Details about image augmentation and hyperparameters

[henrywu829]

Search before asking

• I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

I am new to yolov5 and I have some questions about image-augmentation.

I find some hyperparameters about image-augmentation in the hyp.scratch-low.yaml file, such as flipud, fliplr, copy_paste, mixup and mosaic, I'm curious if I would get better precision and recall if i apply these augmentation methods in the same time?

And I also want to know what's the difference between degree, flipud and fliplr

Thanks~

Additional

no

[github-actions]

👋 Hello @henrywu829, 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.

Requirements

Python>=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

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

• Notebooks with free GPU:
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Amazon Deep Learning AMI. See AWS Quickstart Guide
• Docker Image. See Docker Quickstart Guide

Status

If 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

[glenn-jocher]

@henrywu829 hello there! Welcome to YOLOv5 community!

Augmentations are an important aspect of training Object Detection models. Each augmentation operation serves a particular purpose and unique use-case. Combining different augmentations have proven effective in improving model performance in some cases, but not always.

In YOLOv5, the hyp.scratch-low.yaml file lists the most commonly used augmentations used in training.

degree is an augmentation for 'Rotation'. flipud and fliplr serve for 'Vertical flip' and 'Horizontal flip' respectively.

As for whether applying all augmentations at the same time produces better performance, this depends on the specific use-case and dataset being used. It is advised to experiment with different augmentation combinations and study their effects on model performance.

Hope this helps! Let me know if there's anything else I can help with :)

[henrywu829]

Thank you, this is helpful for me ~ I will try applying these methods.

[glenn-jocher]

@henrywu829 you're welcome! Best of luck with your training! Don't hesitate to ask if you have any further questions. Happy learning!