but this could be so amazing if it all comes together
2017-02-08 03945, 2017
mayhem
what is challenging about the datasets?
2017-02-08 03928, 2017
alastairp
just so many cool ideas
2017-02-08 03946, 2017
mayhem
I know that feeling, but not in that conext.
2017-02-08 03948, 2017
mayhem
context.
2017-02-08 03937, 2017
alastairp
mostly that it's very easy to say what we want to do
2017-02-08 03942, 2017
alastairp
but actually doing it takes a bunch more
2017-02-08 03949, 2017
alastairp
artist filtering is a good example
2017-02-08 03957, 2017
alastairp
"artist filtering" - 2 words!
2017-02-08 03912, 2017
mayhem
yet, ugly bag of nastiness.
2017-02-08 03921, 2017
alastairp
it touches dataset editor, training, access to the mb database
2017-02-08 03922, 2017
mayhem
what are you filtering artists on? what is the goal?
2017-02-08 03907, 2017
alastairp
the general accepted practise in machine learning is that you shouldn't have more than one example in a dataset by the same artist
2017-02-08 03935, 2017
alastairp
because there is the chance of the algorithm learning the style of an artist, or worse, the production style of an album
2017-02-08 03958, 2017
alastairp
so the idea is to let someone make a dataset, but when they go to the summary view, we can say "you have 150 items in this class, but once we compare to all the other classes you have we have to remove 48 of them because there are items in other class that share the same artist"
2017-02-08 03910, 2017
mayhem
and even there are tricky things. are you going to consider performed as artists? i.e. snoop dogg and calvin broadus in the same set.
2017-02-08 03922, 2017
alastairp
you should write a paper about this!
2017-02-08 03909, 2017
alastairp
I would put money on the fact that no one has ever considered this
2017-02-08 03923, 2017
mayhem
thanks, but... no.
2017-02-08 03949, 2017
alastairp
:D
2017-02-08 03914, 2017
Freso
alastairp_: But we consider it all the time?
2017-02-08 03932, 2017
mayhem
not sure if this is misplaced optimism, but I think that the data sets that MeB/UPF produce are going to be far more interesting and thought out than other things that exist out there.
2017-02-08 03941, 2017
alastairp
anyway, in principal easy, but the live feedback in the dataset editor is somthing I'd really like, which opens up a few more questions
2017-02-08 03959, 2017
alastairp
do we use metadata in the lowlevel files? get it from musicbrainz?
2017-02-08 03908, 2017
alastairp
merged artist ids are the same thing
2017-02-08 03913, 2017
alastairp
Freso: yeah, yeah
2017-02-08 03924, 2017
Freso
;)
2017-02-08 03930, 2017
alastairp
yeah, that's point 1.4
2017-02-08 03941, 2017
alastairp
we already know that the hacky datasets that we made are better
2017-02-08 03947, 2017
alastairp
but they're still not live
2017-02-08 03952, 2017
kyan joined the channel
2017-02-08 03942, 2017
alastairp_ has quit
2017-02-08 03959, 2017
alastairp_ joined the channel
2017-02-08 03905, 2017
alastairp_ has left the channel
2017-02-08 03959, 2017
alastairp
zas: wiki http doesn't redirect to https. is that expected?
2017-02-08 03922, 2017
ZarkBit has quit
2017-02-08 03956, 2017
zas
nope (at least since we decided to move to https), can you create a ticket for it ?
2017-02-08 03940, 2017
alastairp
will do
2017-02-08 03905, 2017
Slurpee has quit
2017-02-08 03920, 2017
alastairp
Freso: are we planning on splitting out the ideas page for SoC like last year's one?
2017-02-08 03929, 2017
alastairp
if yes, I'd be happy to help you work towards that tomorrow morning
2017-02-08 03918, 2017
Freso
alastairp: I don't know.
2017-02-08 03933, 2017
Freso
alastairp: I didn't do the splitting last year, and I'm not really involved with GSoC this year.
2017-02-08 03951, 2017
alastairp
ah! ok
2017-02-08 03905, 2017
mayhem
alastairp: yes, that'd be nice.
2017-02-08 03915, 2017
mayhem
we just need to lift the stuff from last year's page.
2017-02-08 03937, 2017
hibiscuskazeneko joined the channel
2017-02-08 03940, 2017
alastairp
mayhem: done
2017-02-08 03947, 2017
alastairp
splitting now
2017-02-08 03909, 2017
alastairp
I just copied everything directly. We [that is, not me] will need to edit stuff that's change
2017-02-08 03904, 2017
hibiscuskazeneko has quit
2017-02-08 03925, 2017
hibiscuskazeneko joined the channel
2017-02-08 03932, 2017
alastairp
I hate mediawiki
2017-02-08 03936, 2017
alastairp
someone else can fix picard :)
2017-02-08 03947, 2017
SothoTalKer
tsk
2017-02-08 03904, 2017
Freso
alastairp: There you go! SothoTalKer volunteered!
2017-02-08 03913, 2017
alastairp
thanks, SothoTalKer!
2017-02-08 03906, 2017
CallerNo6
alastairp, you can always dump mediawiki drudgery on me
2017-02-08 03953, 2017
CallerNo6
on my plate? Idioms are hard.
2017-02-08 03922, 2017
SothoTalKer
oh, let's see whoever is faster. i need to prepare some food :)
2017-02-08 03939, 2017
ibrahimsharaf joined the channel
2017-02-08 03949, 2017
ibrahimsharaf
Hello developers
2017-02-08 03957, 2017
mayhem
CallerNo6: you did the awesome ideas page for last year, yeah?
2017-02-08 03906, 2017
mayhem
I'd love the same treatment for this year, please.
2017-02-08 03923, 2017
mayhem
In fact, it might be good to make a template for us to copy, so we don't reinvent the wheel each year.
2017-02-08 03946, 2017
CallerNo6
copy]
2017-02-08 03927, 2017
Freso
Hi ibrahimsharaf
2017-02-08 03902, 2017
ibrahimsharaf
I've been researching for GSoC 2017 ideas, and I've been interested in AcousticBrainz (New machine learning infrastructure)
2017-02-08 03928, 2017
ibrahimsharaf
I am good with C++, python, and I have basic ML knowledge
2017-02-08 03943, 2017
ibrahimsharaf
I've been playing with scikit learn for some time, solved some kaggle problems
2017-02-08 03953, 2017
alastairp
mayhem: I've already copied it
2017-02-08 03955, 2017
ibrahimsharaf
So how can I start?
2017-02-08 03921, 2017
alastairp
CallerNo6: this year picard is on the list. I tried to make a nice table with a same-size logo as all the other projects, but I failed
2017-02-08 03936, 2017
alastairp
ibrahimsharaf: wow, I just wrote that 15 minutes ago!
ibrahimsharaf: we're still working out exactly what we want this project to involve
2017-02-08 03925, 2017
CatQuest
Freso: I see that as a weird kissyface emoticon o_O
2017-02-08 03925, 2017
gcilou
Freso: ?
2017-02-08 03941, 2017
alastairp
it's also worth noting that SoC doesn't start for a long time! We have some people interested in working on these projects before SoC starts
2017-02-08 03952, 2017
gcilou
Oh
2017-02-08 03954, 2017
gcilou
Yeah
2017-02-08 03955, 2017
CatQuest
Soc?
2017-02-08 03900, 2017
alastairp
we'd love for people to participate in our projects outside of the program too
2017-02-08 03901, 2017
gcilou
Summer of code
2017-02-08 03906, 2017
alastairp
CatQuest: it's that time of year again
2017-02-08 03909, 2017
CatQuest
arg, stay with oe acronym!
2017-02-08 03917, 2017
CatQuest
one*
2017-02-08 03925, 2017
alastairp
it's always been SoC!
2017-02-08 03935, 2017
alastairp
(it's not GCI)
2017-02-08 03935, 2017
gcilou
Or GSoc
2017-02-08 03939, 2017
CatQuest
before people used GosC
2017-02-08 03942, 2017
alastairp
oh
2017-02-08 03944, 2017
CatQuest
erh GSoC
2017-02-08 03948, 2017
alastairp
sorry, to me they're identical
2017-02-08 03910, 2017
CatQuest
ಠ_ಠ
2017-02-08 03913, 2017
ibrahimsharaf
@Freso I'll check the link out, thanks
2017-02-08 03916, 2017
Freso
Google Service on Chip
2017-02-08 03931, 2017
CatQuest
wow just learned if I shift alt on the - key i get —
2017-02-08 03946, 2017
CatQuest
oohh noo not the hypens!
2017-02-08 03955, 2017
Freso
That's not a hyphen.
2017-02-08 03959, 2017
SothoTalKer
endash
2017-02-08 03914, 2017
alastairp
ibrahimsharaf: if you are interested in AcousticBrainz development in general, it would be a good idea to follow the acousticbrainz-specific getting-started guide at https://wiki.musicbrainz.org/Development/Summer_o… and set up the server too
2017-02-08 03917, 2017
CatQuest
I've officially given up a long time ago so you might as well not bother :D
2017-02-08 03952, 2017
Slurpee joined the channel
2017-02-08 03952, 2017
Slurpee has quit
2017-02-08 03952, 2017
Slurpee joined the channel
2017-02-08 03925, 2017
SothoTalKer
the picard logo is bigger, that's why it also is bigger in the wiki
2017-02-08 03946, 2017
alastairp
yep, I got that far :)
2017-02-08 03925, 2017
SothoTalKer
gcilou was the one with the image editing skills, no? :D
2017-02-08 03940, 2017
gcilou
Si
2017-02-08 03901, 2017
bitmap
mayhem: do you think we should copy over any of the MB ideas from 2016? I don't remember anyone expressing interest in any of them last year...
