liveBook · Manning

Ultralytics YOLO format

As explained in Machine Learning Platform Engineering

The Ultralytics YOLO format is a dataset organization and annotation convention used by Ultralytics YOLO models for object detection.

Ultralytics YOLO format organizes images, labels, and dataset configuration for training an object detector such as YOLOv8. Prepare the data in this layout before wiring it into a repeatable data-preparation pipeline. section 7.3.1

Keep the configuration and files aligned:

  • Set path to the dataset root; point train, val, and test to their image directories. Each split also needs a corresponding labels directory. section 7.3.1
  • Match image and label basenames: images/id0042.png pairs with labels/id0042.txt. Map zero-indexed class numbers to labels under names. section 7.3.1
Listing 7.9 Configuration training YOLOv8 with a custom dataset
path: /home/jovyan/data/
train: "train/images"
val: "val/images"
test: "test/images"
names:
  0: id_card

This YOLOv8 configuration points at all three image splits and maps class 0 to id_card. section 7.3.1

When automating the split, copy images and their labels into matching split directories. Check the expected layout before implementation: detector formats do not share one universal configuration or naming convention. section 7.3.1

How Machine Learning Platform Engineering covers it

How deep it goes
Specifies configuration fields, directories, and filename pairing but does not detail annotation contents or conversion.
Examples
Uses an ID-card class, a YOLOv8 configuration, and an image–label filename pair.
What it assumes
Assumes a converted dataset while explaining the configuration and split layout needed to train with it.
What you can do after
Operate/configure it by aligning split directories, image–label pairs, and class indices for YOLOv8 training.
How it connects
Connects object detection, dataset splitting, class labels, and Kubeflow components.
Cover of Machine Learning Platform Engineering
liveBookmanning.com
Read Machine Learning Platform Engineering on liveBook Get the book at manning.com

This explainer was generated by Manning's knowledge graph from the text of Machine Learning Platform Engineering. The section links point to the parts of the book it draws on.