This repository hosts a collection of machine learning projects implemented in PyTorch, covering object detection, classification, and more.
The object detection projects in this repository are inspired by the work of Aladdin Persson. You can find the original source code in the following GitHub repository:
- Repository: Machine-Learning-Collection
- Author: Aladdin Persson
I have adapted the object detection implementations from this repository as a foundation for our projects.
...
This repository is licensed under the MIT License.
This repository presents a variation of the YOLOv1 model where instead of height and width in bounding boxes, radius is used. All functions are based on detecting circles instead of rectangles. Additionally, a model has been trained for detecting apples, achieving a final mAP (mean Average Precision) of 25% over the range from 0.5 to 0.95.