-
-
Notifications
You must be signed in to change notification settings - Fork 16.3k
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
Unable to get NMS in TFJS exported model #11728
Comments
👋 Hello @Abhishekvats1997, 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>=3.7.0 with all requirements.txt installed including PyTorch>=1.7. 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 |
@Abhishekvats1997 hi there! Thank you for bringing this issue to our attention. I understand that you are experiencing trouble with the To investigate this issue further, it would be helpful if you could provide a minimal reproducible example, including the steps to export the model and any relevant code snippets. This will allow us to better understand the problem and assist you with finding a solution. Furthermore, if you are willing to submit a pull request to address this issue, we greatly appreciate your contribution! Our team will be more than happy to review and merge your changes. Please ensure that your proposed changes follow our contribution guidelines. Thank you again for your report, and we look forward to working with you to address this issue. |
Code used to export After a bit of digging I could see the reason for this as The "or" here is causing the model conversion to default to class-agnostic NMS in the case of tfjs. After trying to remove this condition I get errors highlighting that the CombinedNonMax Suppression layer is not implemented on TFJs. Is this the reason that you have introduced the above-mentioned "or" condition? If yes, can you suggest a possible solution to use yolov5 on tfjs with per class NMS and limit objects per class other than the manual for loops in JS :( . |
@Abhishekvats1997 hello! Thank you for bringing this issue to our attention. I understand that you're experiencing difficulty with the After further investigation, it appears that the "or" condition you noticed in the export script is indeed causing the model conversion to default to class-agnostic NMS in the case of TensorFlow.js. Unfortunately, the CombinedNonMax Suppression layer is not implemented in TensorFlow.js, which is why the condition is in place. As of now, there is no straightforward solution to accomplish per-class NMS and limit the number of objects per class in TensorFlow.js without resorting to manual for loops in JavaScript. It would involve implementing custom JavaScript code for performing per-class NMS and enforcing the object limits. We understand that this could lead to potential performance overhead due to the need for additional computation in JavaScript. If this functionality is critical for your use case, we recommend considering alternative options such as using Python or exploring other object detection frameworks that support per-class NMS and object limits in TensorFlow.js. We hope this information clarifies the situation. Please let us know if you have any further questions or concerns. We appreciate your interest in YOLOv5 and the Ultralytics team's work. Thank you for your understanding and contribution to the YOLO community! |
👋 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 ⭐ |
Search before asking
YOLOv5 Component
No response
Bug
I tried exporting my trained .pt model to tfjs. The --nms flag and --topk-per-class flags do not seem to have any effect on the exported model.
Environment
No response
Minimal Reproducible Example
No response
Additional
No response
Are you willing to submit a PR?
The text was updated successfully, but these errors were encountered: