Improve inference results

[AlexandreDiPiazza]

Search before asking

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

Question

Hello,

My goal is to boost the inference results as much as possible, with yolov5-small on a private dataset.

I am doing inference on images at full resolution.

I tried test-time augmentations (combinations of horizontal flips and resize) and ensemble of multiple model checkpoints, but it doesn't really have an impact. My model already has a good performance ( [email protected] of around 0.6), but I would like to boost the results, especially on some minor categories.

Do you have any hints on other techniques that can be useful to boost inference results?

Additional

No response

[github-actions[bot]]

👋 Hello @AlexandreDiPiazza, 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>=3.8.0 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

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]

@AlexandreDiPiazza hello!

Great to hear that you're achieving good performance with YOLOv5-small on your dataset! 🎉 Here are a few suggestions to potentially boost your inference results further:

1. Data Augmentation: Continue experimenting with different augmentation strategies during training, not just inference. This can help the model generalize better to unseen data.

2. Fine-tuning: If you haven't already, fine-tune your model on your specific dataset starting from the pre-trained weights.

3. Class Weights: Adjust the loss function to give more importance to minor categories that are underrepresented or harder to detect.

4. Hyperparameter Tuning: Experiment with different hyperparameters, such as learning rate, batch size, or optimizer. Use the YOLOv5 hyperparameter evolution feature to automate this process.

5. Model Architecture: Try using a larger variant of YOLOv5, such as YOLOv5-medium or YOLOv5-large, if your computational resources allow it.

6. Post-processing: Adjust the confidence threshold and non-maximum suppression (NMS) threshold to filter out false positives and refine the detections.

7. Error Analysis: Perform a thorough error analysis to understand where the model is failing and focus on improving those specific areas.

Remember, improving on already good performance can be challenging and may require a lot of experimentation. Good luck, and keep us posted on your progress in the discussions section!

For more detailed guidance, please refer to our documentation at https://docs.ultralytics.com/yolov5/.

Happy detecting! 🚀

[github-actions[bot]]

👋 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:

• Docs: https://docs.ultralytics.com
• HUB: https://hub.ultralytics.com
• Community: https://community.ultralytics.com

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 ⭐