mlpack IRC logs, 2018-03-01

Logs for the day 2018-03-01 (starts at 0:00 UTC) are shown below.

March 2018
--- Log opened Thu Mar 01 00:00:33 2018
00:47 < robertohueso> Thank you zoq and rcurtin, I'll take a look at it tomorrow
00:47 -!- robertohueso [] has left #mlpack []
01:04 -!- marioloko [5aa50751@gateway/web/freenode/ip.] has quit [Ping timeout: 260 seconds]
02:43 -!- Vss [uid283149@gateway/web/] has quit [Quit: Connection closed for inactivity]
03:41 -!- govg [~govg@unaffiliated/govg] has joined #mlpack
05:23 -!- Rithesh [daf82e6b@gateway/web/freenode/ip.] has joined #mlpack
05:25 -!- Rithesh [daf82e6b@gateway/web/freenode/ip.] has quit [Client Quit]
05:43 -!- daivik [7d10a738@gateway/web/cgi-irc/] has joined #mlpack
07:05 -!- dhoulihan [] has quit [Ping timeout: 265 seconds]
08:17 -!- daivik [7d10a738@gateway/web/cgi-irc/] has quit [Quit: - A hand crafted IRC client]
08:17 -!- daivik [7d10a738@gateway/web/cgi-irc/] has joined #mlpack
08:33 -!- Fen [76b91b9b@gateway/web/freenode/ip.] has joined #mlpack
09:03 -!- Fen [76b91b9b@gateway/web/freenode/ip.] has quit [Ping timeout: 260 seconds]
09:15 -!- ASamir [~asamir@] has joined #mlpack
09:15 -!- ASamir is now known as Samir
09:40 -!- sanghaisubham [2a684d92@gateway/web/freenode/ip.] has joined #mlpack
10:18 -!- sumedhghaisas2 [~yaaic@] has joined #mlpack
10:18 -!- sumedhghaisas [] has quit [Read error: Connection reset by peer]
10:19 -!- sumedhghaisas2 [~yaaic@] has quit [Read error: Connection reset by peer]
10:20 -!- sumedhghaisas2 [~yaaic@] has joined #mlpack
10:21 -!- Samir_ [~asamir@] has joined #mlpack
10:24 -!- Samir [~asamir@] has quit [Ping timeout: 248 seconds]
10:27 -!- Samir_ [~asamir@] has quit [Ping timeout: 240 seconds]
10:29 -!- sanghaisubham [2a684d92@gateway/web/freenode/ip.] has quit [Ping timeout: 260 seconds]
10:30 -!- Samir [~asamir@] has joined #mlpack
10:37 -!- sumedhghaisas2 [~yaaic@] has quit [Read error: Connection reset by peer]
10:37 -!- sumedhghaisas [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has joined #mlpack
10:38 -!- vivekp [~vivek@unaffiliated/vivekp] has quit [Ping timeout: 256 seconds]
10:40 -!- vivekp [~vivek@unaffiliated/vivekp] has joined #mlpack
10:47 -!- qwert123 [ab3dbb6a@gateway/web/freenode/ip.] has joined #mlpack
10:48 < qwert123> hi, I wanted to know: are all the project proposals on the mlpack GSOC ideas page open -- even the ones without any open tickets?
10:52 -!- Samir [~asamir@] has quit [Ping timeout: 240 seconds]
10:59 -!- Samir [~asamir@] has joined #mlpack
11:05 -!- daivik [7d10a738@gateway/web/cgi-irc/] has quit [Quit: - A hand crafted IRC client]
11:05 -!- Samir [~asamir@] has quit [Ping timeout: 248 seconds]
11:07 -!- Samir [~asamir@] has joined #mlpack
11:19 -!- Samir_ [~asamir@] has joined #mlpack
11:23 -!- Samir [~asamir@] has quit [Ping timeout: 245 seconds]
11:33 -!- ShikharJ [0e8bc218@gateway/web/freenode/ip.] has joined #mlpack
11:34 < ShikharJ> qwert123: Yes. Apart from that you are free to propose your own ideas.
11:42 -!- vivekp [~vivek@unaffiliated/vivekp] has quit [Ping timeout: 276 seconds]
11:43 -!- witness [uid10044@gateway/web/] has joined #mlpack
11:44 -!- Samir_ [~asamir@] has quit [Remote host closed the connection]
11:45 -!- Samir [~asamir@] has joined #mlpack
11:45 -!- Samir [~asamir@] has quit [Remote host closed the connection]
11:45 -!- Samir [~asamir@] has joined #mlpack
11:49 < luffy1996> @zoq, did you go through my pr request?
11:49 -!- vivekp [~vivek@unaffiliated/vivekp] has joined #mlpack
12:10 -!- qwert123 [ab3dbb6a@gateway/web/freenode/ip.] has quit [Ping timeout: 260 seconds]
12:17 -!- nikhilgoel1997 [dfbd5e72@gateway/web/freenode/ip.] has joined #mlpack
12:18 -!- ShikharJ [0e8bc218@gateway/web/freenode/ip.] has quit [Ping timeout: 260 seconds]
12:19 -!- nikhilgoel1997_ [uid219701@gateway/web/] has joined #mlpack
12:19 -!- Samir [~asamir@] has quit []
12:22 -!- nikhilgoel1997 [dfbd5e72@gateway/web/freenode/ip.] has quit [Ping timeout: 260 seconds]
12:22 -!- nikhilgoel1997_ is now known as nikhilgoel1997
12:35 -!- samidha [d2d461b0@gateway/web/freenode/ip.] has joined #mlpack
12:46 -!- qi [2d4cd25a@gateway/web/freenode/ip.] has joined #mlpack
12:46 -!- qi is now known as Guest5039
12:47 < samidha> @manish7294: Sorry for the extremely late reply, I fell sick. Here's the very simple code that I failed to execute with the error code:
12:49 < samidha> If anyone else could help me with this, it'd be highly appreciated!
12:50 -!- samidha [d2d461b0@gateway/web/freenode/ip.] has quit [Quit: Page closed]
12:53 -!- Trion [~trion@] has joined #mlpack
12:55 < Guest5039> Why there is no label parameter in k-nearest-neighbor?
12:56 -!- Guest5039 [2d4cd25a@gateway/web/freenode/ip.] has quit [Quit: Page closed]
14:22 -!- witness [uid10044@gateway/web/] has quit [Quit: Connection closed for inactivity]
15:07 -!- travis-ci [] has joined #mlpack
15:07 < travis-ci> mlpack/mlpack#4193 (master - bdbe654 : Ryan Curtin): The build has errored.
15:07 < travis-ci> Change view :
15:07 < travis-ci> Build details :
15:07 -!- travis-ci [] has left #mlpack []
15:32 -!- ImQ009 [~ImQ009@unaffiliated/imq009] has joined #mlpack
15:35 < caladrius[m]> @rcurtin Okay
16:02 -!- avtansh [2bfafdfa@gateway/web/freenode/ip.] has joined #mlpack
16:24 -!- Trion [~trion@] has quit [Quit: Entering a wormhole]
16:30 -!- travis-ci [] has joined #mlpack
16:30 < travis-ci> mlpack/mlpack#4194 (master - 6021e08 : Ryan Curtin): The build passed.
16:30 < travis-ci> Change view :
16:30 < travis-ci> Build details :
16:30 -!- travis-ci [] has left #mlpack []
16:31 < dk97[m]> zoq: rcurtin I have a few questions...
16:34 < rcurtin> dk97[m]: we can try and answer them
16:35 < dk97[m]> is there any separate loss module?
16:35 < rcurtin> I don't follow the question, can you provide some more context please? mlpack is a large library...
16:37 < dk97[m]> Like there are separate files for activation functions, initialization, are there separate files for loss functions like cross entropy and KL Divergence?
16:37 -!- nikhilgoel1997 [uid219701@gateway/web/] has quit [Quit: Connection closed for inactivity]
16:40 < dk97[m]> rcurtin:
16:55 -!- vivekp [~vivek@unaffiliated/vivekp] has quit [Ping timeout: 240 seconds]
17:01 -!- vivekp [~vivek@unaffiliated/vivekp] has joined #mlpack
17:01 -!- travis-ci [] has joined #mlpack
17:01 < travis-ci> mlpack/mlpack#4195 (master - 9f69eb6 : Ryan Curtin): The build passed.
17:01 < travis-ci> Change view :
17:01 < travis-ci> Build details :
17:01 -!- travis-ci [] has left #mlpack []
17:15 -!- qi [2d20fec0@gateway/web/freenode/ip.] has joined #mlpack
17:15 -!- qi is now known as Guest30413
17:17 < Guest30413> How to set labels while training k-nearest-neighbours?
17:19 -!- govg [~govg@unaffiliated/govg] has quit [Ping timeout: 260 seconds]
17:22 < rcurtin> dk97[m]: I guess you are talking about the neural network code
17:23 < rcurtin> have you looked in src/mlpack/methods/ann/ at all?
17:23 < rcurtin> the files are all laid out there
17:23 < rcurtin> Guest30413: the mlpack KNN code solves k-nearest-neighbor search but is not a KNN classifier so labels are not applicable
17:25 < Guest30413> Is ther any other alternative classifier in mlpack?
17:32 -!- Guest30413 [2d20fec0@gateway/web/freenode/ip.] has quit [Quit: Page closed]
17:51 -!- sumedhghaisas [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has quit [Read error: Connection reset by peer]
17:52 -!- sumedhghaisas [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has joined #mlpack
17:57 -!- sumedhghaisas [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has quit [Read error: No route to host]
17:59 -!- sumedhghaisas [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has joined #mlpack
18:01 -!- sumedhghaisas [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has quit [Read error: Connection reset by peer]
18:02 -!- sumedhghaisas [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has joined #mlpack
18:08 -!- daivik [6ad7b755@gateway/web/cgi-irc/] has joined #mlpack
18:18 -!- sumedhghaisas [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has quit [Read error: Connection reset by peer]
18:18 -!- sumedhghaisas [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has joined #mlpack
18:20 < caladrius[m]> zoq: Hey! I tried implementing the Atrous convolution layer. I have a couple of doubts regarding that. First, I couldn't figure out exactly how backpropogation is working in this. But I guess BackwardConvolution and GradientConvolution take care of that. Please point out if I am wrong. Second, I cannot figure out how things are handled in the gradient calculations if we want to have padding. Could you please help me out
18:20 < caladrius[m]> with that? I have opened a PR for the same. All the code is there.
18:22 < dk97[m]> rcurtin: Yes, but I could not find a separate file for Losses like KL Divergence. Rather it is implemented directly in the Sparse Autoencoder.
18:43 -!- daivik [6ad7b755@gateway/web/cgi-irc/] has quit [Quit: - A hand crafted IRC client]
18:44 -!- daivik [6ad7b755@gateway/web/cgi-irc/] has joined #mlpack
18:52 < zoq> dk97[m]: You are right, there is no KL Divergence implemented for the ann code. Not sure it's possible to merge the Sparse Autoencoder code with the existing ann code.
18:53 < zoq> caladrius[m]: Sure, I'll take a look at the code once I get a chance.
18:53 < dk97[m]> Yes, it isnt right now.
18:54 < dk97[m]> Also, exactly how will the framework be? Is it like a model?
18:54 < dk97[m]> I am talking about the VAE GSoC project
18:54 < zoq> dk97[m]: Ah, yeah it will implement the same iterface as the FNN class.
18:54 < dk97[m]> Okay
18:55 < zoq> dk97[m]: Which implementes the Add function to add a new model/layer to an existing model.
18:55 < dk97[m]> Also, is mlpack available on GPU?
18:56 < dk97[m]> I could not find any instructions for it.
18:56 < zoq> dk97[m]: You could use NVBLAS to get GPU acceleration for BLAS functions.
18:57 < zoq> dk97[m]: Take a look at the armadillo documentation for further informations.
18:57 < dk97[m]> Ah okay! I will have a look.
18:57 < dk97[m]> Thanks a lot.
18:58 < dk97[m]> Also, I was thinking of implementing the deconv layer to get the final workings of making a layer and writing tests in mlpack.
18:58 < zoq> dk97[m]: Sounds good.
18:59 < dk97[m]> Is it alright to do so?
18:59 < zoq> dk97[m]: Yes, I think this is a great way to get familair with the codebase.
18:59 < dk97[m]> Also, could you have a look at the Dropout PR as well as the alpha dropout PR when you have time? The builds are passing.
18:59 < zoq> dk97[m]: Sure, I'll take a look at the PR once I get a chance.
19:00 < dk97[m]> Great then, I will let you know if I face any troubles. 🙂
19:00 < dk97[m]> Thanks for helping out!
19:00 < zoq> dk97[m]: Here to help :)
19:01 -!- ShikharJ [0e8bc218@gateway/web/freenode/ip.] has joined #mlpack
19:06 -!- manish7294 [a5e32959@gateway/web/cgi-irc/] has joined #mlpack
19:07 -!- daivik [6ad7b755@gateway/web/cgi-irc/] has quit [Quit: - A hand crafted IRC client]
19:09 < ShikharJ> dk97[m]: Would deconv layers be required in VAE project?
19:10 < dk97[m]> No they wont be required.
19:11 < dk97[m]> I just want to take up the task so that I can understand the code base and flow, and get familiar with making a layer.
19:12 < dk97[m]> ShikharJ:
19:12 < ShikharJ> Ah okay, actually I mentioned this as Deconv layers would be required in one of the ideas that I'm proposing on GANs, and I thought maybe if you could delay that till the projects are announced.
19:12 < ShikharJ> Mentioning this because i'm not too sure of the exact API that would be required.
19:13 -!- sumedhghaisas2 [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has joined #mlpack
19:13 < ShikharJ> So I might have to refactor the code again.
19:13 -!- sumedhghaisas [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has quit [Ping timeout: 252 seconds]
19:13 < ShikharJ> zoq: What should be done in you opinion?
19:14 < dk97[m]> Okay, sure. I will look for another layer to implement. 🙂
19:15 < ShikharJ> dk97[m]: I'm sorry for being a pain.
19:15 -!- daivik [6adb6a78@gateway/web/cgi-irc/] has joined #mlpack
19:15 < dk97[m]> Its alright, not a problem really. I had not started it.
19:16 < zoq> ShikharJ: Ideally, the VAE project follows the existing interface, so I'm not sure you have to refactor the code, but if dk97[m] is happy to switch to another layer, that's fine for me as well.
19:16 -!- daivik [6adb6a78@gateway/web/cgi-irc/] has quit [Client Quit]
19:17 < manish7294> zoq: rcurtin: As LMNN is based on optimizing the accuracy of KNN classifier. So, is it a good idea to propose a KNN classifier along with LMNN as the project will totally need it?
19:17 -!- daivik [6ad7e2e6@gateway/web/cgi-irc/] has joined #mlpack
19:18 < zoq> manish7294: Yes, that sounds reasonable, if I remember right there was a discussion about a KNN classifier based on the existing code.
19:20 < manish7294> zoq: So, we may use existing neighbor search solver. Right?
19:20 < zoq> manish7294: here you go
19:21 -!- ShikharJ [0e8bc218@gateway/web/freenode/ip.] has quit [Ping timeout: 260 seconds]
19:22 -!- ShikharJ [0e8bc218@gateway/web/freenode/ip.] has joined #mlpack
19:22 < daivik> rcurtin: zoq: With regard to issue #356 (Heirarchical Clustering Methods), I'd like to implement BIRCH clustering (its quite popular - next to kmeans, dbscan and a few others...and also scikit-learn has an implementation for it). I've read the paper (; and it doesn't look too difficult
19:22 < daivik> to implement - maybe about a week to implement and write tests. I haven't given the exact implementation and what sort of functions to expose a lot of thought right now - just wanted opinions on how good of an idea this is (worth pursuing? worth binning? any other inputs?)
19:24 < manish7294> zoq: Thanks for providing the link. So, I think here using existing neighbor search solver should simplify the implementation.
19:32 < dk97[m]> zoq: is it alright if I implement quantized fully connected layers?
19:32 < dk97[m]>
19:32 < dk97[m]> This is the paper
19:32 < dk97[m]> The idea can then be extended to conv layers
19:33 < dk97[m]> Also, what is your take on having a separate module for losses?
19:33 -!- sumedhghaisas2 [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has quit [Read error: Connection reset by peer]
19:34 -!- sumedhghaisas [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has joined #mlpack
19:34 < dk97[m]> It could contain the MSE, Absolute, KL Divergence loss.
19:35 < dk97[m]> Also, I think caladrius had done something on huber loss. It all could go in one folder.
19:36 -!- sumedhghaisas2 [~yaaic@] has joined #mlpack
19:38 -!- sumedhghaisas [~yaaic@2a00:79e0:d:fd00:f43d:e2e1:2cef:f129] has quit [Ping timeout: 256 seconds]
19:43 -!- sumedhghaisas2 [~yaaic@] has quit [Ping timeout: 240 seconds]
19:43 -!- ShikharJ [0e8bc218@gateway/web/freenode/ip.] has quit [Quit: Page closed]
19:44 -!- sumedhghaisas [~yaaic@] has joined #mlpack
19:47 < zoq> dk97[m]: I'll take a closer look at the paper later today.
19:48 < dk97[m]> Okay do let me know your views on the paper, as well as the separate module for losses.
19:51 < zoq> daivik: I'm not familiar with BIRCH, so I can't help you right now, maybe Ryan has, I'm sure he will respond once he has a chance.
19:54 -!- daivik [6ad7e2e6@gateway/web/cgi-irc/] has quit [Quit: - A hand crafted IRC client]
19:55 -!- sumedhghaisas [~yaaic@] has quit [Read error: Connection reset by peer]
19:56 -!- sumedhghaisas [] has joined #mlpack
20:02 < rcurtin> daivik: BIRCH would be nice, but there is not time for me to review it now or anytime soon, unfortunately
20:14 -!- sumedhghaisas2 [] has joined #mlpack
20:14 -!- sumedhghaisas [] has quit [Read error: Connection reset by peer]
20:18 -!- djhoulihan [] has joined #mlpack
20:19 -!- daivik [dfb0b01a@gateway/web/cgi-irc/] has joined #mlpack
20:24 -!- manish7294 [a5e32959@gateway/web/cgi-irc/] has quit [Remote host closed the connection]
20:26 < daivik> rcurtin: Thats too bad, I thought that it would have been a nice little self contained thing to work on. I know that the issue is old - but I guess its still open for a reason? Is there some particular deliverable you expect from it?
20:26 < rcurtin> what you described is a good contribution, but what I am saying is that even if you do what you plan to I don't have time anytime soon to review it
20:27 < rcurtin> that said, I don't think would necessarily hurt to close the issue; I had it open as a general entrypoint for contribution, but the GSoC load is now too much for me to keep up with effectively
20:34 < daivik> sorry for flogging the idomatic dead horse, but I really don't mind waiting for a review. And having seen the paper recently, I am itching to code it up. So is it a firm "no, don't do this" from your end or can I work on this and maybe in a distant future (possibly after the current gsoc session ends), you'll review it?
20:38 < rcurtin> I think that is reasonable, but set your expectations for "a year or more"
20:38 < rcurtin> it's not hard for me to quickly review simple contributions but more complex contributions where I need to read the paper, convince myself I am familiar with the paper, etc... this is very time-consuming
20:40 < daivik> right, I completely understand. I guess I'll go ahead with it (with my expectations set to "a year or more") .. and if it is to be it will eventually be :)
20:43 < daivik> On the topic of simple contributions, I was writing tests for mlpack_hmm_generate CLI binding -- there appears to be something wrong with it. When I run $ mlpack_hmm_generate --model_file model-file.xml --length 3; I get this:
20:43 < daivik> [FATAL] Attempted to access parameter --model as type N6mlpack3hmm8HMMModelE, but its true type is PN6mlpack3hmm8HMMModelE!
20:43 < rcurtin> hm, I think this will mean that somewhere in hmm_generate_main.cpp there is a CLI::GetParam<HMMModel>(...) where it should be CLI::GetParam<HMMModel*>(...)
20:43 < rcurtin> try making that change, I bet that will fix it
20:44 < daivik> okay.. will do. Thanks for your inputs.
20:46 < rcurtin> let me know if that doesn't work
20:50 -!- daivik [dfb0b01a@gateway/web/cgi-irc/] has quit [Quit: - A hand crafted IRC client]
21:19 < dk97[m]> zoq: I think before the quantized fully connected paper, we could have a losses module that contains the definitions of different loss functions. That would be more helpful in the general case. Do let me know your thoughts.
21:24 -!- ImQ009 [~ImQ009@unaffiliated/imq009] has quit [Quit: Leaving]
22:17 -!- travis-ci [] has joined #mlpack
22:17 < travis-ci> mlpack/models#22 (master - b0b627f : yash sharan): The build passed.
22:17 < travis-ci> Change view :
22:17 < travis-ci> Build details :
22:17 -!- travis-ci [] has left #mlpack []
--- Log closed Fri Mar 02 00:00:34 2018