04:44 < sreenik[m]> MLJens: I am no expert too but I can try. There are a couple of things that you can take note of. Firstly, the folder where you have the dataset train.csv should be accessible without admin priviledges. Secondly, you can also try and increase your batch size. Unfortunately I don't have the dataset so I cannot try and reproduce the error. Let us know if this solves your issue.

05:05 < jenkins-mlpack2> Project docker ensmallen nightly build build #9: ABORTED in 50 min: http://ci.mlpack.org/job/docker%20ensmallen%20nightly%20build/9/

07:44 < jenkins-mlpack2> Project docker mlpack nightly build build #451: STILL UNSTABLE in 3 hr 30 min: http://ci.mlpack.org/job/docker%20mlpack%20nightly%20build/451/

08:07 < MLJens> jenkins-mlpack2: Hi, the trainingdata are available without admin rights.

08:07 < jenkins-mlpack2> MLJens did you mean me? Unknown command 'Hi,'

08:07 < jenkins-mlpack2> Use '!jenkins help' to get help!

08:07 < MLJens> !jenkins-mlpack2: Hi, the trainingdata are available without admin rights.

08:07 < jenkins-mlpack2> MLJens did you mean me? Unknown command '-mlpack2:'

08:07 < jenkins-mlpack2> Use '!jenkins help' to get help!

08:07 < MLJens> !jenkins-mlpack2: Changing the batch size does not lead to any changing.

08:07 < jenkins-mlpack2> MLJens did you mean me? Unknown command '-mlpack2:'

08:07 < jenkins-mlpack2> Use '!jenkins help' to get help!

09:59 < captain_rubik> Hi, I am a graduate student from IISc India. I am looking to contribute to ML Pack for the next GSoC. Can someone guide me about the projects available?

11:13 < sreenik[m]> MLJens: Haha, jenkins-mlpack2 is not a human. I guess you meant me. Can you upload the dataset (or a similar one) somewhere so that I can replicate your issue? Also, I assume you have you tried the original mnist program (DigitRecognizerCNN) and it works on your system?

11:14 < MLJens> Well, than congratulation - well done your jenkins :)

11:14 < MLJens> I uploaded a file containing only seven fonts to reduce reading time. You can get it from:

11:14 < MLJens> https://drive.google.com/open?id=1UDtuYW3e0pw81piPvXyFgzOnTlcty02G

11:14 < MLJens> The usual MNist example runs well on my machine.

11:15 < sreenik[m]> MLJens: Okay let me try

11:15 < MLJens> I also created a stackoverflow report here: https://stackoverflow.com/questions/57836972/access-violation-in-training-model

11:16 < MLJens> There you'll find my code and the link to my training files.

11:37 < sreenik[m]> MLJens: I don't know if it's fortunate or unfortunate, but your code is working in my system. I got two epochs successfully run before cntrl-c ing out of it.

11:38 < MLJens> Yes, thats perfekt *ironie*

11:38 < MLJens> you run in debug or release mode?

11:39 < sreenik[m]> I don't think I understand completely what you mean by this? Are you using qt creator?

11:40 < sreenik[m]> And is your code compiling without any warning or error?

11:47 < sreenik[m]> I suggest you compile it in the terminal/cmd and run it from there itself. Also, I wouldn't be of much help if this is a Windows centric problem; I have little experience with Windows as far as coding is concerned :(

11:50 < MLJens> I'm compiling in Windows using VS2017.

11:51 < MLJens> I'm searching for a ml framework for c++.

11:51 < MLJens> mlpack was quite nice until now :/

11:51 < MLJens> What I meant is, that if the code is running on your system, thats the worst answer you colud give :) but thats not your fault.

11:51 < MLJens> Because then, the error can be everywhere.

11:56 < sreenik[m]> Try compiling with this (don't forget to replace the file name, you may also exclude -lopenblas if it's not available): g++ -o FontModel FontModel.cpp -lboost_serialization -lboost_program_options -larmadillo -lopenblas -fopenmp -lmlpack

11:57 < sreenik[m]> Oh and exclude -fopenmp also if you aren't using it

12:01 < MLJens> usually, I'm using openmp but I can disable it. I also disabled all optimization stages

12:01 < MLJens> Will try that. But it's really hard, as mlpack is not really telling you, whats wrong.

12:04 < MLJens> Just added a screenshot of my exception to the stackoverflow content.

12:13 < sreenik[m]> I have never seen this before. rcurtinor zoq might be able to help you better here

12:16 < MLJens> Thank you very much anyways!

12:17 < MLJens> I try to get more debug information, because this seams to be a problem in Armadilo

16:50 < travis-ci> mlpack/mlpack#7973 (master - fecb360 : Ryan Curtin): The build has errored.

16:50 < travis-ci> Change view : https://github.com/mlpack/mlpack/compare/444559d71e1d...fecb3601319f

16:50 < travis-ci> Build details : https://travis-ci.org/mlpack/mlpack/builds/582352070

18:31 < Aditi2362> Hello, I'm Aditi Singh, I would love to contribute. I've never contributed earlier thus I'm a little lost as to what to do. Any help would be appreciated..

19:17 < travis-ci> mlpack/mlpack#7975 (master - d716e31 : Ryan Curtin): The build has errored.

19:17 < travis-ci> Change view : https://github.com/mlpack/mlpack/compare/fecb3601319f...d716e312f641

19:17 < travis-ci> Build details : https://travis-ci.org/mlpack/mlpack/builds/582381447