02:55 < travis-ci> mlpack/mlpack#1845 (master - 341a6ca : Marcus Edel): The build was fixed.

02:55 < travis-ci> Change view : https://github.com/mlpack/mlpack/compare/26e35e9ec1aa...341a6ca7d8a5

02:55 < travis-ci> Build details : https://travis-ci.org/mlpack/mlpack/builds/204046917

03:04 < rcurtin> zoq: fixed now, I just had to put the password in quotes

03:07 < zoq> rcurtin: is it working for you?

03:27 < rcurtin> it was a minute ago...

03:27 < rcurtin> maybe the htaccess broke it

03:29 < zoq> rcurtin: Can you also checkout the latest version?

03:30 < rcurtin> zoq: ok, fixed

03:33 < zoq> rcurtin: Thanks, I need to check why shogun failed for almost any dataset. But now it's time to get some sleep.

03:40 < zoq> Also since, we are not only collecting the overall runtime you can also see e.g. for the ALLKNN method at which step hlearn or mlpack is faster; by using the "Metric plots for an algorithm/parameter/dataset combination" view

15:09 < rcurtin> zoq: any comments about the DBSCAN implementation? I figure, I can merge that, then Tham's faster CSV reader, and that's good for mlpack 2.2.0

15:10 < rcurtin> I guess it should be 2.2.0 and not 2.1.2 since we are adding "public API" functionality

15:21 < travis-ci> mlpack/mlpack#1846 (master - b336bd4 : Ryan Curtin): The build passed.

15:21 < travis-ci> Change view : https://github.com/mlpack/mlpack/compare/341a6ca7d8a5...b336bd46d835

15:21 < travis-ci> Build details : https://travis-ci.org/mlpack/mlpack/builds/204218537

17:28 < zoq> rcurtin: I haven't really looked at the PR, let me do this right now.