-
-
Notifications
You must be signed in to change notification settings - Fork 15.8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Extremely low precision but high mAP #12940
Comments
👋 Hello @eric80739, thank you for your interest in YOLOv5 🚀! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution. If this is a 🐛 Bug Report, please provide a minimum reproducible example to help us debug it. If this is a custom training ❓ Question, please provide as much information as possible, including dataset image examples and training logs, and verify you are following our Tips for Best Training Results. RequirementsPython>=3.8.0 with all requirements.txt installed including PyTorch>=1.8. To get started: git clone https://github.com/ultralytics/yolov5 # clone
cd yolov5
pip install -r requirements.txt # install EnvironmentsYOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):
StatusIf this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training, validation, inference, export and benchmarks on macOS, Windows, and Ubuntu every 24 hours and on every commit. Introducing YOLOv8 🚀We're excited to announce the launch of our latest state-of-the-art (SOTA) object detection model for 2023 - YOLOv8 🚀! Designed to be fast, accurate, and easy to use, YOLOv8 is an ideal choice for a wide range of object detection, image segmentation and image classification tasks. With YOLOv8, you'll be able to quickly and accurately detect objects in real-time, streamline your workflows, and achieve new levels of accuracy in your projects. Check out our YOLOv8 Docs for details and get started with: pip install ultralytics |
Hi Eric! 👋 It's great that you're diving into training YOLOv5n, even with a small dataset. The peculiar case of low precision yet high mAP often indicates a model might be leaning towards making many detections to ensure it doesn't miss the target, but this can lead to a high number of false positives. Given your unique situation (tiny dataset, clear single-object targets), here are a few suggestions:
Remember, with such a small dataset, overfitting is almost given, but since you're okay with overlooking it for now, these suggestions should help improve performance without needing more images. Feel free to dive into our docs https://docs.ultralytics.com/yolov5/ for more detailed guidance on training adjustments and techniques. Keep experimenting, and good luck! 🚀 |
👋 Hello there! We wanted to give you a friendly reminder that this issue has not had any recent activity and may be closed soon, but don't worry - you can always reopen it if needed. If you still have any questions or concerns, please feel free to let us know how we can help. For additional resources and information, please see the links below:
Feel free to inform us of any other issues you discover or feature requests that come to mind in the future. Pull Requests (PRs) are also always welcomed! Thank you for your contributions to YOLO 🚀 and Vision AI ⭐ |
Discussed in #12928
Originally posted by eric80739 April 16, 2024
I'm using YOLOv5n to train on a very small dataset with only around 10 images.
Image size is 512*512, and the targets are very clear. I want to detect the only one green dot in the image below:
Therefore, I expect the model to converge quickly, ignoring the overfitting.
However, the training isn't going as expected, and it's taking a long time to converge. During training, I noticed a unreasonable phenomena, very low precision but unusually high mAP:
The training performance of the model on my other dataset is excellent, as shown in the graph below:
So, we can probably rule out issues with parameter settings. I would like to inquire about how to improve the training method without increasing the number of images.
Thank you!
The text was updated successfully, but these errors were encountered: