addressalign-toparrow-leftarrow-rightbackbellblockcalendarcameraccwcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-checkcircle-with-crosscircle-with-pluscrossdots-three-verticaleditemptyheartexporteye-with-lineeyefacebookfolderfullheartglobegmailgooglegroupshelp-with-circleimageimagesinstagramFill 1linklocation-pinm-swarmSearchmailmessagesminusmoremuplabelShape 3 + Rectangle 1ShapeoutlookpersonJoin Group on CardStartprice-ribbonShapeShapeShapeShapeImported LayersImported LayersImported Layersshieldstartickettrashtriangle-downtriangle-uptwitteruserwarningyahoo

"Official"­ BARUG August 2013 Meeting

Agenda
6:30 PM - Pizza and networking
7:00 - announcements
7:10 - Norm Matloff on Rdsm
7:45 - John Mount - Science as and interdisciplinary exercise

Norm will present his "much Improved" version of Rdsm, an R package that provides a threads-type programming environment for R, usable both on a multicore machine and across a network of multiple machines. Tests done so far indicate that Rdsm
is sometimes substantially faster than the 'parallel' library in core R. Also, due to the optional use of backing store, it can handle very large matrices 

John writes: I will talk about data science as an interdisciplinary exercise.  Data science is often confused with big data engineering, statistics and machine learning.  It is in fact a collaborative exercise requiring listening, education and consensus building.   I will share my experiences (good and bad) and tricks (good and bad) using R as a platform for data science with an emphasis on topics related to my upcoming book: "Practical Data Science with R."

Dr. Norm Matloff is a professor of computer science (formerly professor of statistics) at the University of California at Davis, author of the Art of R Programming and organizer of the BARUG.

John Mount, has a Ph.D. in computer science from Carnegie Mellon (for work on randomized algorithms and optimization).  He has worked as a computational scientist in biotechnology, a stock trading algorithm designer and managed a research team for a major online shopping site. John Mount is now a principal consultant at Win-Vector LLC and writes for the Win-Vecto blog


Join or login to comment.

  • Andrea R.

    Hey John,

    Is there any reason why you didn't mention data.table? I am starting using them recently (after having had speed issues with plyr) and I think they are great. I have never used sqldf so I would be curious how they compare with sqldf in term of performance and usability.

    Cheers!

    August 22, 2013

    • Alex

      Also data.table's index spec requires you to declare your intentions about data structure too soon - I'd rather this was automatic.

      August 22, 2013

    • Norm M.

      You are correct; bigmemory does not comply with the general FP nature of R. That's "either a bug or a feature," depending on one's point of view. To me, it's absolutely a feature. I've never been a fan of FP personally [but hey, I like global variables, the alleged Scourge of the Programming Universe :-) ]. Michael Kane, bigmemory's coauthor, tells me that the non-FP nature of bigmemory is the reason people like the package; they need the performance. And keep in mind that arguably R itself is moving away from FP somewhat, with its new reference classes, and it has always had some things with side effects.

      August 23, 2013

  • Basanta Dhungana, P.

    I would be excited to hear on tricks of working with big data in R given the memory constraint. Like to see some real life example.

    1 · August 16, 2013

  • allan m.

    Excellent presentations, very informative and helpful. Are Norm Matloff's slides available online? Thank you.

    August 22, 2013

  • John M.

    Thank you for being a generous audience and great questions (and corrections). Here are my slides: http://www.win-vector.com/dfiles/PDSwRTalk.pdf

    August 21, 2013

    • Andrea R.

      Thanks John for the excellent, entertaining, and extremely educative talk.

      August 22, 2013

  • Ashish

    interesting talks - great questions

    August 21, 2013

  • A former member
    A former member

    Sorry, for the last minute change; an issue came up at work that needed tending.

    August 21, 2013

  • Vincent L.

    Thanks

    August 20, 2013

  • Daniel F.

    Sorry Mtn View is out of range for Wednesday. Have fun.

    August 17, 2013

  • A former member
    A former member

    SORRY! I can't make it to Mountain View with a broken ankle. Hope to see you next time.

    August 16, 2013

  • Guillaume G.

    R enthousiast. Fan of Caret package.

    August 16, 2013

  • Guillaume G.

    R enthousiast. Fan of Caret package.

    August 16, 2013

  • akshar d.

    We can do it at Softnet office in Sunnyvale if needed. Pl let me know and I can make arrangements.

    August 15, 2013

  • Juan H.

    +1

    August 12, 2013

  • Rabi K.

    looking forward to listen two great speakers and contributers

    1 · July 31, 2013

Our Sponsors

  • RStudio

    Financial support for meetings

  • O'Reilly

    Our March 29th BARUG meeting onsite at Strata Hadoop World

Sign up

Meetup members, Log in