search

ultralytics / yolov5

YOLOv5 🚀 in PyTorch > ONNX > CoreML > TFLite
https://docs.ultralytics.com
GNU Affero General Public License v3.0
50.58k stars 16.31k forks source link

Data Augumentation to improve YOLOV5 Training accuracy #12380

Closed naveenvj25 closed 10 months ago

naveenvj25 commented 11 months ago

Search before asking

Question

@glenn-jocher I have a custom dataset with annotation labels. Now, I am planning to increase the dataset with some standard augmentation. Does the yolov5 model automatically perform augmentation? If so, by updating the default hyperparameters, can it increase the images by augmentation and further improve MAP and other matrix values? Or should I do augmentation manually, but if so, I need to do labeling also, which might take more time.

Additional

No response

glenn-jocher commented 11 months ago

@naveenvj25 hi there! YOLOv5 doesn't currently perform data augmentation automatically during training. However, you can easily apply it using popular libraries like Albumentations or OpenCV, and then adapt the augmented images and bounding boxes accordingly. This typically improves model accuracy. You can learn more on data augmentation in the training section of the Ultralytics Docs. If you have further questions, feel free to ask!

naveenvj25 commented 11 months ago

@glenn-jocher hi..! Actually, by default, the following hyperparameters related to augmented are called in the train.py file from the hyp file. So my doubt is, does some augmentation happening in the images while training..?

hsv_h: 0.015 # image HSV-Hue augmentation (fraction) hsv_s: 0.7 # image HSV-Saturation augmentation (fraction) hsv_v: 0.4 # image HSV-Value augmentation (fraction) degrees: 0.0 # image rotation (+/- deg) translate: 0.1 # image translation (+/- fraction) scale: 0.5 # image scale (+/- gain) shear: 0.0 # image shear (+/- deg) perspective: 0.0 # image perspective (+/- fraction), range 0-0.001 flipud: 0.0 # image flip up-down (probability) fliplr: 0.5 # image flip left-right (probability) mosaic: 1.0 # image mosaic (probability)

glenn-jocher commented 11 months ago

@naveenvj25 Those parameters in the hyp file are indeed related to augmentations. YOLOv5 does perform built-in augmentations during training using these parameters. Feel free to tweak these values to optimize your training based on your specific dataset and requirements. If you need further assistance or have more questions, don't hesitate to ask!

naveenvj25 commented 11 months ago

Ok, thank you...I will explore that..!

glenn-jocher commented 11 months ago

@naveenvj25 you're welcome! If you have any more questions as you explore, feel free to reach out. Best of luck with your exploration!

github-actions[bot] commented 10 months ago

👋 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 ⭐