konioy
konioy
### Search before asking - [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and found no similar bug report. ### YOLOv5 Component _No response_ ### Bug I know there are some...
### Search before asking - [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and [discussions](https://github.com/ultralytics/yolov5/discussions) and found no similar questions. ### Question Same data, same code, I trained it twice. The...
### Search before asking - [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and [discussions](https://github.com/ultralytics/yolov5/discussions) and found no similar questions. ### Question In build_targets, the GT box is extended, plus neighboring...
In you pre_trained Resnet50 model, you set k = 3,5,5,7. But i think that k = 3.5.5.5. Am I mistake？
微信群满了 qq加了没人通过 方便拉我？qq:928482720
在DCQ的前向中，对im_k做了一次shuff（对应函数_batch_shuffle_ddp），然后又做了一次unshuff（对应函数_batch_unshuffle_ddp），我认为得到的结果和直接对im_k做前向的结果是一样的。这样做的目的是什么？
Why clean and organize this data? Is it to reduce the amount of data and speed up training?
How do I convert to ONNX? I'm encountering errors when trying to convert to ONNX. `RuntimeError: The serialized model is larger than the 2GiB limit imposed by the protobuf library....
Will using more data during MNTP and SimCSE training improve performance? Are there any datasets you would recommend?