Discriminator using activation sigmoid in Gan

#1

Hi,

In this video when we create the model for discriminator we are using sigmoid activation for output layer.

See Below code: -

discriminator.add(Dense(1,activation=‘sigmoid’))

In the training tricks video, it was mentioned we need 0.9 as real image output.
But sigmoid will produce 1 right? - for real image output.

So how will that be handled?

Thanks

#2

hey @lintasheelkumar ,
sigmoid provides an output as a probability from 0 to 1.
Which means that if our prediction is greater than 0.5 , it is a true else False.

I hope this helps.

#3

The next video ha cleared my doubt. We are mutliplying that one with 0.9 in the code. Thanks!

#4

That’s good.
I request you to kindly mark this doubt as resolved in your course’s doubt section and do provide your valuable feedback.

Thank You and Happy Learning :slightly_smiling_face:.

#5

I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.

On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.