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/
chandramouli_r has quit [Ping timeout: 252 seconds]
kanishq244[m] has quit [Ping timeout: 252 seconds]
skrpl[m] has quit [Ping timeout: 252 seconds]
Sergobot has quit [Ping timeout: 250 seconds]
ani1238[m] has quit [Ping timeout: 250 seconds]
shashank-b[m] has quit [Ping timeout: 250 seconds]
johnsoncarl[m] has quit [Ping timeout: 252 seconds]
harias[m] has quit [Ping timeout: 252 seconds]
lozhnikov_ has joined #mlpack
ani1238[m] has joined #mlpack
skrpl[m] has joined #mlpack
harias[m] has joined #mlpack
shashank-b[m] has joined #mlpack
Sergobot has joined #mlpack
chandramouli_r has joined #mlpack
johnsoncarl[m] has joined #mlpack
kanishq244[m] has joined #mlpack
Suryo has joined #mlpack
< Suryo>
Hello everyone! Glad that I could be here to contribute to mlpack and ensmallen over summer! :)
< Suryo>
rcurting, zoq: As per my plan, I should be making a list of objective functions that I would like to optimize using PSO. I figured that wikipedia already has a list of them. Plus, since ensmallen has some of these built in, I've got a good place to start off.
< Suryo>
And as for discrete valued functions - there has been a lot of work that has been done on applying genetic algorithms and other metaheuristics to problems like the TSP, knapsack, etc. and I'm just pulling out some of my favourite ones. We can compare the performance of PSO to exhaustive search approaches as well as bounded approximation algorithms, and see where we get.
< Suryo>
Another thing that we would need to discuss for my project is the design for constrained optimization objectives. I'll be brainstorming for that for sometime from now.
< jeffin143>
lozhnikov_, Hi , would like to update you on my plan regarding the work for gsoc, So I am done will all my pending PRs , and now would start for the fresh one . There is one PR left for dictionary encoding , please review it whenever you are free pr#1814,
< jeffin143>
The first plan is to built basic function for cleaning of string dataset , which I would start to implement soon . Would also keep you updated on my plans .
travis-ci has joined #mlpack
< travis-ci>
mlpack/ensmallen#280 (ensmallen-1.14.4 - 799bd82 : Ryan Curtin): The build passed.
< zoq>
Suryo: Agreed, let us setup a test suite, I like the problems you suggested also what do you think about: https://bbcomp.ini.rub.de?
< zoq>
Suryo: I don't mind to finish the non-constrained version first, what do you think?
< zoq>
chandramouli_r: Sorry for the slow response; About the research opportunity, perhaps we can find a neat trick to speed it up or find another trick that does improve the performance.
< zoq>
chandramouli_r: What do you think, should we start with something more concrete?