How to train YOLOv9 Object Detection on Custom Dataset | step by step Tutorial | Google Colab
On February 21st, 2024, Chien-Yao Wang, I-Hau Yeh, and Hong-Yuan Mark Liao released the latest installation in the YOLO series “YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information'' paper.
YOLOv9 is released in four models, ordered by parameter count: v9-S, v9-M, v9-C, and v9-E. To improve accuracy, it introduces programmable gradient information (PGI) and the Generalized Efficient Layer Aggregation Network (GELAN). PGI prevents data loss and ensures accurate gradient updates and GELAN optimizes lightweight models with gradient path planning.
Concepts covered:
How to annotate data for YOLO
How to create a train-val split
How to create the dataset.yaml file
How to train or fine-tune yolov9 model on custom data
How to do inferencing on images and videos
How to convert the model from PyTorch to ONNX and TensorRT or trt
How to do inferencing using the yolov9 TensorRT model
YOLOv9 Github: https://github.com/WongKinYiu/yolov9
YOLOv9 research paper: https://arxiv.org/abs/2402.13616
Data: https://www.kaggle.com/datasets/deepa...
How to annotate data: • How to annotate images for object detection
Special Thanks to:
Roboflow: • YOLOv9 Tutorial: Train Model on Custom Dat...
Altaf Shah: https://www.pexels.com/video/drone-fo...
Altaf Shah: https://www.pexels.com/video/drone-fo...
Recommended books:
1. HANDS ON MACHINE LEARNING WITH SCIKIT LEARN, KERAS & TENSORFLOW: https://amzn.to/3wkAqeG
2. Deep Learning with PyTorch: Build, train, and tune neural networks using Python tools: https://amzn.to/3QX5ivo
Recommended laptops for Machine Learning and Deep Learning:
1. Lenovo Legion 5 AMD Ryzen 5 4600H 15: https://amzn.to/3XSGEOP
2. Acer Nitro 5: https://amzn.to/3kE3oU7
3. Acer Predator Helios 300: https://amzn.to/3XvAkgi
Best budget setup:
1. BenQ GW2480 24-inch (60.5 cm) 1080p monitor: https://amzn.to/3iY6oKO
2. Office Chair: https://amzn.to/3XQEK0Z
3. Table: https://amzn.to/3R2b1QI
4. Mic: https://amzn.to/3Wv4leT