Roboflow 100 Benchmarking Tutorial with Google Colab and Docker

Roboflow 100 (RF100) is a crowdsourced object detection benchmark. It consists of 100 datasets, 224,714 images, 804 label classes divided into 7 domains. RF100 should be used in tandem with COCO to measure models performance. This video is a step-by-step tutorial showing how to use RF100 and benchmark models in Google Colab and Docker. We use YOLOv5 and YOLOv7 as examples. Chapters: 0:00 WTF COCO 2:02 Roboflow 100 3:13 Plan of Attack 4:02 Google Colab YOLOv7 benchmark example 9:22 Docker YOLOv5 benchmark example 11:33 YOLOv5 vs YOLOv7 comparison 12:29 Outro Website: https://www.rf100.org/ Blog post: https://blog.roboflow.com/roboflow-100/ Github: https://github.com/roboflow-ai/roboflow-100-benchmark Paper: https://arxiv.org/abs/2211.13523 Papers with Code: https://paperswithcode.com/dataset/rf100 Google Colab Tutorial: https://colab.research.google.com/github/roboflow-ai/notebooks/blob/main/notebooks/how-to-use-rf100.ipynb Roboflow Notebooks: https://github.com/roboflow-ai/notebooks Explore the datasets: https://universe.roboflow.com/roboflow-100 Stay up to date with the projects I'm working on at https://github.com/roboflow-ai and https://github.com/SkalskiP! ⭐