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/
< jonpsy[m]>
Hey zoq, wouldn't it be nice if have a Trait check for ```typename... ArbitraryFunctionTypes```, to ensure each of them has an ```Evaluate()``` method?
< jonpsy[m]>
so, inside NSGA2, I could ```CheckMultiArbitraryFunctionAPI<ArbitraryFunctinoTypes...>```, sounds cool? :)
ImQ009 has joined #mlpack
< AakashkaushikGit>
Hey, I posted some of the things i wanted to discuss in the mailing list, I will be glad if anyone can take a look.
< RishabhGargGitte>
Hello @rcurtin @zoq , I would like to apologise in advance for tagging you for this, but it would be really helpful for me if I could get some feedback on my project idea. As today is the last day for project discussion, so if I could get some slight feedback, then I can focus on making my proposal and I can properly plan about the project. I hope you understand me. Thanks a lot :)
< rcurtin[m]1>
what do you mean, 'today is the last day for project discussion'? the application period is open for two weeks... you don't have to submit a perfect application on March 29th, there is still a lot of time
< RishabhGargGitte>
Yes, you are right, but I think making proposal is an iterative process, so if my deliverables and idea are finalised, then I can start with the proposal early on to get it reviewed.
< RishabhGargGitte>
I was following the timeline on the website and it said "Interested students can now begin discussing project ideas with accepted mentor organizations." for the March 10 to March 29 period. Did I miss something here ?
< jonpsy[m]>
It didn't say "thou shalt not discuss idea any further than 29th of match ;)
< jonpsy[m]>
It's fine, we have till April :)
< zoq>
jonpsy[m]: Do you mean CheckArbitraryFunctionTypeAPI? This already checks for Evaluate().
< jonpsy[m]>
Yep, but only for single objective
< zoq>
jonpsy[m]: I see, right, yes would be a good idea to add another trait.
< jonpsy[m]>
Can I extract the current type from std::tuple<ArbitraryFuntype... >
< jonpsy[m]>
<zoq "jonpsy: I see, right, yes would "> + Ryan gave this amazing idea for FunctionType test for multiobjectives.
< ABHINAVANAND[m]>
Hey zoq, did you read my idea to improve the speed of pooling layer.
< jonpsy[m]>
We're heading towards a well built MOO support :D
< zoq>
jonpsy[m]: Only if you unpack the parameter pack.
< zoq>
ABHINAVANAND[m]: Not yet, no.
< jonpsy[m]>
Yep, that's what has been bothering me
< jonpsy[m]>
I think for a type trait for multiobjectiveArbitrary, I'd have to use recursion?
< zoq>
jonpsy[m]: Yes, don't think there will be any performance penalty, especially since we are doing this only once.
< jonpsy[m]>
Aight, that's on my list then 👍
< rcurtin[m]1>
yeah, I think that all of the work should happen at compile time too, so hopefully the call itself ends up getting compiled out
< jonpsy[m]>
Fingers crossed 🤞
< rcurtin[m]1>
it's actually not impossible to check, but, it is certainly tedious :)
< jonpsy[m]>
Ill just hope the user has a reasonable compiler