Why delete the sigmoid in the output layers?

[tongxueqing]

Thanks for your wonderful work. I wonder why you delete the sigmoid in the output layers in DenseNet.py in the latter version? And I guess it will be reasonable if self.fc_dist(out) is followed by softmax , since the sum of gt_distribution is one .(https://github.com/fnzhan/Illumination-Estimation/blob/master/RegressionNetwork/DenseNet.py)

[fnzhan]

Hi, I just find including sigmoid will make the network more difficult to converge during training.

[tongxueqing]

Thanks for your timely reply. Can you share how large is the subset when you begin to add schedule for learning rate? And I found you do not save ambient term in the test.py, is it due to the ambient term is not important?

[tongxueqing]

https://github.com/fnzhan/Illumination-Estimation/blob/master/RegressionNetwork/DenseNet.py
The if condition does not hold, it is a bug or you intend to do so?