Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Data Analysis and Visualization Using R (2014)

varianceexplained.org

Post reply on HN
Print view View original

21–26 of 26 posts

Re: Data Analysis and Visualization Using R (2014)

#21

Post by michaelsbradley »

sgt101 wrote: post #9

I love R, but I have two problems with it that I would like suggestions to deal with. 1. Debugging seems way more primitive than in other languages; I get cryptic messages and really struggle to pinpoint what is happening. Debugging in (free) shiny is even harder, the page says connection closed and I have to guess what has happened. 2) Code structure. R is simply fantastic in REPL and/or RStudio mode for digging aro…

For (2), you may want to look at a couple of module systems for R:

https://github.com/klmr/modules

https://github.com/wahani/modules

Neither is perfect, but I've found them helpful in my projects. They involve much less overhead than writing packages, especially when the modularization I'm trying to achieve is purely internal to my project and I don't intend to publish the code. At the same time, they provide much better encapsulation compared to `base::source`.

Re: Data Analysis and Visualization Using R (2014)

#22

Post by var_explained »

minimaxir wrote: post #2

These tutorials are from 2014. While they provide a good overview of R syntax, a lot has been added to the R-verse such as dplyr, which the author primarily used for his Trump Tweets blog post yesterday. If you are interested in learning R, you may want to read the R for Data Science book ( http://r4ds.had.co.nz/ ) book by dplyr (and ggplot2) author Hadley Wickham. Relatedly, I have my own (slightly more complicated)…

Course author here; I agree about most of the lessons being outdated in the last two years, and that R for Data Science is a great modern source.

I'm working with DataCamp to develop an R course that covers dplyr, tidyr, and other newer additions to the R language.

Re: Data Analysis and Visualization Using R (2014)

#23

Post by minimaxir »

var_explained wrote: post #22
minimaxir wrote: post #2

These tutorials are from 2014. While they provide a good overview of R syntax, a lot has been added to the R-verse such as dplyr, which the author primarily used for his Trump Tweets blog post yesterday. If you are interested in learning R, you may want to read the R for Data Science book ( http://r4ds.had.co.nz/ ) book by dplyr (and ggplot2) author Hadley Wickham. Relatedly, I have my own (slightly more complicated)…

Course author here; I agree about most of the lessons being outdated in the last two years, and that R for Data Science is a great modern source. I'm working with DataCamp to develop an R course that covers dplyr, tidyr, and other newer additions to the R language.

Good to hear! :)

Re: Data Analysis and Visualization Using R (2014)

#24

Post by dandermotj »

denzil_correa wrote: post #13

How good is the support for R when data is large and does not fit in memory?

Very good, just look at sparklyr and the likes. People seem to think there is some hang up here. R has some of the best developed packages for working out of memory.

Re: Data Analysis and Visualization Using R (2014)

#25

Post by dandermotj »

vegabook wrote: post #19

I use R multiple hours every day, I love it, I love the ecosystem, but I can't help thinking that it's showing its age. It does a great job on static analysis and visualization of smaller (< 1 gigabyte) data sets, but is seriously challenged by anything significantly larger, and is unfit for purpose if the data is changing rapidly (eg streaming). I unfortunately am slowly coming to the conclusion that Spark and Flink…

I genuinely think these type of thoughts come from having extensive experience as a programmer, that can consider building out systems that might reach the performance ceiling of R. For 99% of people multicore/distributed architecture will never even be a consideration. But I'm with you, in that having these things from a system engineers perspective would be incredible. There are other implementations of R out there (not just R GNU): Hadley Wickham's discusses them in Advanced R somewhere.

Re: Data Analysis and Visualization Using R (2014)

#26

Post by jcheng »

sgt101 wrote: post #14
jcheng wrote: post #12

Earlier quoted context omitted.

FYI, debugging in Shiny has gotten much better in 0.13.1 and later (you now get stack traces at the console, or in your log file if running on your own Shiny Server, or in your admin console if running on ShinyApps.io).

How? I have looked, but shiny says "debugging is private, talk you your admin".... (I'm running ubuntu on aws)

See http://docs.rstudio.com/shiny-server/#application-error-logs
Post reply on HN