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/
mindlifter has joined #mlpack
munch29 has joined #mlpack
munch29 has quit [Remote host closed the connection]
TTtie-M has joined #mlpack
TTtie-M has quit [Remote host closed the connection]
__idiot__ has joined #mlpack
__idiot__ has quit [Remote host closed the connection]
< jonpsy[m]>
zoq: About the remote server, will it retain the files and dev-env setup? Also, who all have access to that server ? Because I might git-login there and it might be "not secure" if anyone other than us 3 have access to the server. Or perhaps there is a better way to go about this, let me know :)
< jonpsy[m]>
oh ..and also, can i know the specs of the system, just excited :D
< zoq>
jonpsy[m]: All files will be retained, yes; but probably not advisable to store secure information on the system, so I would not store e.g. the git credentials.
< zoq>
jonpsy[m]: Will create the account tomorrow.
< zoq>
jonpsy[m]: I guess we have to see if the latency is good enough for you.
< zoq>
Hopefully it's not to bad.
< jonpsy[m]>
Ah yes! If the server is in U.S that'd be problem
< jonpsy[m]>
could you try banglore server?
< zoq>
Germany.
< zoq>
Canada might work as well.
< jonpsy[m]>
canada is too far :)
< zoq>
I'll setup the account, and we will see if it works, I guess.
< jonpsy[m]>
is banglore server out of question...?
< jonpsy[m]>
i suppose you'll have to pay again for it, correct?
< zoq>
I don't have a machine in banglore :(
< jonpsy[m]>
Bummer
< zoq>
In this case the machine is a build node we use for jenkins as well, it's currently not really used.
< jonpsy[m]>
Hmm, i mean if the training is fast (which shouldnt have to do with my latency), i can cope up with it.
< jonpsy[m]>
germany works
UmarJ has quit [Ping timeout: 245 seconds]
nsuperbus_ has joined #mlpack
nsuperbus_ has quit [Remote host closed the connection]
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
nullburrito has joined #mlpack
nullburrito has quit [Read error: Connection reset by peer]
theCorvus has joined #mlpack
theCorvus is now known as Guest46113
Guest46113 has quit [Remote host closed the connection]
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
thykka has joined #mlpack
thykka has quit [Remote host closed the connection]
rak-zero has joined #mlpack
rak-zero has quit [Remote host closed the connection]
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
Panther13 has joined #mlpack
Panther13 has quit [Remote host closed the connection]
UmarJ has joined #mlpack
jindraj3 has joined #mlpack
jindraj3 has quit [Remote host closed the connection]
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
< HarshVardhanKuma>
Hi guys. I'm having trouble with decision trees cli binding. It seems to reject the .bin file input (which I generated by loading a CSV file in armadillo as a matrix and then saving it). The same bin input file works with other mlpack cli bindings like random_forests and adaboost.
TrevorH has joined #mlpack
TrevorH has quit [Remote host closed the connection]
Rounin has joined #mlpack
Rounin has quit [Remote host closed the connection]
kirch has joined #mlpack
kirch has quit [Remote host closed the connection]
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
sponix2ipfw has joined #mlpack
sponix2ipfw has quit [Remote host closed the connection]
ImQ009 has joined #mlpack
AbleBacon has joined #mlpack
AbleBacon has quit [Remote host closed the connection]
nethope has joined #mlpack
nethope has quit [Killed (Sigyn (Spam is off topic on freenode.))]
so26 has joined #mlpack
so26 has quit [Remote host closed the connection]
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
mukmuk29 has joined #mlpack
mukmuk29 has quit [Remote host closed the connection]
RaaFaTKhaLeD-t22 has joined #mlpack
RaaFaTKhaLeD-t22 has quit [Remote host closed the connection]
< HarshVardhanKuma>
Hi guys, there is one more doubt regarding mlpack_adaboost cli binding. It does not expose any option to choose the number of estimators - rather there is an option for the number of iterations. I suppose both are equivalent?
srslypascal__ has joined #mlpack
srslypascal__ has quit [Remote host closed the connection]
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
TecnoBrat23 has joined #mlpack
TecnoBrat23 has quit [Remote host closed the connection]
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
< swaingotnochill[>
zoq is there any way to store the dataset I will be working on server..
< swaingotnochill[>
(edited) ... I will be working on server.. => ... I am working on server..?
UmarJ has quit [Ping timeout: 258 seconds]
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
Davey has joined #mlpack
Davey has quit [Remote host closed the connection]
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
UmarJ has joined #mlpack
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
DataDrake21 has joined #mlpack
DataDrake21 has quit [Remote host closed the connection]
Tokoyami has joined #mlpack
Tokoyami has quit [Remote host closed the connection]
< zoq>
swaingotnochill[: Which dataset are we talking?
< swaingotnochill[>
I want to upload the California housing price dataset
< zoq>
Ahh, there is no automated process for that, if you can post the link here, pretty sure @rcrutin can upload the dataset for you.
< swaingotnochill[>
Okay...then let me just finish the first project..I will then ask Ryan to upload the dataset
< rcurtin[m]>
yep, that's easy for me to do, but I don't have an automated script to do it... just send a URL and I'll add it to datasets.tar.gz :)
haxki has joined #mlpack
haxki has quit [Remote host closed the connection]
UmarJ has quit [Ping timeout: 272 seconds]
UmarJ has joined #mlpack
mindlifter has quit [Remote host closed the connection]
< rcurtin[m]>
zoq: I'm looking at the `Select` layer right now... the comment is that it "selects the specified column from a given input matrix"---but in the ANN toolkit, a column is a single point in a batch. so is the idea here that we are reducing a batch to a single point? or should instead this be selecting a single dimension (or set of dimensions) from each input point?
< rcurtin[m]>
I couldn't find any uses of the `Select` layer in the mlpack codebase, so I wasn't sure; maybe there are some uses in the examples or models repository though
< zoq>
It will select a specific dimension, but honestly if there is no test for it, I would say let's remove the layer.
mindlifter has quit [Remote host closed the connection]
< rcurtin[m]>
oh, it is tested, but just as a standalone layer, not in the context of a larger network
< rcurtin[m]>
I'll adapt it to use a specific dimension instead of a column 👍️
mindlifter has joined #mlpack
< zoq>
rcurtin[m]: looks like #49 is red because of some connection issue
< rcurtin[m]>
yeah, I saw that... transient DNS failure on dealgood it seems like?
< rcurtin[m]>
I'll watch and see if it keeps happening
< zoq>
wondering if it makes sense to print out a warning and continue, because stb is optinal anyway
< rcurtin[m]>
yeah, actually the better question is why STB isn't being detected on the system... I installed `libstb-dev`
< zoq>
ohh you did, in this case yes that is the better question
< swaingotnochill[>
zoq is there any way to read both the categorical and numerical data at same time?
< swaingotnochill[>
zoq is there any way to read both the numerical and categorical data at the same time? the data::Load() only loads the numerical data.
< rcurtin[m]>
zoq: oh! of course. I am running all this inside a docker container. so I just need to rebuild all the containers with stb inside them...
mindlifter has quit [Remote host closed the connection]
mindlifter has joined #mlpack
< HarshVardhanKuma>
I wanted to check if mlpack bindings have any option to run on multicores. I could not find any cli options for number of jobs. I wanted to know if it was possible?
runrin27 has joined #mlpack
< rcurtin[m]>
HarshVardhanKumar (Harsh Kumar): if you compile with OpenMP, or if you are using OpenBLAS, this will happen automatically (depending on the algorithm) :)
runrin27 has quit [Remote host closed the connection]
< rcurtin[m]>
also I saw your other messages... can you open an issue for the .bin problem you had with `mlpack_decision_tree`? and you are right that the number of iterations is the number of estimators for AdaBoost 👍️
< zoq>
swaingotnochill[: You can read numerical and categorical data at the same time.