rcurtin_irc changed the topic of #mlpack to: mlpack: a scalable machine learning library (https://www.mlpack.org/) -- channel logs: https://libera.irclog.whitequark.org/mlpack -- NOTE: messages sent here might not be seen by bridged users on matrix, gitter, or slack
<rcurtin[m]> I'm thinking of releasing a new version once [#3487](https://github.com/mlpack/mlpack/pull/3487) and [#3493](https://github.com/mlpack/mlpack/pull/3493) are merged; anything else I should wait for? maybe [#3491](https://github.com/mlpack/mlpack/pull/3491)? (Mayank Raj zoq )
<zoq[m]> <rcurtin[m]> "I'm thinking of releasing a..." <- We can include #3491, will approve it today and we need to add it to the History file.
<rcurtin[m]> nice 👍️
<rcurtin[m]> that LARS fix took forever... I thought it would be easy but I'm 20+ hours in... so far
<rcurtin[m]> very much hoping that the build is green this time around
<rcurtin[m]> it was just one rabbit hole after another
<zoq[m]> ohh, okay, just seen some of the comments
<rcurtin[m]> yeah, it's always the little details of the paper that get you... the LARS paper says things like "we assume that features are not fully linearly dependent", but that's not the reality in practice... they say "you can make features linearly independent by adding jitter"... but I have a const reference to the user's data... so I end up having to derive things again a little differently, which is pretty tedious
<rcurtin[m]> luckily our test suite is pretty good at unintentionally including degenerate situations
robobub has quit [Ping timeout: 250 seconds]
robobub has joined #mlpack