Skip to content

Latest commit

 

History

History
14 lines (9 loc) · 562 Bytes

File metadata and controls

14 lines (9 loc) · 562 Bytes

Indoor-RFDETR

Welcome to Indoor-RFDETR!
This repository contains scripts and utilities for building an end-to-end pipeline to train an RFDETR-Nano model on annotated indoor object datasets.

The provided scripts were used to:

  • Download annotated indoor object data
  • Convert annotations into COCO format
  • Train and evaluate an RFDETR-Nano object detection model

NOTE: Currently doesn't have automatically to balance classes, which could result in poorer training results. Updates *ideally coming soon!

Thanks for stopping by!