anirudhacharya commented on a change in pull request #13612: add pos_weight for
SigmoidBinaryCrossEntropyLoss
URL: https://github.com/apache/incubator-mxnet/pull/13612#discussion_r242276230
##########
File path: python/mxnet/gluon/loss.py
##########
@@ -194,6 +194,14 @@ class SigmoidBinaryCrossEntropyLoss(Loss):
L = - \sum_i {label}_i * \log({pred}_i) +
(1 - {label}_i) * \log(1 - {pred}_i)
+ A value `pos_weights > 1` decreases the false negative count, hence
increasing
Review comment:
please update the formulas above with the new `pos_weight` parameter. The
documentation should specify how the formula changes when `from_sigmoid` is
True and False and where in the formula `pos_weight` is used.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
