[mlpack] mlpack video meeting: April 4

[Ryan Curtin]

Hey there everyone,

After some discussion, schedule wrangling, and playing with
videoconferencing software, we've decided that we'll have the first
mlpack video meeting on

    Thursday, April 4 at 1600-1700 UTC

(so to convert that to some common time zones, from west to east: 9am
PST, 12pm EST, 4pm GMT, 6pm CEST, 7pm MST, 9:30pm IST), and we'll use
the open-source Jitsi videoconferencing software to meet at

https://meet.mlpack.org/mlpack-meeting

If you can't make it, we'll send a followup set of notes to the list for
any further discussion.

I hope to make these semi-regular, so let's see how it goes.

Here's a list of topics we'd like to discuss:

 - Introductions.  (I think very few of us have actually met before!)

 - A review of recent PRs that have been merged and the changes they
   incurred.  I'll put this part together and it'll take no more than 5
   to 10 minutes.

 - Work out a list of things that should be included in mlpack 3.1.0,
   which we should release as soon as possible.  We can go through the
   open PRs, see how much work needs to be done, and decide what to
   include and what can wait until next time.

 - I'd like to decide on a 'direction' (or multiple directions) to push
   mlpack towards in the future, in part because there are a lot of
   toolkits out there, and if we want to make our software useful to
   people, we should have some target we are aiming at, so we can
   differentiate from other toolkits in that respect.  Here are some
   possibilities that we can discuss in a group; we can see which of
   these ideas (or more that get mentioned) are interesting to each of
   us.

    * Low-power and embedded devices.

    * Interoperability with other toolkits.

    * Automatic selection of algorithms for speed (i.e. pick the right
      k-means implementation for a dataset).

    * Better Windows accessibility.

    * Automated release process.

    * Arbitrary-precision data support.

    * Availability of cutting-edge neural network models at levels that
      other toolkits don't easily support.

    * Summer of Docs / Code-In / any other open source
      initiatives like this.

That's a lot of things already, so I think that we'll probably fill up
the hour for sure.  That means there almost certainly won't be time to
discuss individual GSoC projects or proposals; we should focus on the
larger planning goals in this meeting.  We can discuss individual
GSoC details through existing channels, etc., like IRC or the mailing
list. :)

Thanks, and see you then!

Ryan

-- 
Ryan Curtin    | "I'm just going to shoot you once!"
ryan at ratml.org |   - Joseph Dunn