00:46 < jeffin> Karl : could you post error message

00:46 < jeffin> I would try to debug it have a glance at error message

01:40 < Guest33421> zoq: no, I just mean in general. After training a set on an NN with Linear only layers, the MSE is the same as if I did not train it.

02:27 < Karl> jeffin: Thanks for your help. Do you mean post error here or on github?

02:33 < Karl> jeffin: I paste the part reporting error. /var/folders/fr/bf53vrjd5t5c63kql06b_9bw0000gn/T//ccENAysM.s:423594:11: warning: section "__datacoal_nt" is deprecated .section __DATA,__datacoal_nt,coalesced ^ ~~~~~~~~~~~~~ /var/folders/fr/bf53vrjd5t5c63kql06b_9bw0000gn/T//ccENAysM.s:423594:11: note: change section name to "__data" .section __DATA,__datacoal_nt,coalesced ^ ~~~~~~~~~

02:34 < Karl> jeffin: it won't let me post all the error.

02:34 < Karl> jeffin: [ 14%] Linking CXX executable ../../../../bin/mlpack_cf Undefined symbols for architecture x86_64: "boost::program_options::to_internal(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)", referenced from: boost::program_options::basic_parsed_options<char> boost::program_options::parse_command_line<char>(int, char const* const*, boost::program_options::options_description const&, int, boost:

02:35 < Karl> jeffin: this is the starting of error. I finished the `cmake ../` and when I use `make -j6`, it report errors.

02:36 < Karl> jeffin: the following the end of error.

02:36 < Karl> jeffin: ld: symbol(s) not found for architecture x86_64 collect2: error: ld returned 1 exit status make[2]: *** [bin/mlpack_cf] Error 1 make[1]: *** [src/mlpack/methods/cf/CMakeFiles/mlpack_cf.dir/all] Error 2 make: *** [all] Error 2

02:44 < jeffin> Did u try to decrease the number of cores..??

02:44 < jeffin> May be try with make -j2

02:44 < jeffin> Or something

02:45 < jeffin> Do u have boost library..??

02:56 < Karl> jeffin: I am trying with one core now. but currently I already have some error.

02:56 < Karl> jeffin: [ 14%] Linking CXX executable ../../../../bin/generate_pyx_perceptron dyld: lazy symbol binding failed: Symbol not found: __ZNSsaSEOSs Referenced from: /Users/kungangzhang/Documents/OneDrive/Northwestern/Job/Programming/GSoC/mlpack-master/build/bin/generate_pyx_perceptron (which was built for Mac OS X 10.13) Expected in: /usr/lib/libstdc++.6.0.9.dylib dyld: Symbol not found: __ZNSsaSEOSs Referenced from: /Users/kungangz

02:56 < Karl> jeffin: I have boost 1.60

02:57 < Karl> jeffin: I followed the instruction of building from source and installed all the dependency and make sure version is correct

03:02 < Guest33421> jeffin: try adding -lboost_system and -lboost_serialization to your flags

03:02 < jeffin> Run make clean once and try to run with one core*

03:06 < Karl> jeffin: the flags are on make -j1 like this 'make -j1 -lboost_system and -lboost_serialization', is that correct?

03:15 < jeffin> Yup does the error still persist..??

03:22 < Karl> jeffin: I just run, the error still here.

03:22 < Karl> [ 55%] Built target python [ 55%] Linking CXX executable ../../../../bin/mlpack_adaboost Undefined symbols for architecture x86_64: "boost::program_options::to_internal(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)", referenced from: std::vector<std::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::basic_string<char, std::char_traits<char>, std::allocator<c

03:22 < Karl> .cpp.o vtable for boost::program_options::typed_value<std::basic_string<char, std::char_traits<char>, std::allocator<char> >, char> in adaboost_main.cpp.o ld: symbol(s) not found for architecture x86_64 collect2: error: ld returned 1 exit status make[2]: *** [bin/mlpack_adaboost] Error 1 make[1]: *** [src/mlpack/methods/adaboost/CMakeFiles/mlpack_adaboost.dir/all] Error 2 make: *** [all] Error 2

03:23 < Karl> jeffin: I will delete files in build. And cmake ../ and make again

03:23 < Karl> this is mlpack-master

03:23 < Karl> not mlpack-3.0.4

03:24 < Karl> I am running `cmake -D DEBUG=ON -D PROFILE=ON -D MATHJAX=ON ../` first

03:24 < Karl> and then `make -j1 -lboost_system and -lboost_serialization`

03:25 < Karl> `make -j1 -lboost_system -lboost_serialization`

03:25 < Karl> waiting for results

03:43 < jeffin> It's showing undefined symbols

03:43 < jeffin> That means either it's not built or issues with something it depends on

05:28 < Karl> jeffin: I recheck the boost library. Looks like I once changed the version of boost and might be not linked well. I reinstall and boost library can be used.

05:28 < Karl> jeffin: there still some `fail`. I don't know if it is error. I am still waiting for the installing results

05:29 < Karl> jeffin: [ 22%] Linking CXX executable ../../../../bin/generate_pyx_emst dyld: lazy symbol binding failed: Symbol not found: __ZNSsaSEOSs Referenced from: /Users/kungangzhang/Documents/OneDrive/Northwestern/Job/Programming/GSoC/mlpack-master/build/bin/generate_pyx_emst (which was built for Mac OS X 10.13) Expected in: /usr/lib/libstdc++.6.0.9.dylib dyld: Symbol not found: __ZNSsaSEOSs Referenced from: /Users/kungangzhang/Documen

05:29 < Karl> jeffin: https://github.com/lanl/LaGriT/issues/107

05:33 < Karl> this post talked about this issue, but seems no illuminating solution.

06:08 < Shine_> Guest33421: about your question on a NN with just linear layers. The NN will learn but it will just act in the same way as a linear regression model because it does not have any non-linearity in between different layers.

06:08 < Shine_> So you may think of it as a model with just 1 linear layer

07:24 < grixx> HI

07:49 < s_tian> Hi all! I'm Shi Tian, a yr3 Computer Science student, I'm currently working on Spiking Neural Networks in undergrad research and would really like to know the possibility of implementing them (overhaul of perceptron model and layers including forward/back propagation) in mlpack for GSoC this year?

11:21 < mulx10> Karl: Was your cmake successful?

12:13 < rcurtin> zoq: I woke up to find your visualization on the front page of Hacker News:

12:13 < rcurtin> https://news.ycombinator.com/item?id=19467891

12:13 < rcurtin> :)

12:16 < Udit> Hi.

12:17 < Udit> Hi I am interested in working in mlpack organization on a project under gsoc. I wanted to know where I can find the list of projects and would any language other than C++ be needed...

12:19 < sreenik> Udit: Hi! You can find the ideas' list here https://github.com/mlpack/mlpack/wiki/SummerOfCodeIdeas

12:19 < Udit> I am a first year BTech student in IIIT Delhi, India. I am well versed in C++, python and have decent knowledge of Java.

12:20 < Udit> Also I am very interested in machine learning and also have fair knowledge in this feild

12:21 < sreenik> mlpack is a C++ based library but I think one of the ideas for this year's GSOC is for bindings to other languages like Java, etc. You can take a look. I am just another aspirant like you, the maintainers would be able to guide you better

12:21 < Udit> sreenik thanks a lot for the guidance

12:22 < sreenik> :)

13:23 < Udit> I have seen the list of projects among other things on the website. String Processing Utilities project is what I decided to go for. It felt interesting to me and I felt it to be within my scope to do it.

13:24 < Udit> Besides I wanted to learn somethings which would be a part of this project. Sir let me know when I can apply and how are the applications shortlisted..

14:17 < zoq> rcurtin: https://twitter.com/rasbt/status/1108496200154730496 ... this one is nice

14:17 < zoq> Udit: Hello, see https://github.com/mlpack/mlpack/wiki/Google-Summer-of-Code-Application-Guide for how to apply.

14:18 < zoq> Udit: Also http://mlpack.org/gsoc.html should be helpful.

14:21 < rcurtin> zoq: very cool, I hope it is useful for his class :)

14:26 < Udit> Thanks for the reply..

16:18 < gauravcr7rm> zoq: Hello , please brief me about test we have to incorporate in our proposal . What tests we have to incorporate other than benchmark test and semantic test?

16:21 < rf_sust2018> hello there!

16:22 < rf_sust2018> i'm interested in gsoc 2019

16:23 < rf_sust2018> let me know if there were any attempt to build automatic bindings for mlpack in Java/Scala the previous year or earlier

16:23 < rf_sust2018> i've browsed through the recent mailing list and have found out that someone is interested in automatic bidings in ruby

22:24 < ReemGody_> Hi everyone :D . I am Reem and I am senior computer engineering student. I am interested in participating in GSOC this year (I am mainly interested about the NEATS project) . I have tried writing my neural network before and I am really eager to contribute to mlpack

22:26 < ReemGody_> What I have currently done is that I have read the guide lines for participating in GSOC , I built the library from src and I have started exploring the code and answering questions on github. I want to contribute to issues, but whenever I open an issue, I find that someone else is working on it. I would really be very glad if some one can guide me in my first steps :D