07:33 < KimSangYeon-DGU> The logs for 24th, 25th, and 26th seem to be missing.

08:39 < jenkins-mlpack2> Yippee, build fixed!

08:39 < jenkins-mlpack2> Project docker mlpack nightly build build #201: FIXED in 3 hr 25 min: http://ci.mlpack.org/job/docker%20mlpack%20nightly%20build/201/

16:52 < Suryo> Hello Team!

16:53 < Suryo> On the GSoC ideas page < https://github.com/mlpack/mlpack/wiki/SummerOfCodeIdeas >, I read that one of the potential projects involves developing PSO methods for constraint optimization methods.

16:54 < Suryo> I'm interested to pursue this as I am familiar with this type of a problem.

16:55 < Suryo> You can see my paper here: https://arxiv.org/pdf/1805.08858.pdf

16:55 < Suryo> A little bit about myself: My name is Suryoday Basak < suryodaybasak.info >. I am currently a masters' student in computer science at the University of Texas at Arlington.

16:57 < Suryo> I am familiar with C++ (although I've lost a bit of practice in the past few years, owing to the rising prominence of Python). I have taken coursework in C/C++ at the undergraduate level and am familiar with templating.

16:59 < Suryo> I'd like to be a part of the mlpack project as I am moving towards a career in machine learning and (I am really honest about this), in the past few months to a year, I've been wanting to make a transition to implementing more ML methods in C++ than python.

17:00 < Suryo> And since I'm a student, I'd like to apply for the GSoC program, and be a part of it.

17:02 < Suryo> The paper that I mentioned to you was presented at IEEE Symposium Series in Computational Intelligence, http://ieee-ssci2018.org/

17:15 < zoq> Suryo: Hello there, sounds like a good fit to me, will take a closer look at the paper.

17:15 < zoq> Suryo: There is an open PR which implements a generic PSO, so I guess a good start would be to finish this one (move it into mlpack/ensmallen).

17:29 < Suryo> Dear zoq, thank you for your timely response.

17:30 < Suryo> And thank you for the heads up. I will definitely take a closer look at it. Kindly give me a few days to familiarize myself with the codebase. As I am a student, I have some deadlines coming up over the weekend.

17:31 < Suryo> I will try to move as fast as possible through this. And this includes reviewing the old C++ concepts, understanding the flow of mlpack, and taking a look at the PR. I'll be active on this IRC meanwhile.

17:32 < Suryo> Found the PR.

17:49 < zoq> Suryo: No worries, GSoC hasn't even started yet.

17:50 < Suryo> I've found ensmallen

17:50 < Suryo> Looking through, but will take a week for me to grasp everything that's going on.

17:51 < zoq> Sounds good, let me know if I should clarify anything.

17:51 < Suryo> Will do. Thank you.

17:52 < glaz77> Hi, I am an undergraduate at III-H and am pursuing a bachelors in Computer Science and a Masters in Computational Linguistics and Natural Language Processing. I am pursuing research actively in the NLP lab at IIIT-H under the guidance of Dr. Manish Shrivastava. I have worked on various projects in NLP such as Named Entity Recognition, Grammatical Error Correction, Authorship Detection, Generating Cross-Lingual Word Embeddings, and Fa

17:54 < glaz77> I have undertaken courses in Machine Learning and Deep Learning and have read quite a few research papers in the process. I also have experience of working in C++. I am really interested in working on the project of 'NeuroEvolution of Augmenting Topologies' as mentioned on the Github page for ideas for GSOC 2019.

17:55 < glaz77> It would be great to work and build under mlpack for GSOC 2019

18:00 < zoq> glaz77: Hello, glad you like the project, I think a good starting point is to go through the references that are listed on the project ideas page.

18:05 < glaz77> ok, I will go through them

21:30 < Suryo> zoq, I have a couple of questions. I saw the open PR about the PSO method by chintans111. However, at the time that the pull request was merged into the master branch, the changes went into the following directory: mlpack/src/mlpack/core/optimizers/

21:31 < Suryo> However, at this time, the optimizers directory does not exist in mlpack/src/mlpack/core/

21:32 < Suryo> So the optimizers need to go in mlpack/ensmallen

21:33 < Suryo> And you would like to move the PSO methods to ensmallen. Is that right?

21:35 < zoq> Correct, last year moved the optimization framework out of mlpack into it's own repository, and now use ensmallen as a dependency.

21:49 < rcurtin> KimSangYeon-DGU: I'm really sorry, I have no idea what is going on with the IRC log creation process

21:49 < rcurtin> I'm instrumenting the generators now to provide output that maybe will be helpful

21:52 < suryo> zoq: yeah, I guessed that when I tried installing mlpack sometime back and saw ensmallen as a dependency. But thanks for clarifying.

21:53 < suryo> And by saying that you want to move the open PR on PSO to mlpack/ensmallen, do you mean that I should try to move the pso optimizer to ensmallen/include/ensmallen_bits/ and try to integrate it?

22:04 < zoq> suryo: Yeah, basically just open a new PR including the test in the ensmallen repo, if you build the ensmallen tests there should be some errors since ensmallen doesn't know much about mlpack.

22:51 < Suryo> zoq: okay I'm beginning to understand what's required. I'll spend some more time reading and understanding the codebase of ensmallen. It'll take me some time, like I mentioned earlier, but I'll keep trying to make progress. Thanks.