verne.freenode.net changed the topic of #mlpack to: http://www.mlpack.org/ -- We don't respond instantly... but we will respond. Give it a few minutes. Or hours. -- Channel logs: http://www.mlpack.org/irc/
travis-ci has joined #mlpack
< travis-ci> mlpack/mlpack#224 (master - ef29032 : Marcus Edel): The build was broken.
travis-ci has left #mlpack []
< zoq> hm, maybe I can use static weights, let's see
< naywhayare> is it just a random test failure?
< zoq> yeah
< naywhayare> I hate tracking those down... I always end up running 'while(true); do bin/mlpack_test -t SuiteName/TestName; done' and watching it for 100 iterations or so
< naywhayare> the HMM tests and the GMM tests are especially bad for that. I had to keep widening the tolerances; I wonder if the tests are still meaningful...
< zoq> The easiest way is to go with random weights and just increase the iterations, but this would also increase the run time. So, I decided to go with static weights, which might decrease the meaningfulness of the test.
< zoq> I wouldn't call the build broken, because of one failed test ...
travis-ci has joined #mlpack
< travis-ci> mlpack/mlpack#225 (master - f5893d5 : Marcus Edel): The build was fixed.
travis-ci has left #mlpack []
< zoq> http://big.cc.gt.atl.ga.us/job/mlpack%20-%20git%20commit%20test/2439/testReport/ Luckily travis uses a 'good' seed for the CoverTreeAlternateMetricTest test
< naywhayare> I have never been able to track down the actual issue underlying that bug
< naywhayare> the cover tree is incredibly complex and debugging anything to do with it takes forever...
< naywhayare> (for instance, the FastMKS bug I fixed a day or two ago, I'd been working on periodically since May)
< naywhayare> maybe I can get that metric test bug fixed by December :)
< zoq> wow ... on the other side it's nice to have all these tests
< naywhayare> :)
travis-ci has joined #mlpack
< travis-ci> mlpack/mlpack#226 (master - 424383c : Ryan Curtin): The build passed.
travis-ci has left #mlpack []
benchmark has joined #mlpack
benchmark has quit [Client Quit]
< naywhayare> the fast benchmark results are super useful
< naywhayare> now we can see if I improved it :)
travis-ci has joined #mlpack
< travis-ci> mlpack/mlpack#228 (master - 2bd5e78 : Ryan Curtin): The build was broken.
travis-ci has left #mlpack []
< zoq> there it is the CoverTreeAlternateMetricTest failed on travis :(
< naywhayare> yep, I saw that... I guess I'll have to boost the priority of working on that
benchmark has joined #mlpack
benchmark has quit [Client Quit]
benchmark has joined #mlpack
benchmark has quit [Client Quit]
naywhaya1e has joined #mlpack
zoq has quit [Read error: No route to host]
naywhayare has quit [Ping timeout: 244 seconds]
zoq_ is now known as zoq