Skip to content

DetectKit

DetectKit is the detection model training tool, launched via detectkit.

Purpose

Train, evaluate, and manage YOLO detection models for use in TrackerKit.

Launch

detectkit

Workflow

  1. Curate detection training datasets from TrackerKit exports or external sources.
  2. Configure YOLO training parameters (epochs, batch size, image size).
  3. Launch training and monitor loss curves.
  4. Evaluate model performance on validation sets.

Key Features

  • Dataset panel for assembling and inspecting training data
  • Training panel with configurable hyperparameters
  • Integration with TrackerKit's dataset generation exports