07:07 < travis-ci> Soonmok/models#20 (gan - d24c55b : Soonmok): The build passed.

07:07 < travis-ci> Change view : https://github.com/Soonmok/models/compare/e93373be22ce...d24c55b738e4

07:07 < travis-ci> Build details : https://travis-ci.com/Soonmok/models/builds/106115912

09:40 < Shubhangi> I tried all classification algorithms and getting a segmentation fault for all and also when debugged showing the same error for all algorithms in Mat_meat.hpp file. Is this issue is specific to my set up of mlpack. Maybe there is some binding problem in my set up of mlpack.Does anybody else also facing this issue?

10:37 < jeffin> rcurtin : checkmatrices ( ) function doesn't have all the datypes check like there is no option to compare integer datatype in matrix, i think we could implement template programming for that. OR is it by choice that function is implemented without template.

10:38 < ShikharJ> Shubhangi: Mat_meat.hpp is a file in Armadillo for dense matrices. Maybe what you are observing is a issue of compatibility with an older version of Armadillo installation, or less probably, a deviation in the underlying API of the current Armadillo master. You might wanna open an issue for that, and mention your build information, the scripts used and the error message.

10:39 < jeffin> I guess may be you intentionally didn't use template*

11:46 < abhiji> can i get the update of the mlpack regarding the idea list?

13:24 < rcurtin> jeffin: yeah, if we need to check integer matrices, we have to use BOOST_REQUIRE_EQUAL() inside instead of BOOST_REQUIRE_CLOSE() so it's a bit different

13:24 < rcurtin> I don't mind adding another overload though

13:24 < rcurtin> in fact I thought there already was an overload for arma::Mat<size_t>, correct me if I'm wrong

13:27 < jeffin> Yes * it was there

13:27 < jeffin> Rcurtin : But we couldn't have check for negative number if its size_t

13:30 < rcurtin> jeffin: yeah, so I guess if you have a matrix that may have negative numbers, then you can't use that overload and might need to add another one

13:31 < jeffin> Okay thanks

13:34 < jeffin> rcurtin : one more query , is there any function under preprocessing step to scale the features, like maxabsolute scaler - to scale each feature by its maximum absolute value

13:34 < jeffin> And many other such scaler functions*

13:39 < rcurtin> jeffin: no, I don't think we currently have any support for that

14:00 < jeffin> Okay

15:21 < Shubhangi> I have armadillo version >6.500.0, I don't think so that there would be any compatibility issue.

16:25 < jenkins-mlpack2> Project mlpack - git commit test build #137: STILL UNSTABLE in 41 min: http://ci.mlpack.org/job/mlpack%20-%20git%20commit%20test/137/

16:25 < jenkins-mlpack2> Ryan Curtin: Merge branch 'Yashwants19-master'

16:39 < ReemGody> Hello everyone, I am Reem. I have expressed my interest in working on the NEAT api this summer as part of GSOC and now I have written my proposal . I hope you could take a look at it. I have sent it on the mailing list. I will really appreciate your feedback.

17:29 < Reem> I dont know whether the mailing list was the correct place to send my proposal . I am afraid it is not

17:30 < rcurtin> Reem: no, it's just fine to post it there, but there are a lot of proposals so mentors may or may not be able to find the time to review it

19:16 < jenkins-mlpack2> Yippee, build fixed!

19:16 < jenkins-mlpack2> Project mlpack - git commit test build #138: FIXED in 45 min: http://ci.mlpack.org/job/mlpack%20-%20git%20commit%20test/138/

19:16 < jenkins-mlpack2> noreply: Merge pull request #1820 from MuLx10/python_docs

23:51 < KimSangYeon-DGU> ShikharJ: I want to know your opinions about https://github.com/mlpack/mlpack/pull/1666.

23:51 < KimSangYeon-DGU> ShikharJ: Is it fine to leave two-spaces in comment, because they are used all over the place?