I have seen people using stacked softmax layers right at the output of neural networks designed for classification. I'm trying to understand this. Does it give any additional value? I think this could "sharpen" decisions on the boundaries.

model.add(Dense(10, activation='sigmoid'))
model.add(Dense(1, activation='sigmoid'))

Seen here.

  • $\begingroup$ could you give an example? i dont think this does anything $\endgroup$ – k.c. sayz 'k.c sayz' Jan 27 '18 at 0:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.