This page describes the data format you will need to use to train a YOLOv9 model. Note: YOLOv9 uses the same format as YOLOv7.
Roboflow supports converting 30+ different object detection annotation formats into the TXT format that YOLOv9 needs and we automatically generate your YAML config file for you. Plus, all 1,000,000+ datasets available on Roboflow Universe (opens in a new tab) are available in YOLOv7 format for seamless use in custom training.
Format Description
Below, learn the structure of YOLOv9 PyTorch TXT.
Each image has one txt file with a single line for each bounding box. The format of each row is
class_id center_x center_y width height
where fields are space delimited, and the coordinates are normalized from zero to one.
Note: To convert to normalized xywh from pixel values, divide x (and width) by the image's width and divide y (and height) by the image's height.