Add --label-smoothing eps argument to train.py (default 0.0) (#2344)

* Add label smoothing option

* Correct data type

* add_log

* Remove log

* Add log

* Update loss.py

remove comment (too versbose)

Co-authored-by: phattran <phat.tranhoang@cyberlogitec.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>

utils/loss.py

@@ -97,7 +97,7 @@ class ComputeLoss:
         BCEobj = nn.BCEWithLogitsLoss(pos_weight=torch.tensor([h['obj_pw']], device=device))
 
         # Class label smoothing https://arxiv.org/pdf/1902.04103.pdf eqn 3
-        self.cp, self.cn = smooth_BCE(eps=0.0)
+        self.cp, self.cn = smooth_BCE(eps=h.get('label_smoothing', 0.0))  # positive, negative BCE targets
 
         # Focal loss
         g = h['fl_gamma']  # focal loss gamma