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/
seewishnew has joined #mlpack
seewishnew has quit [Remote host closed the connection]
seewishnew has joined #mlpack
seewishnew has quit [Ping timeout: 250 seconds]
big has joined #mlpack
seewishnew has joined #mlpack
big has quit [Quit: Page closed]
Mulx10 has joined #mlpack
< Mulx10> zoq: thanks for the reviews.
Mulx10 has quit [Ping timeout: 256 seconds]
seewishnew has quit [Remote host closed the connection]
seewishnew has joined #mlpack
seewishnew has quit [Ping timeout: 268 seconds]
seewishnew has joined #mlpack
seewishnew has quit [Ping timeout: 250 seconds]
sreenik has joined #mlpack
astadnik has quit [Quit: Page closed]
govg has joined #mlpack
govg has quit [Ping timeout: 255 seconds]
pd09041999 has joined #mlpack
favre49 has joined #mlpack
< favre49> zoq: rcurtin: How long does testing an implementation usually take?
favre49 has quit [Client Quit]
< zoq> favre49: difficult to put a number on it, sometimes it takes less time as the actual implementation sometimes it takes more time.
Hemal has joined #mlpack
braceletboy has joined #mlpack
< braceletboy> zoq: In the mlpack's benchmarking frame work, why is the knn benchmark named allknn?
< braceletboy> Does that 'all' have any significance
< zoq> braceletboy: allknn was the old name, mlpack_allknn was renamed to knn in mlpack 3.0. Never thought about renaming the benchmark script as well.
< braceletboy> It
< braceletboy> It's okay if I rename it right
< braceletboy> To avoid confusion
< zoq> sure
< braceletboy> Okay. Thanks
< Hemal> Hi zoq, I was going through the available codes for NEAT. I found one on Kenneth O. Stanley's Website: http://nn.cs.utexas.edu/?neat-c
< Hemal> I am trying to figure out how to run it
< Hemal> And a number of softwares are available : https://eplex.cs.ucf.edu/neat_software/#NEAT
< Hemal> From our last chat, you had mentioned "The specification isn't correct", what does that mean ?
< zoq> Hemal: At some point in the learning process the model dosn't evolve anymore, so there is some issue in the Reproduce method, perhaps IsSameSpecies is wrong.
< Hemal> Ok I'll have a look at the code
< Hemal> And should I explore the codes available on the websites that I mentioned?
< zoq> Hemal: That might be a good starting point, also we could use that at some pint for testing as well.
< Hemal> Ok :)
seewishnew has joined #mlpack
seewishn_ has joined #mlpack
seewishnew has quit [Ping timeout: 250 seconds]
< Hemal> zoq: I ran the code from the Website that I mentioned. I ran it 6 times(for XOR). The code executed successfully only twice. Other 4 times, the cursor was blinking on the terminal. Do you think that the original NEAT would have some flaws and perhaps the advanced version of NEAT would have solved them?
< braceletboy> @zoq
< zoq> Hemal: Not sure, maybe the defaults aren't the best?
< braceletboy> In one of the benchmarking scripts, I saw that the predictions were made on the training data itself in the absence of test data
< braceletboy> Should I do this or should not do any prediction at all when test data is not available
< braceletboy> in the new benchmarking scripts?
< zoq> braceletboy: Hm, Ideally someone would pass a test file.
< zoq> braceletboy: I don't mind to use the training set if no test set was passed.
< braceletboy> Yeah I thought the same but in the pull request you opened I found such a file. So, wanted to double check.
< braceletboy> zoq: I think I will go with the former. If someone wants to include the predictions also in the timing metric, he/she will have to give a test file
< zoq> agreed
favre49 has joined #mlpack
< favre49> zoq: Alright, thanks
< favre49> Hemal: The python implementation of NEAT works fine
< favre49> I also found it the best starting point for understanding how to code NEAT as compared to SharpNEAT and other libraries
favre49 has quit [Client Quit]
< Hemal> Thanks ! @favre49
favre49 has joined #mlpack
< Hemal> Thanks! @favre
< favre49> Glad to help :)
favre49 has quit [Quit: Page closed]
seewishn_ has quit [Remote host closed the connection]
seewishnew has joined #mlpack
seewishnew has quit [Remote host closed the connection]
seewishnew has joined #mlpack
hey has joined #mlpack
hey has quit [Client Quit]
seewishnew has quit [Ping timeout: 250 seconds]
seewishnew has joined #mlpack
seewishnew has quit [Remote host closed the connection]
seewishnew has joined #mlpack
seewishnew has quit [Ping timeout: 240 seconds]
seewishnew has joined #mlpack
ahmedmaheralamir has quit [Read error: Connection reset by peer]
ahmedmaheralamir has joined #mlpack
seewishnew has quit [Ping timeout: 250 seconds]
braceletboy has quit [Ping timeout: 246 seconds]
braceletboy has joined #mlpack
Hemal has quit [Quit: Leaving.]
seewishnew has joined #mlpack
seewishnew has quit [Remote host closed the connection]
seewishnew has joined #mlpack
seewishnew has quit [Ping timeout: 250 seconds]
Hemal has joined #mlpack
braceletboy has quit [Ping timeout: 245 seconds]
KimSangYeon-DGU_ has joined #mlpack
< KimSangYeon-DGU_> rcurtin, zoq, ShikharJ: I submitted my draft for GSoC 2019. It would be really great if I get your opinions about it. Can you please have a look at it, if you have a chance?
< KimSangYeon-DGU_> Thanks
KimSangYeon-DGU_ has quit [Quit: Page closed]
pd09041999 has quit [Ping timeout: 245 seconds]
braceletboy has joined #mlpack
< rcurtin> KimSangYeon-DGU_: yes, I'd like to take a look tonight. I would have looked yesterday but I didn't see it on the list of draft proposals
< rcurtin> in any case I can't remember who has asked me to review proposals so I am just selecting them semi-randomly and reviewing :)
pd09041999 has joined #mlpack
severus has joined #mlpack
severus has quit [Client Quit]
saksham189 has joined #mlpack
vivekp has joined #mlpack
naman_ has joined #mlpack
saksham189 has quit [Ping timeout: 256 seconds]
saksham189 has joined #mlpack
naman_ has quit [Ping timeout: 256 seconds]
braceletboy has quit [Ping timeout: 255 seconds]
Im has joined #mlpack
big_ has joined #mlpack
big_ has quit [Ping timeout: 256 seconds]
< Im> Hello! I want to contribute to the RL project.
bhavya01 has joined #mlpack
< bhavya01> rcurtin: I have submitted a proposal on the GSOC website for mlpack-translator project. Please let me know your comments when you get some time.
Kd_ has joined #mlpack
< Kd_> Hey can anybody tell me where should I draft my proposal for review thank u in advance
Kd_ has quit [Client Quit]
< rcurtin> bhavya01: sure, I'll try and take a look if I have time
jeffin143 has joined #mlpack
braceletboy has joined #mlpack
bhavya01 has quit [Quit: Ex-Chat]
braceletboy has quit [Ping timeout: 245 seconds]
braceletboy has joined #mlpack
big_ has joined #mlpack
< big_> hey, zoq or whomever let me know if there's a problem with my code or otherwise https://gitlab.com/hexrays/my-error
< big_> I can't figure out why I'm getting a Mat() out of bounds error
< big_> my data is all there, and I believe the dimensions are correct as well
saksham189 has quit [Ping timeout: 256 seconds]
< Im> I dont understand. On the github Ideas page, since the ideas are already given, what do we have to state in the proposal, the complete algorithm?
Im has quit [Quit: Page closed]
robertohueso has joined #mlpack
saksham189 has joined #mlpack
Hemal has quit [Quit: Leaving.]
ahmedalamir has joined #mlpack
ahmedmaheralamir has quit [Ping timeout: 268 seconds]
sreenik has quit [Quit: Page closed]
vivekp has quit [Ping timeout: 245 seconds]
big_ has quit [Quit: Page closed]
Hemal has joined #mlpack
rish has joined #mlpack
< rish> i want to know that what do you expect from me ?
< rish> I mean do i need to propose something on machine learning or anything related to C++
< zoq> rish: regarding GSoC?
< zoq> there are some ideas listed on https://github.com/mlpack/mlpack/wiki/SummerOfCodeIdeas but feel free to propose anything else
< zoq> but yeah this is a C++ machine learning library
braceletboy has quit [Remote host closed the connection]
jeffin143 has quit [Ping timeout: 256 seconds]
big has joined #mlpack
< big> lol i hope a full NEAT implementation gets merged this summer
Hemal has quit [Quit: Leaving.]
< big> has anyone taken it up yet?
< zoq> big: we received some applications for the project
big has quit [Quit: Page closed]
ahmedalamir has quit [Read error: Connection reset by peer]
< saksham189> zoq: I had submitted my proposal last week and was wondering if you could give me some feedback when you get time.
< zoq> saksham189: sure
< zoq> saksham189: hm, what's the title of the application?
< saksham189> Implementing Essential Deep Learning Modules
akfluffy has joined #mlpack
< zoq> okay thanks
< akfluffy> notifications
< akfluffy> oops sorry new IRC client
< zoq> no worries
witness has joined #mlpack
saksham189 has quit [Ping timeout: 256 seconds]
pd09041999 has quit [Quit: Leaving]
< akfluffy> I think the (submat) error has something to do with my data, because I've tried every combination of layers, rho, etc that I can and all of them do the same thing
< akfluffy> Just not entirely sure what, because mlpack doesn't catch it
Shubhangi has joined #mlpack
< Shubhangi> Is it possible to create own custom layer in mlpack?
< akfluffy> It sure is. Although this might not be the best example, there is one demonstrating a custom Sigmoid layer at https://github.com/mlpack/mlpack/blob/master/src/mlpack/tests/custom_layer.hpp
saksham189 has joined #mlpack
< akfluffy> what should rho be for an RNN? the amount of timesteps (slices) in the dataset?
witness has quit [Quit: Connection closed for inactivity]
Toshal has joined #mlpack
< Toshal> zoq rcurtin ShikharJ: I have shared my proposal on GSOC website. It would be really great to have your feedback on it.
saksham189 has quit [Quit: Page closed]
< Toshal> Also I was having one doubt do we have support for python language for ANN's?
Toshal has quit []
rish has quit [Ping timeout: 256 seconds]
Shubhangi has quit [Ping timeout: 256 seconds]
Hemal has joined #mlpack