ChanServ changed the topic of #mlpack to: "mlpack: a fast, flexible machine learning library :: We don't always respond instantly, but we will respond; please be patient :: Logs at http://www.mlpack.org/irc/
< AnushKiniGitter[>
> `Shah Anwaar Khalid` Hi! While writing examples for GANs, I realized the way GAN class has been implemented, it will only show discriminator's errors during training. Often, it's useful to monitor both generator's as well as discriminator's errors to make sure one is not overpowering the other (like in mode collapse). Perhaps, it makes sense to have a dual optimizer for GANs, one for generator and another for
< AnushKiniGitter[>
discriminator. Please let me know your thoughts.
< AnushKiniGitter[>
I agree with this. I found the need for this when I tried to train DCGAN models using mlpack.
< kaushal07[m]>
hey I was testing the changes on the file src/methods/ann/layer/recurrent.hpp and src/methods/ann/layer/recurrent_impl.hpp and recurrent _attention file as well it shows this erorr all the time, anyone can help on this.
ImQ009 has joined #mlpack
< AnushKiniGitter[>
Interested in the idea **Improve mlpack's tree ensemble support**. Adding to this, I believe the ensemble models in mlpack do not have a method to get the importance of each feature. Adding support for this is also something I am considering taking up.
< AnushKiniGitter[>
(edited) ... feature. Adding ... => ... feature. similar to scikit-learn. Adding ...
< AnushKiniGitter[>
(edited) ... idea **Improve ... => ... idea to **Improve ...
< AnushKiniGitter[>
(edited) ... each feature. similar to scikit-learn. ... => ... each feature, like in scikit-learn. ...
mrshu[m] is now known as mrshu
mrshu has quit [Quit: authenticating]
mrshu has joined #mlpack
petris has quit [Remote host closed the connection]