rcurtin_irc changed the topic of #mlpack to: mlpack: a scalable machine learning library (https://www.mlpack.org/) -- channel logs: https://libera.irclog.whitequark.org/mlpack -- NOTE: messages sent here might not be seen by bridged users on matrix, gitter, or slack
<ShubhamAgrawal[m> <zoq[m]> "Wondering if in a previous..." <- Actually the issue is we are hardcoding the location but for some reason the location of directory changed for linux
<ShubhamAgrawal[m> It kind of takes like cpython-37
<ShubhamAgrawal[m> Something on these lines
<ShubhamAgrawal[m> * these lines now instead of 3.12
<rcurtin[m]> if I am remembering right, I was angry that I even had to write that script because I didn't have a safe way to get the generated directory name from setuptools
<rcurtin[m]> so... I am altogether unsurprised if setuptools changed things and it's now wrong. ugh... if you know the new formula to make it right, feel free to make a change there, but I guess we also have to figure out what version of setuptools it changed with, and then consult the version of setuptools to see which formula to use...
<rcurtin[m]> thank you for looking into that!
<NabanitaDash[m]> I am unable to submit proposal through Gsoc website.
<zoq[m]> Do you get any error?
<NabanitaDash[m]> No, I dont know how to submit one
<zoq[m]> Not sure I can be of much help here, since what we see is different.
<zoq[m]> Perhaps some else can provide some help?
<NabanitaDash[m]> Alright!
<zoq[m]> I know you have to register as a contributor
<EshaanAgarwal[m]> NabanitaDash[m]: You would need to sign in on Gsoc website. After that , on top right you can select or go to top right on profile/dashboard. There you should see the option i guess 😅
<NabanitaDash[m]> I have registered already
<NabanitaDash[m]> Yes, but i dont see any option for adding a proposal
<NabanitaDash[m]> Maybe I will use a different browser
<zoq[m]> Maybe someone can take a screenshot of that button?
<GopiMTatiraju[m]> I was able to submit successfully
<zoq[m]> Good thing we have 24h? To figure this out.
<GopiMTatiraju[m]> I just did it right now to test
<NabanitaDash[m]> I don't see any option to submit proposal
<EshaanAgarwal[m]> NabanitaDash[m]: pH right you need to search mlpack in browse organization
<EshaanAgarwal[m]> * OH right you need to search mlpack in browse organization. There you should see that option.
<GopiMTatiraju[m]> On the program home go to Browse all 2022 organizations
<GopiMTatiraju[m]> There you can find mlpack and the option to submit proposal
<NabanitaDash[m]> After choosing mlpack, it open in a separate tab and i get logged out of the site. Once i log in, it shows "no proposals submitted" screen
<NabanitaDash[m]> I don't see any submit proposal button
<EshaanAgarwal[m]> NabanitaDash[m]: Maybe delete all cache ?👀
zoq[m]1 has joined #mlpack
<zoq[m]1> Looks like you are doing from a mobile device?
<zoq[m]1> Not sure that is a problem?
<NabanitaDash[m]> No the ss is the same. Thats why i sent it here
<NabanitaDash[m]> I am using my laptop
<NabanitaDash[m]> I think it was my browser.
<NabanitaDash[m]> I got it when i used chrome
<NabanitaDash[m]> Thanks!
<zoq[m]> What browser did you use before?
<NabanitaDash[m]> I was using mozilla
<zoq[m]> Might be good to report that to Google if that is a problem.
<EshaanAgarwal[m]> NabanitaDash[m]: What worked ?
<NabanitaDash[m]> zoq[m]: Let me check how to do that
<NabanitaDash[m]> EshaanAgarwal[m]: Changing browser to chrome
<NabanitaDash[m]> I will email them the issue
<ShubhamAgrawal[m> Is there an option to give 2 proposal to same org?
<zoq[m]> Pretty sure you can submit up to 3 applications, following the same steps.
GopiMTatiraju[m] has quit [*.net *.split]
_slack_mlpack_37 has quit [*.net *.split]
_slack_mlpack_U0 has quit [*.net *.split]
077AAIAM5 has quit [*.net *.split]
SlackIntegration has quit [*.net *.split]
_slack_mlpack_U0 has joined #mlpack
SlackIntegration has joined #mlpack
_slack_mlpack_37 has joined #mlpack
M077AAIAM5 has joined #mlpack
GopiMTatiraju[m] has joined #mlpack
<NabanitaDash[m]> Hey zoq , rcurtin Can you look into another proposal also I have resend it today?
striker_7 has joined #mlpack
<EshaanAgarwal[m]> > <@eshaanagarwal:matrix.org> rcurtin zoq: I have sent link to my proposal. Actually it's still incomplete ( small parts are still left here and there) but I would love to hear out your suggestions and reviews on the rest of it.... (full message at https://libera.ems.host/_matrix/media/r0/download/libera.chat/65aef757840918e30039f28bb1f9d66841900e96)
striker_7 has quit [Quit: Client closed]
<zoq[m]> > <@eshaanagarwal:matrix.org> I have made some modifications in my proposal and worked upon your suggestions.
<zoq[m]> >
<zoq[m]> > rcurtin zoq @marcusedel:matrix.org: Could you please do a final review of my proposal before I can submit it tomorrow ?
<zoq[m]> I'm not sure if I can get to it before the submission deadline, so my recommendation is not wait for another review.
<EshaanAgarwal[m]> <zoq[m]> "> <@eshaanagarwal:matrix.org..." <- Thank You!
<EshaanAgarwal[m]> I will wait till the same time tomorrow.
<EshaanAgarwal[m]> Otherwise, I will submit it.
<GopiMTatiraju[m]> zoq: Will adding the timeline and function list would do for me? Or should I add anything more?
<zoq[m]> <GopiMTatiraju[m]> "zoq: Will adding the timeline..." <- I'm usually looking and checking for a clear and sound application, that said we have an application guide, that has some more details for what we are looking for.
<zoq[m]> What I said applies to all applications, I know this is not the answer you are looking for, ideally I could tell you if you do that and that your application will be accepted. But GSoC is a competitive program we have a limited number of slots/mentors and time.
<GopiMTatiraju[m]> Definitely, I had my experience 😂, I actually meant that are there any more comments?
<zoq[m]> <GopiMTatiraju[m]> "Definitely, I had my experience..." <- I might be able to do another round later today, but I would't wait on it.
<GopiMTatiraju[m]> Okayyy
<GopiMTatiraju[m]> What kinda `shuffle` does `arma` uses?
<GopiMTatiraju[m]> Will we implement the same algo for `coot` as well?
<rcurtin[m]> It might be reasonable to try the same thing first, but it may not perform very well
<rcurtin[m]> at least for sorting, sorting on GPUs is hard; I haven't thought about whether that is the same for shuffling
<rcurtin[m]> I wonder if, for shuffle, it would be faster to generate the random permutation list in CPU memory, then do the shuffling on the GPU
<GopiMTatiraju[m]> Yup that was I was thinking
<GopiMTatiraju[m]> Do we really need GPU for the whole implementation?
<rcurtin[m]> yeah, we can try whatever---basically the first step is just a working implementation, then we can time it and if it is a bottleneck, we can try more complicated strategies
<GopiMTatiraju[m]> Oh yupp, that makes sense, I was going through `notes.txt` from bandicoot...