[mlpack-git] [mlpack] Create a function to visualize the features learned by sparse autoencoder (#465)

stereomatchingkiss notifications at github.com
Wed Nov 25 00:55:11 EST 2015


>Does my idea make sense here, or maybe am I misunderstanding something?

I am not sure about this one, reshape the maximal inputs into a square block looks like a default way to visualize the features learned by autoencoder, don't know the other settings(block width * block height) meaningful or not. 

However, if you prefer to make the class more general, I will make the constructor provide a constructor with blockWidth and blockHeight with default value equal to 0, explain the purpose of them by the comments.

---
Reply to this email directly or view it on GitHub:
https://github.com/mlpack/mlpack/pull/465#issuecomment-159499081
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/mlpack-git/attachments/20151124/a80ecff2/attachment-0001.html>


More information about the mlpack-git mailing list