<jonpsy[m]>
zoq: updated the PR, made some major changes. Let's move our convo there
<jonpsy[m]>
* zoq: say4n updated the
texasmusicinstru has quit [Remote host closed the connection]
texasmusicinstru has joined #mlpack
fusta has quit [Quit: Leaving]
heymike[m] has quit [Quit: You have been kicked for being idle]
<zoq[m]>
Hello everyone, today is the GSoC mentor summit, so what about we postpone the mlpack meeting to next week?
TarunKumar[m] has joined #mlpack
<jonpsy[m]>
so, meeting for gsoc'21 mentors?
<rcurtin[m]>
does anyone have a link to info about the mentor summit? I forgot to register 🤦
<rcurtin[m]>
jonpsy: yeah, each year GSoC holds a "mentor summit" where all of the mentors can join some talks/discussions on things like feedback, planning moving forward, some talks, etc.; I've found it in the past to be super useful
<zoq[m]>
I think you only have to register for the breakout sessions, not sure.
<rcurtin[m]>
nice---I'll fit in the stream where I can (juggling meetings as usual...)
<zoq[m]>
hehe, same here, I have meetings at the same time, but I'll try to keep the stream open.
<zoq[m]>
"15:15 - 15:50 - 2021 highlights, GSoC 2022 changes and the Future of GSoC " is what I'm interested in
<rcurtin[m]>
I actually have a lunch meetup today, which is very unusual... it will be strange to see my coworkers in person for once 😄 but that is right in the middle of the stream, so I guess I have to miss some of it
<rcurtin[m]>
yeah luckily that part is before I have to leave---that's what I'm most interested in also
<shrit[m]>
I will be listening
<shrit[m]>
I will let you knows what happens
<shrit[m]>
s/knows/know/
yugansharora01yu has joined #mlpack
<yugansharora01yu>
least "8.400.0" (found C:/mlpack/armadillo/include)
<yugansharora01yu>
Could NOT find Armadillo: Found unsuitable version "..", but required is at
<yugansharora01yu>
this is the error i am getting when i run cmake for mlpack
<jonpsy[m]>
I've heard after you've reached a senior level, most of the job involves meetings and not coding. Is that true? I find it a little depressing that our entire time is sharded in multiple meetins
<yugansharora01yu>
any suggestion what should i do?
<shrit[m]>
jonpsy: If you are senior by 28 would that involves not doing coding? I am not sure
<shrit[m]>
yugansharora01 (yugansharora01): Would you provides more information
<rcurtin[m]>
jonpsy: yeah, that can happen, but it depends a lot on the company and the path that you choose through it. there comes a point where, as a senior developer, it is more worthwhile for you to spend time mentoring and teaching others instead of working on things alone
<rcurtin[m]>
but I think it is also completely possible to continue to be an individual contributor for one's entire career; you just have to be clear that that is what you want to do, and choose accordingly (so, e.g., if you have the opportunity to become a manager of a group, you might say no if you want to keep getting your hands dirty in the code 😄)
<rcurtin[m]>
in my current situation, I have a lot of meetings, but in many of these I only need to listen and I write a lot of code in the background (yesterday was a start-to-finish 1k+ line PR for instance); the reason I need to listen is to make sure that the things that I am developing will fit well with other parts of the product that are being worked on
<rcurtin[m]>
so, that is not ideal, because ideally I would have a manager who would be doing all that listening for me and keeping me apprised of these types of things, but... nothing is ideal in a startup 😄 (in fact I think it is reasonable to extrapolate and say nothing is ideal anywhere...)
<yugansharora01yu>
@shrit:matrix.org i am trying to build mlpack for windows by following this page "https://www.mlpack.org/doc/mlpack-3.4.2/doxygen/build_windows.html" and have completed every step till building mlpack (dependencies are done) and now if i run the command for cmake it is giving me error
<rcurtin[m]>
I guess, I would also add, in a company everything is about making your value and contribution clear (hmm I suppose that is the same in academia too...); even if I was spending all my time all day coding, I would still need to make sure my management chain---and preferably other managers too---recognized the importance/value of the work I was doing
<shrit[m]>
ads, ads, ads
<shrit[m]>
yugansharora01 (yugansharora01): are you using autodownloader
<shrit[m]>
mlpack needs the same
<rcurtin[m]>
ha, maybe I could pay facebook to serve ads to people specifically inside my company... "have you seen the great work Ryan is doing?? click here to learn more!"
<rcurtin[m]>
I always like the publicity part the least... but in some sense that is kind of a really important part
<yugansharora01yu>
@shrit:matrix.org what is auto-downloader? this is my first time doing something like this..
<jonpsy[m]>
I see... It's just that i noticed you guys are busy with meetings after meetings and are only able to spare 30-40mins for normal meetings which makes me scared for my future self 😅
<shrit[m]>
jonpsy: I am happy to do meeting if I can find a job 😆
<rcurtin[m]>
it could be a timezone overlap thing... for instance, in my case, most of my meetings are in the late morning/early afternoon---but my early mornings and late afternoons tend to be pretty free. however, late morning/early afternoon seems to also coincide with when things are going on here 😄
<shrit[m]>
yugansharora01 (yugansharora01): What version of mlpack you are using?
<yugansharora01yu>
@shrit:matrix.org 3.4.2
<rcurtin[m]>
yugansharora01 (yugansharora01): can you tell us a little more about how you installed Armadillo? it seems strange that it is not detecting the version correctly
<jonpsy[m]>
<shrit[m]> "jonpsy: I am happy to do meeting..." <- I came across some opening for data scientist at Microsoft but they're for India
<yugansharora01yu>
@ryan:ratml.org actually now it is giving other error when i wrote the command myself instead of copying from the web page
<rcurtin[m]>
this is the part of the build where mlpack needs to find all its dependencies, so all this output you are seeing is just the results of the search. mlpack needs to be able to find boost and armadillo, at the very least, so it's important to make sure the paths you are passing to the CMake command match where the libraries are actually installed
<rcurtin[m]>
so, does `C:/dev/boost/` have a `boost/` subdirectory with lots of header files in it? and does `C:/dev/boost/lib64-mscv-14.2/` have lots of .dlls inside it?
<yugansharora01yu>
umm no
<zoq[m]>
I guess an easy solution for all of this is to clone the master branch and use the auto-downloader?
<rcurtin[m]>
hmm, that might work---I think that could work on Windows
<rcurtin[m]>
I guess our CI builds do test that after all :)
<zoq[m]>
Getting the dependencies right on Windows is always a little bit tricky.
<yugansharora01yu>
@ryan:ratml.org how do i install boost
<yugansharora01yu>
i just downloaded it and did nothing like armadillo
<zoq[m]>
Correct, the CI installs every dependency manually. So what is described in the build tutorial should still work.
<yugansharora01yu>
i think i downloaded wrong boost
<rcurtin[m]>
agreed, yugansharora01 (yugansharora01) I think that the easier path here might be to clone the git `master` branch from https://github.com/mlpack/mlpack, and then run the CMake command without worrying about any dependencies like boost or armadillo---the branch on Github *should* automatically download and set them up
<zoq[m]>
You have to run cmake with the download option enabled e.g. `cmake -DDOWNLOAD_DEPENDENCIES=ON ..`
<yugansharora01yu>
let me just give it one more shot then i'll do the cloning
<yugansharora01yu>
@marcusedel:matrix.org @ryan:ratml.org thanks for guiding me
<zoq[m]>
rcurtin[m]: It's only the default in the master branch.
<yugansharora01yu>
what is the command for Go bindings...
<rcurtin[m]>
hmm, yugansharora01 (yugansharora01) , you can just be paranoid and disable all bindings: `-DBUILD_JULIA_BINDINGS=OFF -DBUILD_GO_BINDINGS=OFF -DBUILD_PYTHON_BINDINGS=OFF -DBUILD_R_BINDINGS=OFF`