👋 Hello @xxxkkw, 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

@xxxkkw hello,

Thank you for providing a detailed description of the issue you're encountering with YOLOv5 on your Mac device using MPS acceleration. It seems like you've done a thorough job of setting up your environment and troubleshooting the problem. Let's address the issues step-by-step:

Deprecation Warning

The warning message you encountered:

FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.

This is a known issue with the latest versions of PyTorch. While it doesn't affect the training process, it indicates that the codebase needs to be updated to align with the latest PyTorch API changes. You can safely ignore this warning for now, but it's good to keep your packages updated to avoid such warnings in the future.

Video Detection Issues with MPS

The issue with the detection results being incorrect when using MPS but correct when using CPU suggests a potential problem with the MPS backend in PyTorch. Here are a few steps you can take to troubleshoot and potentially resolve this issue:

1. Verify with Latest Versions: Ensure you are using the latest versions of YOLOv5 and PyTorch. Sometimes, bugs are fixed in newer releases.

   git pull  # update YOLOv5 repo
   pip install --upgrade torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cpu

2. Test with Stable PyTorch: While nightly builds are great for accessing the latest features, they can sometimes introduce instability. Try using the latest stable release of PyTorch.

   pip uninstall torch torchvision torchaudio
   pip install torch torchvision torchaudio

3. Check for Known Issues: Look through the PyTorch GitHub issues for any known problems with MPS on macOS. If you find a related issue, you can track its progress or contribute additional information.

4. Fallback to CPU: If MPS continues to cause issues, you might need to fallback to using the CPU for inference on your Mac. While not ideal, it ensures that your results are accurate.

Using Pretrained Models

The fact that using yolov5s.pt works correctly while your custom-trained model does not suggests there might be differences in the training process or data. Ensure that your training data and process are consistent and that there are no discrepancies.

Example Code for CPU Inference

Here's an example of how you can run inference using the CPU to avoid MPS-related issues:

python detect.py --weights your_custom_model.pt --source your_video.mp4 --device cpu

Reporting Bugs

If the issue persists, consider reporting it to the PyTorch team with detailed information about your setup and the problems you're encountering. This helps improve the MPS backend for everyone.

Thank you for your patience and for being a part of the YOLO community. If you have any further questions or need additional assistance, feel free to ask!