Manuel Breitenstein

Do not use reduce_sum before returning to loss wrapper.

2

# Description Brief Description of the PR: Handle reduction with loss wrapper not with this function. Currently when using `tfa.losses.SigmoidFocalCrossEntropy(reduction: str = tf.keras.losses.Reduction.NONE)`, the loss is still reduced by summing...

Is this project still actively developed?

2

Is there any plan to keep this project active? I do not see any active development since quiet a long time. Would be a shame. I think it's very useful...

The network architecture weights `alphas_network` are initialized non zero even for connections that do not exist. For example in the first layer we only have two connections, but three weight...