New Release: Ultralytics v8.3.7

Discussion
,
1

:star2: Ultralytics v8.3.7 Release Announcement

Hello Ultralytics Community!

We’re excited to announce the release of v8.3.7, packed with enhancements and fixes designed to improve your experience with YOLO. This update focuses on refining custom predictor functionality, boosting performance, and optimizing resource management.

:bar_chart: Key Changes

  • Custom Predictor Argument Fix: We’ve corrected the argument handling in model.predict() for custom predictors, ensuring smoother operations.
  • Docker Image Update: The base Docker image now includes PyTorch 2.4.1, CUDA 12.1, and cuDNN 9, enhancing training and inference speed.
  • New Script for Synthetic Datasets: A new function to create synthetic COCO datasets has been added, simplifying data testing and augmentation.
  • Enhanced AutoBatch Memory Management: Improved GPU memory handling during autobatching for optimized resource usage.
  • Added OMP_NUM_THREADS=1: Adjustments in Docker configurations for better CPU management and performance.

:dart: Purpose & Impact

  • Improved Custom Predictor Functionality: Accurate argument handling for custom predictors enhances user experience.
  • Performance Boost: The Docker update leverages the latest PyTorch improvements for faster model operations.
  • Simplified Data Handling: The synthetic dataset script aids developers in preparing data for model validation.
  • Efficient Resource Management: Optimized memory and CPU usage ensure better performance across systems.
  • Streamlined Development: These updates collectively enhance workflows, reduce bugs, and improve code clarity.

:arrows_counterclockwise: What’s Changed

Full Changelog: v8.3.6…v8.3.7

We encourage you to try out the new release and share your feedback. Your insights are invaluable in helping us improve and innovate. Thank you for being a part of our community!

Release URL