[R] Suprising R behaviour

[David L Carlson]

My original answer works fine as long as fin_hyp contains only one definition for the line, but if it contains more than one, R will use a different line for each test which is probably not what you want (e.g. the first point will be tested against the first line and the second point against the second line, etc). In that case a loop or apply function would be needed:

> data <- data.frame(x = c(1,2,3,1,1,1), y = c(1,2,3,4,6,7))
> fin_hyp <- data.frame(slope = c(2, 1, -2), constant = c(1, -1, 7))
> outputs <- apply(fin_hyp, 1, function(z) data$y > z[1] * data$x  + z[2])
> outputs
      [,1] [,2]  [,3]
[1,] FALSE TRUE FALSE
[2,] FALSE TRUE FALSE
[3,] FALSE TRUE  TRUE
[4,]  TRUE TRUE FALSE
[5,]  TRUE TRUE  TRUE
[6,]  TRUE TRUE  TRUE

The first column is the result for the first equation (row in fin_hyp) and so on.

David L. Carlson
Department of Anthropology
Texas A&M University


-----Original Message-----
From: David L Carlson 
Sent: Saturday, November 14, 2015 9:57 AM
To: 'Duncan Murdoch' <murdoch.duncan at gmail.com>; Ilgaz S <ilgaz.somer at gmail.com>; r-help at r-project.org
Subject: RE: [R] Suprising R behaviour

You definitely need to learn about the differences between R and the other languages that you are familiar with. You've done some work since you mention the apply() family, but those are just another way of programming a loop. In many cases (including this one), a loop is not needed. Here's your code in a plain text message with some additions:

data <- data.frame(x = c(1,2,3,1,1,1), y = c(1,2,3,4,6,7))
# fin_hyp could just as easily be a data frame or a matrix
fin_hyp <- list(slope = 2, constant = 1)
# R vectorizes the following command and automatically computes the
# result for each row in "data"
outputs <- data['y'] > fin_hyp['slope'] * data['x']  + fin_hyp['constant']
outputs
# Add a plot showing points above and below the line
# ifelse is vectorized so it creates a vector with
# 16 (symbol for solid circle) if above the line and 
# 1 (open circle) if below the line
sym <- ifelse(outputs, 16, 1)
plot(y~x, data, pch=sym)
abline(a=fin_hyp$constant, b=fin_hyp$slope)


David L. Carlson
Department of Anthropology
Texas A&M University

-----Original Message-----
From: R-help [mailto:r-help-bounces at r-project.org] On Behalf Of Duncan Murdoch
Sent: Friday, November 13, 2015 1:16 PM
To: Ilgaz S <ilgaz.somer at gmail.com>; r-help at r-project.org
Subject: Re: [R] Suprising R behaviour

On 13/11/2015 8:11 AM, Ilgaz S wrote:
> Hello everybody, I am new to R and I discovered something that suprise me
> and I have a question about it.
> Today I wanted to return a bit array which represents this:
>
> if( arbitrary point above the line)
>       return TRUE
> else
>       return FALSE
>
> First I tought I would use for loop and access every element of the data.
> Then I tend to use lapply function.
>
> At the end, I accidently done that without using any if/else statement. (
> or for loop )  Here is the code:

I can't read your code (you posted in HTML, don't do that), but it 
sounds as though you have discovered vectorized operations.  These are 
central to good R programming, and are well described in the 
Introduction to R manual.

Duncan Murdoch
>
> data <- data.frame(x= c(1,2,3,1,1,1), y = c(1,2,3,4,6,7))fin_hyp <-
> list(slope=2,constant=1)outputs <- data['y'] > fin_hyp['slope'] *
> data['x']  +fin_hyp['constant']outputs
>
> What is R doing here? It is using loop somewhere inside? Is this code
> more efficient than other methods I mentioned?
>
> Thank you, I.S.
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

______________________________________________
R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.