeline data flow are explicit.

Install from PyPI in a Python 3.10+ environment:

python -m pip install ml-pipes-ultralytics

The package installs ml-pipes-core, ml-pipes-vision, and Ultralytics. See the Reference for the public operator surface and Coverage for the native Ultralytics API comparison.

Start with the runnable examples:

from ml_pipes.core import Pipeline
from ml_pipes.standard import Select
from ml_pipes.ultralytics import results, yolo

pipeline = Pipeline(
    [
        yolo.Predict("yolo26n-seg.pt", conf=0.25),
        Select(0),
        results.Plot(color_mode="instance"),
    ]
)

Licensing

ml-pipes-ultralytics is licensed under the Apache License 2.0. It requires Ultralytics, which is licensed separately under AGPL-3.0 or an Ultralytics Enterprise License. Installing or using this package does not grant rights to Ultralytics software or model weights. Community Ultralytics users must comply with AGPL-3.0; Enterprise users must ensure their Ultralytics agreement covers their intended use.

Built with Ultralytics x ml-pipes

Example	Upstream source	Note
run_detect_and_crop.py	Object Cropping	Detects objects, saves native result crops, and renders the annotated image.
run_segment_and_annotate.py	Instance Segmentation and Tracking	Runs segmentation and renders instance-coloured masks.
run_object_blurrer.py	Object Blurring	Tracks and blurs COCO person detections.
run_detection_video.py	Ultralytics predict mode	Uses an explicit OpenCV capture loop and one non-streaming prediction pipeline call per frame.
run_track_objects.py	Ultralytics track mode	Preserves native tracker state and draws native tracking IDs plus explicit motion traces.