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

MTL R User Group MaRch Madness

  • Mar 25, 2014 · 6:00 PM
  • Radialpoint

# AGENDA

• 6:00 PM - Free Pizza & Beer thanks to eightysix labs.

• 6:25 PM - Hello World!

• 6:30 PM - Lightning TalkeRs

• 7:30 PM - Intermission

• 8:00 PM - Monthly SpeakeR - Ary Bressane

• 9:00 PM - So long & thanks for all the fish.

# SPONSOR

# PREPARE

Have RStudio downloaded and installed with the following packages:

• dplyr (might need to update the rcpp pacakge)

• ggvis (hosted on github)

# PACKAGE OF THE MONTH (PotM)

data.table - Fast aggregation of large data (e.g. 100GB in RAM), fast ordered joins, fast add/modify/delete of columns by group using no copies at all, list columns and a fast file reader (fread). Offers a natural and flexible syntax, for faster development.


# LIGHTNING TALKERS (5 - 10 mins)

Ira Sutherland

Title: Statistical models of forest recovery indicate dynamic land use trade-off's ... we think. 

Abstract: As background, Ill present my 3 minute thesis (Masters) then jump into the progress and hindrance of my attempt to model non-linear forest recovery. Were talking tweedie distributions (?), heteroscedasticity, known biologically-based hypotheses and other fun challenges. 

Bio: Ira is from Vancouver, BC in his first year of his masters in Elena Bennetts's lab at McGill. His background is in forest ecology and conservation; not so much a statistician but he is stoked on R and here to learn. 

Toby Hocking

Title: Reproducible analysis by writing package versions in R code.

Abstract: Since packages change so frequently, it is often difficult to reproduce old results. In this talk I introduce the works_with_R("3.0.2", changepoint="0.8") declaration which can be used to (1) record package versions, (2) install packages, and (3) warn if installed does not match recorded package versions.

Bio: Toby Dylan Hocking is a post-doc in the McGill Human genetics department who has been writing R code since 2003.

Tyler Rudolph

Title: Modeling woodland caribou space use in northern Quebec using R!

Abstract: An overview of some of the products I have developed to inform conservation planning for the woodland caribou, a threatened species in the boreal forest, including: a) deriving biological seasons from individual movement behaviour (rpart, REEMtree, zoo), b) estimating a weighted kernel density probability function for a population (KernSmooth, raster, maptools, rgdal), c) habitat selection modeling (spatstat, raster, survival), and d) spatial model prediction under different forest management planning scenarios.

Bio: I am a wildlife biologist/consultant conducting applied research and conservation planning for species-at-risk in the boreal forest.  The favourite part of my job is writing R code to explore and analyze ecological data, and I am always seeking better and more efficient ways to do this.  Some of my current goals include writing better functions, interfacing seamlessly with GRASS and multi-core processing.  I am always stoked to interact with and learn from other R enthusiasts.


# MONTHLY SPEAKER (20 - 30 mins)

Ary Bressane

R at Strata, ggvis & dplyr

In this talk Ary will present an overview on R at the O'Reilly Strata Conference (California) and a quick hands-on workshop covering two new packages developed by Hadley Wickham (the man behind ggplot, plyr, lubridate & R Studio): 

• ggvis for visualisation

• dplyr for data manipulation

Ary is Data Scientist at Radialpoint SafeCare Inc., co-organizer of the MTL R User Group and an R enthusiast.


Join or login to comment.

  • jean-philippe m.

    Hey guys ! For those interested in learning R I am organizing a training session that would take place a Saturday (date to be confirm) around 6 hours of training for 50$ (low price = beta testing) The main statistical techniques that will be covered during this session are :

    - Factor analysis
    - Logistics & Linear Regression
    - Clustering analysis
    - Survival analysis
    - Neural network

    If you are interested email me at [masked]. Do not hesitate to propose subject and/or give me comments.

    See you at the next meetup !!

    Cheers,

    Jp

    April 21, 2014

  • Vaughn D.

    Here is Toby's .RPofile if you'd like to automatic have the "works_with_R" function he showed us! https://github.com/tdhock/dotfiles/blob/master/.Rprofile

    March 29, 2014

  • Éric de M.

    just saw the pizza.
    sorry new job out of town i won't be able to make it tonight

    1 · March 25, 2014

    • Vaughn D.

      You missed a great one Eric. Luckily there's always next month!

      March 26, 2014

  • Vaughn D.

    Please remeber to bring your laptop to follow along with Ary's talk. You'll need RStudio installed, along with the following packages:

    • data.table
    • dplyr (might need to update the rcpp pacakge)
    • ggvis (hosted on github)

    If you run into any issues, let us know. See you all tomorrow evening.

    March 24, 2014

  • Tahir

    Hi Ary,
    Both packages (ggvis & dplyr ) are not available (for R version 3.0.2) . So what to do?
    Thanks

    March 18, 2014

    • Vaughn D.

      The solution was to update the Rcpp package. After that, there were a few other packages that had to be updated, but it was different on both platform. Linux was something "httpuv" followed by "shiny". On windows I had to update devtools, to run the ggvis command. That should fix it for you hopefully! Let us know if it worked.

      March 20, 2014

    • Tahir

      Hi, Thank you guys. I did as you suggested and "dplyr" is installed but still "ggvis" is not available (for R version 3.0.2). Btw I am using Windows. Thanks

      March 21, 2014

  • A former member
    A former member

    Hi Vaughn and Matthieu, I would really love to attend and I wouldn't even mind giving a lightning talk but your R meet-up always conflicts with Montreal All-Girl Hack Night. They are also happening in the last Tuesday of the month. Is there anyway you can change it, say to the third Tuesday of the month? You are potentially losing out on a significant women programmer population in Montreal.

    March 18, 2014

    • Vaughn D.

      Hey Annie, you make a good point. We will try and address a better date for the next meetup. Sorry you won't be able to make it this month!

      March 18, 2014

    • Vaughn D.

      Does anyone have a preferred day of the week?

      March 18, 2014

  • Wendy Z.

    Can't make it because foundation concert tonight at 1219 rue crescent Brutopia brew pub! Dropping by if you like music and most important donation to children cancer foundation

    1 · March 18, 2014

  • Vaughn D.

    We are still looking for lightning speakers willing to share 5 - 10 minutes of content on anything R related. If you're interested, let us know by:

    1. commenting below
    2. sending Matthieu an email ([masked])
    3. posting on our **NEW** discourse forum -> http://mtlruser.cloudapp.net

    March 17, 2014

  • A former member
    A former member

    Hey guys, Im new to R. Just working through my first R modelling class at McGill. For this class I am statistically modelling the recovery patterns of coastal BC forest values: things like timber, habitat for an endangered seabird, and carbon storage. I have data from a 240 year chronosequence that measured forests at different ages. Basically it covers forest recovery from clear-cut to old-growth condition. My goal is to figure out which curve functions and parameters I can use to plot recovery. Forest age is on the x axis and forest value (eg timber volume) is the y-axis. I would like these trajectories to reflect known biological processes of forest development.

    R seems the program for the task, but a tough challenge it may be. Id love to present my project (maybe 10-15min) and get some feedback on whats possible. Maybe thoughts or advice.

    1 · March 6, 2014

    • Vaughn D.

      Awesome Ira, this is a great idea!

      March 8, 2014

32 went

Our Sponsors

People in this
Meetup are also in:

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy