Hello. First of all, I would like to thank learnopencv for sharing these awesome learning materials.
Currently, I am trying to learn how to use the yolox for object detection. I found your article (https://learnopencv.com/yolox-object-detector-paper-explanation-and-custom-training/) and the github repo. I simply downloaded YOLOX_training_on_custom_drone_dataset.ipynb from the given Google Colab link, and ran the code inside. During the training step, it trained the first epoch and crashed during evaluation process before starting the next epoch. The error says:
ValueError: operands could not be broadcast together with shapes (24,5) (0,)
(see attached train_log.txt for details). I just downloaded yolox this past week. I wondered if the current yolox version causes the problem.
Added 20240827 -- I think this is the same issue here Megvii-BaseDetection/YOLOX#1701
Thank you very much.
Warm regards,
KB
train_log.txt