Instructions to use orasul/deki-yolo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ultralytics
How to use orasul/deki-yolo with ultralytics:
from huggingface_hub import hf_hub_download from ultralytics import YOLO # pick the weights file from this repo's "Files and versions" tab weights = hf_hub_download("orasul/deki-yolo", "<weights>.pt") model = YOLO(weights) source = 'http://images.cocodataset.org/val2017/000000039769.jpg' model.predict(source=source, save=True) - Notebooks
- Google Colab
- Kaggle
Dataset
#1
by patricklef - opened
Hi, really cool work on this. I’m building something similar and was curious to extend the dataset and train an RT-DETR. If the dataset published somewhere?
Hi, unfortunately dataset is private, but I also have plan to extend it and use newest OT models. Will share it here too.
Thanks
Have you tried with RT-DETR? Might work well.
What’s your process for collecting training data? Manual labeling or fetching it from device.