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/
ib07 has quit [Ping timeout: 240 seconds]
ImQ009 has joined #mlpack
ib07 has joined #mlpack
ImQ009 has quit [Quit: Leaving]
ImQ009 has joined #mlpack
ImQ009 has quit [Client Quit]
ImQ009 has joined #mlpack
GitterIntegratio has quit [Quit: killed]
kartikdutt18Gitt has quit [Quit: killed]
ShikharJaiswalGi has quit [Quit: killed]
TanayMehtaGitter has quit [Quit: killed]
anjishnu[m] has quit [Quit: killed]
KritikaGuptaGitt has quit [Quit: killed]
pafmaf[m] has quit [Quit: killed]
bisakh[m] has quit [Quit: killed]
ChaithanyaNaik[m has quit [Quit: killed]
MrityunjayTripat has quit [Quit: killed]
ArunavShandeelya has quit [Quit: killed]
RudraPatil[m] has quit [Quit: killed]
kshitijandmojoGi has quit [Quit: killed]
ZanHuang[m] has quit [Quit: killed]
HimanshuPathakGi has quit [Quit: killed]
PrinceGuptaGitte has quit [Quit: killed]
RoHitRushilGitte has quit [Quit: killed]
RyanBirminghamGi has quit [Quit: killed]
AdityaKandekarGi has quit [Quit: killed]
AbhinavTalariGit has quit [Quit: killed]
franciscoferioli has quit [Quit: killed]
outmanipulateGit has quit [Quit: killed]
saksham189Gitter has quit [Quit: killed]
TaapasAgrawalGit has quit [Quit: killed]
ShivamShaurya[m] has quit [Quit: killed]
ArijitRoyGitter[ has quit [Quit: killed]
Manav-KumarGitte has quit [Quit: killed]
RV784Gitter[m]1 has quit [Quit: killed]
yc7871[m] has quit [Quit: killed]
nishantkr18[m] has quit [Quit: killed]
Valliappan_CAGit has quit [Quit: killed]
GauravSinghGitte has quit [Quit: killed]
JoelJosephGitter has quit [Quit: killed]
AbhishekNimje[m] has quit [Quit: killed]
VSaicharanGitter has quit [Quit: killed]
AniThoGitter[m] has quit [Quit: killed]
RishabhGoel[m] has quit [Quit: killed]
TrinhNgo[m] has quit [Quit: killed]
HavshxjdnaggzGi4 has quit [Quit: killed]
SaraanshTandonGi has quit [Quit: killed]
MostafaNabiehGit has quit [Quit: killed]
AnirudhSinghChau has quit [Quit: killed]
jeffin143[m] has quit [Quit: killed]
AakashkaushikGi4 has quit [Quit: killed]
Eddie-XiaoGitte4 has quit [Quit: killed]
SriramSKGitter[4 has quit [Quit: killed]
PulkitGitter[m] has quit [Quit: killed]
RishabhGarg108Gi has quit [Quit: killed]
RohitKartikGitt4 has quit [Quit: killed]
_slack_mlpack_U0 has quit [Quit: killed]
Saksham[m] has quit [Quit: killed]
theloudspeaker has quit [Quit: killed]
AbdullahKhilji[m has quit [Quit: killed]
EL-SHREIFGitter[ has quit [Quit: killed]
LolitaNazarov[m] has quit [Quit: killed]
kunal12298Gitter has quit [Quit: killed]
Param-29Gitter[m has quit [Quit: killed]
GauravTirodkar[m has quit [Quit: killed]
darkharmony9999[ has quit [Quit: killed]
AryamanBhagatGit has quit [Quit: killed]
DirkEddelbuette4 has quit [Quit: killed]
geek-2002Gitter[ has quit [Quit: killed]
UmarGitter[m] has quit [Quit: killed]
KumarArnav[m] has quit [Quit: killed]
NishantKumarGitt has quit [Quit: killed]
khimrajGitter[m] has quit [Quit: killed]
JatoJoseph[m] has quit [Quit: killed]
AbhinavvermaGitt has quit [Quit: killed]
yashwants19[m] has quit [Quit: killed]
AtharvaKhandaitG has quit [Quit: killed]
say4n has quit [Quit: killed]
jeffinsamGitter[ has quit [Quit: killed]
vigsterkr[m]1 has quit [Quit: killed]
zoq[m] has quit [Quit: killed]
SakshamRastogiG4 has quit [Quit: killed]
PulkitgeraGitte4 has quit [Quit: killed]
abernauer[m] has quit [Quit: killed]
sakshamb189[m] has quit [Quit: killed]
AmeetKumarRanaGi has quit [Quit: killed]
DivyanshKumar[m] has quit [Quit: killed]
AnushKiniGitter[ has quit [Quit: killed]
RakeshBhugraGitt has quit [Quit: killed]
aadarsh-asthanaG has quit [Quit: killed]
siddhant2001Gitt has quit [Quit: killed]
KhizirSiddiquiGi has quit [Quit: killed]
PratikPriyadarsa has quit [Quit: killed]
canewsin[m] has quit [Quit: killed]
Cadair has quit [Quit: killed]
AbishaiEbenezer7 has quit [Quit: killed]
jonpsy[m] has quit [Quit: killed]
sreenik[m] has quit [Quit: killed]
TejasviGuptaGit4 has quit [Quit: killed]
OmarWagih1Gitter has quit [Quit: killed]
kartikdutt18[m] has quit [Quit: killed]
robotcatorGitter has quit [Quit: killed]
AnjishnuGitter[m has quit [Quit: killed]
M2foryou[m] has quit [Quit: killed]
AbhinavGudipati[ has quit [Quit: killed]
jjb[m] has quit [Quit: killed]
ggalan87[m] has quit [Quit: killed]
shrit[m] has quit [Quit: killed]
PranavReddyP16Gi has quit [Quit: killed]
himanshu_pathak[ has quit [Quit: killed]
jacob-earleGitte has quit [Quit: killed]
yuvraj_2701[m] has quit [Quit: killed]
chopper_inbound[ has quit [Quit: killed]
sailor[m]1 has quit [Quit: killed]
DivyanshKumar[m] has joined #mlpack
sreenik[m] has joined #mlpack
GitterIntegratio has joined #mlpack
pafmaf[m] has joined #mlpack
SlackIntegration has joined #mlpack
canewsin[m] has joined #mlpack
darkharmony9999[ has joined #mlpack
Cadair has joined #mlpack
vigsterkr[m] has joined #mlpack
RoHitRushilGitte has joined #mlpack
jacob-earleGitte has joined #mlpack
jjb[m] has joined #mlpack
PulkitgeraGitter has joined #mlpack
franciscoferioli has joined #mlpack
KumarArnav[m] has joined #mlpack
theloudspeaker has joined #mlpack
Valliappan_CAGit has joined #mlpack
AdityaKandekarGi has joined #mlpack
VSaicharanGitter has joined #mlpack
AakashkaushikGit has joined #mlpack
TejasviGuptaGitt has joined #mlpack
OmarWagih1Gitter has joined #mlpack
zoq[m] has joined #mlpack
siddhant2001Gitt has joined #mlpack
ArijitRoyGitter[ has joined #mlpack
RishabhGoel[m] has joined #mlpack
EL-SHREIFGitter[ has joined #mlpack
outmanipulateGit has joined #mlpack
chopper_inbound[ has joined #mlpack
yc7871[m] has joined #mlpack
UmarGitter[m] has joined #mlpack
kshitijandmojoGi has joined #mlpack
GauravSinghGitte has joined #mlpack
AnirudhSinghChau has joined #mlpack
PrinceGuptaGitte has joined #mlpack
shrit[m] has joined #mlpack
saksham189Gitter has joined #mlpack
anjishnu[m] has joined #mlpack
Eddie-XiaoGitter has joined #mlpack
_slack_mlpack_U0 has joined #mlpack
RohitKartikGitte has joined #mlpack
MrityunjayTripat has joined #mlpack
geek-2002Gitter[ has joined #mlpack
LolitaNazarov[m] has joined #mlpack
RyanBirminghamGi has joined #mlpack
kartikdutt18[m] has joined #mlpack
AryamanBhagatGit has joined #mlpack
khimrajGitter[m] has joined #mlpack
ShikharJaiswalGi has joined #mlpack
AbishaiEbenezerG has joined #mlpack
kartikdutt18Gitt has joined #mlpack
TanayMehtaGitter has joined #mlpack
AtharvaKhandaitG has joined #mlpack
AbhishekNimje[m] has joined #mlpack
NishantKumarGitt has joined #mlpack
say4n has joined #mlpack
AbhinavGudipati[ has joined #mlpack
JatoJoseph[m] has joined #mlpack
GauravTirodkar[m has joined #mlpack
PranavReddyP16Gi has joined #mlpack
Manav-KumarGitte has joined #mlpack
MostafaNabiehGit has joined #mlpack
ZanHuang[m] has joined #mlpack
yashwants19[m] has joined #mlpack
AnjishnuGitter[m has joined #mlpack
AmeetKumarRanaGi has joined #mlpack
KhizirSiddiquiGi has joined #mlpack
robotcatorGitter has joined #mlpack
nishantkr18[m] has joined #mlpack
SaraanshTandonGi has joined #mlpack
himanshu_pathak[ has joined #mlpack
SakshamRastogiGi has joined #mlpack
AbhinavTalariGit has joined #mlpack
Saksham[m] has joined #mlpack
sailor[m] has joined #mlpack
jonpsy[m] has joined #mlpack
KritikaGuptaGitt has joined #mlpack
HavshxjdnaggzGit has joined #mlpack
Param-29Gitter[m has joined #mlpack
ShivamShaurya[m] has joined #mlpack
TaapasAgrawalGit has joined #mlpack
HimanshuPathakGi has joined #mlpack
AnushKiniGitter[ has joined #mlpack
JoelJosephGitter has joined #mlpack
abernauer[m] has joined #mlpack
DirkEddelbuettel has joined #mlpack
RakeshBhugraGitt has joined #mlpack
jeffin143[m] has joined #mlpack
aadarsh-asthanaG has joined #mlpack
RV784Gitter[m] has joined #mlpack
SriramSKGitter[m has joined #mlpack
AniThoGitter[m] has joined #mlpack
bisakh[m] has joined #mlpack
RishabhGarg108Gi has joined #mlpack
yuvraj_2701[m] has joined #mlpack
ggalan87[m] has joined #mlpack
ArunavShandeelya has joined #mlpack
kunal12298Gitter has joined #mlpack
M2foryou[m] has joined #mlpack
RudraPatil[m] has joined #mlpack
jeffinsamGitter[ has joined #mlpack
PratikPriyadarsa has joined #mlpack
AbhinavvermaGitt has joined #mlpack
sakshamb189[m] has joined #mlpack
TrinhNgo[m] has joined #mlpack
PulkitGitter[m] has joined #mlpack
AbdullahKhilji[m has joined #mlpack
ChaithanyaNaik[m has joined #mlpack
rcurtin[m] has joined #mlpack
KritikaGuptaGitt has joined #mlpack
jeffin143[m] has joined #mlpack
PratikPriyadarsa has joined #mlpack
AtharvaKhandaitG has joined #mlpack
ArijitRoyGitter[ has joined #mlpack
MrityunjayTripat has joined #mlpack
AnjishnuGitter[m has joined #mlpack
HavshxjdnaggzGit has joined #mlpack
jonpsy[m] has joined #mlpack
siddhant2001Gitt has joined #mlpack
sreenik[m] has joined #mlpack
RohitKartikGitte has joined #mlpack
vigsterkr[m] has joined #mlpack
AbhinavvermaGitt has joined #mlpack
AbhinavvermaGitt has quit [Changing host]
AbhinavvermaGitt has joined #mlpack
< shrit[m]> @zoq it seems that in sac implementation there is only one constructor with neither copy or move constructor. This will require the user to declare it as a pointer inside a class (if sac object is delared as a data member)
< shrit[m]> if this is intended, maybe we can add SAC(SAC const&) = delete ?
< zoq> shrit[m]: Yeah good idea.
< shrit[m]> OK, I will open a pull request for this one, maybe add this line too? SAC(SAC&&) = default
< zoq> shrit[m]: Yes, great thanks.
< AakashkaushikGit> Hey recently we found out that the implementation of Softmin and Softmax function were considerably slow, look at #2657 for example, so this got me thinking that there can be similar implementation that maybe slow such as some loss function, activation function or similar files so should i create a issue which can maybe marked as a `good first issue` and will greatly help in reducing the overall train and inference
< AakashkaushikGit> time for all the algorithms that call such functions over and over again.
< AakashkaushikGit> So the basic thing that can be improved would be something like
< AakashkaushikGit> (edited) ... something like => ... something like:
< zoq> AakashkaushikGit: I don't think a general issue that says we have a bunch of methods let's see if we can improve them is helpful, especially not for a beginner. If you have a specific method in mind and can provide some hints, that's another story.
< AakashkaushikGit> So the basic things that can be improved would be something like:
< AakashkaushikGit> 1. The mathematical implementation.
< AakashkaushikGit> 2. if there is a for loop used to decide some 0, 1 values for a vector or something similar: a vectorization approach or a arma operation can be used to do that.
< AakashkaushikGit> > `zoq on Freenode` Aakash kaushik (Gitter): I don't think a general issue that says we have a bunch of methods let's see if we can improve them is helpful, especially not for a beginner. If you have a specific method in mind and can provide some hints, that's another story.
< AakashkaushikGit> I guess going through each method and trying to find what can be improved in the mathematical examples would be tuff to point out.
< AakashkaushikGit> but i do feel like there could be improvements made to the way these functions are implemented, so i do feel like we should somehow work on that.
< AakashkaushikGit> we can avoid the beginner part, but if you have some idea on how we can approach this, let me know.
< RishabhGarg108Gi> I think we can find some literature online for the most efficient implementation of softmax. Generally the term in the denominator is what that is computationally heavy
< abernauer[m]> zoq: After updating R and packages multiple times in the last couple weeks. Starting to see why you prefer conda and virtual environments.
< zoq> AakashkaushikGit: As I said, if you have a method in mind go for it; but I don't think a general issue is good. To make a good optimization you have to understand the method from start to end, you have to be familiar with armadillo and C++.
< zoq> Optimization is difficult and often frustrating. You start with something, realize it doesn't work in the way you anticipated and you test out another idea, then you realize it's fine as is.
< zoq> abernauer[m]: haha
< zoq> Maybe others have another opinion, but even if there is not an issue, it doesn't mean you can't improve the codebase.
< AakashkaushikGit> @zoq I believe you are true on the optimization part and so I think I would just create PR if i find something like this and this can be done if someone notices something that can be improved or just eyeing through the code.
ib07 has quit [Ping timeout: 260 seconds]
< rcurtin> AakashKaushikGit I'm agreed with zoq here; if you find something that you can optimize, go ahead and do it, but it can often be tricky to do a good job of that, and I don't know how helpful it would be for beginners
< rcurtin> in my view optimizing existing code is a pretty in-depth skill requiring a lot of low-level knowledge and I don't know how many beginners would actually have that skill
ib07 has joined #mlpack
ib07 has quit [Max SendQ exceeded]
ib07 has joined #mlpack
< abernauer[m]> Yeah I would agree with that statement and would have to imagine that get's a lot more complicated given the complexity of C++. I try to avoid over optimizing code over getting a functional working product first.
rcurtin[m] has quit [Quit: Idle for 30+ days]
< AnushKiniGitter[> > Can I open an Issue regarding this and try to get a PR in?
< AnushKiniGitter[> Hey @zoq , I have opened an issue on the stratified split as discussed earlier with a sample implementation and some tests.
< AnushKiniGitter[> (edited) ... tests. => ... tests. Could you take a look at it?
< AakashkaushikGit> @rcurtin , @abernauer yes I do understand the point here and i also agree with it.
< AakashkaushikGit> also the minimum required ensmallen version when installing mlpack is still 2.10.0 while the examples require a minimum version of 2.13.0 as said by @rcurtin in this https://github.com/mlpack/examples/issues/117
< AakashkaushikGit> (edited) ... this https://github.com/mlpack/examples/issues/117 => ... this https://github.com/mlpack/examples/issues/117 so i am not sure but are there any plans of bumping up the minimum version required ?
< zoq> AnushKiniGitter[: Alright, I'll take a look once I have a chance.
< zoq> AakashkaushikGit: I don't think, there is any plan to bump the ensmallen version before mlpack 4.0, we have to make sure ensmallen is available.
< zoq> AakashkaushikGit: But as Ryan pointed out in the issue, we can modify the example and make sure we have the correct version.
< zoq> AakashkaushikGit: I guess you used the mlpack auto download feature to install ensmallen?
< zoq> AakashkaushikGit: So an easy workaround is to replace the current version with the latest.
< zoq> AakashkaushikGit: ensmallen is header only so just copy https://github.com/mlpack/ensmallen/tree/master/include into the correct location and everything should be fine.
ImQ009 has quit [Quit: Leaving]
< AakashkaushikGit> @zoq i installed the ensmallen version from the ubuntu repo, and the issue was solved I did `make install` ensmallen by downloading it through the ensmallen site.
< AakashkaushikGit> I didn't exactly knew how to include such a check in the make file so i didn't create a PR, i guess i would look at the mlpack make file to see how such checks are performed and will create a PR for that example.
< zoq> AakashkaushikGit: mlpack uses cmake and the examples repo a simple Makefile, so it's not exactly the same, but the check is in - https://github.com/mlpack/mlpack/blob/master/CMake/FindEnsmallen.cmake
< zoq> AakashkaushikGit: If you don't like to work on this, if you can open an issue for the problem, that would be great as well.
< AakashkaushikGit> I would like to work on this but right now i am primarily working on #2647 , so will it be okay if i create a PR after sometime ?
< zoq> AakashkaushikGit: absolutely, thanks
< AakashkaushikGit> : D
gaulishcoin0 has joined #mlpack
gaulishcoin has quit [Read error: Connection reset by peer]
gaulishcoin0 is now known as gaulishcoin
CharanpreetSingh has joined #mlpack
< CharanpreetSingh> How much time does it take ( on an average ) to build mlpack on windows , it has been more than 1.5 hours and it seems to be stuck at a place .
< CharanpreetSingh> Config: i5 8th gen , 8gb ram
< zoq> CharanpreetSingh: Hm, this can take some time, I only have the Azure timings - https://dev.azure.com/mlpack/mlpack/_build/results?buildId=4590&view=results
< zoq> CharanpreetSingh: which is around 1:40
The_LoudSpeaker has quit [Remote host closed the connection]
< CharanpreetSingh> Ohh , but I think it is repeating its steps over and again ( ie.. getting stuck at the same command everytime ). Also , the build almost completed ( just 10%-15% remaining ).
< zoq> Do you see any CPU utilization?
< abernauer[m]> Well I didn't just get a freelance job because I am too much of a generalist. Atleast I got some useful advice from the twenty minute video call.