Hyperparamets

[varghesealex90]

Using Alexy's github page (https://github.com/AlexeyAB/darknet) and using the this cfg
https://github.com/PengyiZhang/SlimYOLOv3/blob/master/cfg/00-unpruned/yolov3-spp3.cfg I was able to mAP of 82. Apart from the number of classes, no other changes in hyper-parameters were made.
I would like to replicate the same using ultralytics. Can someone please tell how to change the hyper-parameters in ultralytics accordingly. I see parameters such as

'giou': 1.582, # giou loss gain
'xy': 4.688, # xy loss gain
'wh': 0.1857, # wh loss gain
'cls': 27.76, # cls loss gain (CE=~1.0, uCE=~20, uBCE=~200,~30)
'cls_pw': 1.446, # cls BCELoss positive_weight
'obj': 21.35, # obj loss gain
'obj_pw': 3.941, # obj BCELoss positive_weight
'iou_t': 0.2635, # iou training threshold
'lr0': 0.002324

Where can see these hyperparametrs in the original yolov3 implementation

[glenn-jocher]

@varghesealex90 closing as duplicate of #392.

I recommend you use the default hyperparameters for your training, and don't change anything.