Glenn Jocher

@zglihs @noreenanwar 👋 hi, thanks for letting us know about this possible problem with YOLOv5 🚀. We've created a few short guidelines below to help users provide what we need...

@yuerlong 👋 hi, thanks for letting us know about this possible problem with YOLOv5 🚀. We've created a few short guidelines below to help users provide what we need in...

@jxzddup yes that's correct. attempt_load() `map_location` argument has been renamed to `device`: https://github.com/ultralytics/yolov5/blob/1ab23fc67f52d44d5f8ce67a895e73c7cbd7aec5/models/experimental.py#L74-L75

@MichaelDanAmar yes you can do that. See RTSP, RTMP, HTTP etc example in Colab notebook Detect section:

@thepycoder ah got it. Simplest quick fix is to scope the import to place it within the `try: import clearml` statement. This will reduce the affected userbase. A more permanent...

@abcsunshine @thepycoder good news 😃! Your original issue may now be fixed ✅ in PR #8915. This is not a permanent solution but rather a scoped import quickfix. To receive...

@abcsunshine @thepycoder I think this shows the need for another CI check at the torch minimum requirement. I'll add this to the YOLOv5 CI to catch these problems earlier in...

@PushpakBhoge 👋 Hello! Thanks for asking about improving YOLOv5 🚀 training results. Low confidence scores are indicative of too short trainings and can be resolved by training for more epochs....

@PushpakBhoge for very small objects you'll benefit most from a larger train/detect --imgsz, i.e. 1280 or higher.

@PushpakBhoge increased image size helps small objects even if expanded beyond native size. COCO performs better at 1280 even though no image is larger than 640. Alternatively you could try...