Juan Natera - Highlights of useR conference - LA R meetup - Sep 2014

Transcript

1. A short introduction
   to dplyr
   Juan Natera
   Los Angeles R Meetup
   09/04/2014
   

   View Slide

2. A bit about me
   • Software Engineer
   • Interested in R and its use for gaining
   insights about data
   • Open Source enthusiast
   • Baseball fanatic
   

   View Slide

3. About dplyr
   • Developed by Hadley Wickham, Chief
   Scientist @ Rstudio.
   • Part of a suite of packages meant to
   facilitate working on the “data pipeline”.
   

   View Slide

4. Why?
   • People spend a lot of time getting data
   ready for analysis
   • Almost no learning curve (just need to
   learn 5 verbs)
   • Improves readability
   • It's FAST
   

   View Slide

5. The data pipeline
   Tidy Transform
   Model
   Visualize
   

   View Slide

6. The 5 verbs
   • flter: remove rows
   • select: choose columns
   • arrange: reorder rows
   • mutate: change data
   • summarize: guess...
   

   View Slide

7. No learning curve, how?
   • First parameter is always a data.frame
   • Other parameters describe what you want
   to do with it.
   • Always returns a new data.frame
   

   View Slide

8. It's Fast
   

   View Slide

9. Let's see some code!
   

   View Slide

10. A great book I picked up at
    useR 2014
    

    View Slide

11. Questions or Comments?
    naterajj@gmail.com
    

    View Slide