Dear R experts,
I am having difficulty using loops productively and would like to please
ask for advice. I have a dataframe of ids and groups. I would like to
break down the dataframe into groups, find the unique sets of ids, then
reassemble. My thought was to use a loop, but I have been unable to finish
this loop in a logical way. I would like to find the unique ids for group
1, group 2, etc., and rbind these back together. However, I am unclear how
to do this because in other attempts my final product is always a part of
the last run loop. My way of working around this has been to write.csv and
then re read at the end, which is so clumsy. Previously, I have used a
store matrix for individual cells. 1. Is there a better way to approach
this? 2. How can I combine parts of matrices to other parts created in
prior loops?
I have created a primitive example below. Each of the groups varies in
number, so my repetitive example below is not accurate. In my real data,
ids repeat often within groups.
Thank you so much, Matt
example <- data.frame(id=rep(
( abs(round(rnorm(50,mean=500,sd=250),digits=0)))
,3), group=rep(1:15,10))
example <-example[with(example,order(id,group)),]
for (i in 1:15) {
ai <- example[example[,2]==i,][!duplicated ( example[example[,2]==i,][,1]
),]
write.csv(ai, paste('a',i,'.csv',sep=""))
}
b1<-read.csv('a1.csv')
b2<-read.csv('a2.csv')
b3<-read.csv('a3.csv')
b4<-read.csv('a4.csv')
b5<-read.csv('a5.csv')
b6<-read.csv('a6.csv')
b7<-read.csv('a7.csv')
b8<-read.csv('a8.csv')
b9<-read.csv('a9.csv')
b10<-read.csv('a10.csv')
b11<-read.csv('a11.csv')
b12<-read.csv('a12.csv')
b13<-read.csv('a13.csv')
unis2 <-
rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind
(b1,b2),b3),b4),b5),b6),b7),b8),b9),b10),b11),b12),b13)
[[alternative HTML version deleted]]
Is this close to what you want:> example <- data.frame(id=rep(+ ( abs(round(rnorm(50,mean=500,sd=250),digits=0))) + ,3), group=rep(1:15,10))> example <-example[with(example,order(id,group)),] > > uniqueIDs <- do.call(rbind+ , lapply(split(example, example$group), function(.grp){ + data.frame(group = .grp$group[1L] + , unique = paste(unique(.grp$id), collapse = ',') + ) + }) + )> > uniqueIDsgroup unique 1 1 175,186,226,318,458,609,633,682,719,869 2 2 290,361,408,487,500,559,773,779,809,1459 3 3 201,283,371,470,479,513,576,719,810,1037 4 4 4,18,118,212,291,482,738,818,1008,1095 5 5 115,243,253,385,396,447,547,625,790,1156 6 6 175,186,226,318,458,609,633,682,719,869 7 7 290,361,408,487,500,559,773,779,809,1459 8 8 201,283,371,470,479,513,576,719,810,1037 9 9 4,18,118,212,291,482,738,818,1008,1095 10 10 115,243,253,385,396,447,547,625,790,1156 11 11 175,186,226,318,458,609,633,682,719,869 12 12 290,361,408,487,500,559,773,779,809,1459 13 13 201,283,371,470,479,513,576,719,810,1037 14 14 4,18,118,212,291,482,738,818,1008,1095 15 15 115,243,253,385,396,447,547,625,790,1156>On Thu, Feb 16, 2012 at 11:15 AM, Matt Spitzer <matthewjspitzer at gmail.com> wrote:> Dear R experts, > I am having difficulty using loops productively and would like to please > ask for advice. ?I have a dataframe of ids and groups. ?I would like to > break down the dataframe into groups, find the unique sets of ids, then > reassemble. ?My thought was to use a loop, but I have been unable to finish > this loop in a logical way. ?I would like to find the unique ids for group > 1, group 2, etc., and rbind these back together. ?However, I am unclear how > to do this because in other attempts my final product is always a part of > the last run loop. ?My way of working around this has been to write.csv and > then re read at the end, which is so clumsy. ?Previously, I have used a > store matrix for individual cells. 1. Is there a better way to approach > this? ?2. How can I combine parts of matrices to other parts created in > prior loops? > I have created a primitive example below. ?Each of the groups varies in > number, so my repetitive example below is not accurate. ?In my real data, > ids repeat often within groups. > Thank you so much, Matt > > example <- data.frame(id=rep( > ( abs(round(rnorm(50,mean=500,sd=250),digits=0))) > ,3), group=rep(1:15,10)) > example <-example[with(example,order(id,group)),] > > for (i in 1:15) { > ai <- example[example[,2]==i,][!duplicated ( example[example[,2]==i,][,1] > ),] > write.csv(ai, paste('a',i,'.csv',sep="")) > } > b1<-read.csv('a1.csv') > b2<-read.csv('a2.csv') > b3<-read.csv('a3.csv') > b4<-read.csv('a4.csv') > b5<-read.csv('a5.csv') > b6<-read.csv('a6.csv') > b7<-read.csv('a7.csv') > b8<-read.csv('a8.csv') > b9<-read.csv('a9.csv') > b10<-read.csv('a10.csv') > b11<-read.csv('a11.csv') > b12<-read.csv('a12.csv') > b13<-read.csv('a13.csv') > > unis2 <- > rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind > (b1,b2),b3),b4),b5),b6),b7),b8),b9),b10),b11),b12),b13) > > ? ? ? ?[[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > 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.-- Jim Holtman Data Munger Guru What is the problem that you are trying to solve? Tell me what you want to do, not how you want to do it.
If you want a row for each unique item:> example <- data.frame(id=rep(+ ( abs(round(rnorm(50,mean=500,sd=250),digits=0))) + ,3), group=rep(1:15,10))> example <-example[with(example,order(id,group)),] > > uniqueIDs <- do.call(rbind+ , lapply(split(example, example$group), function(.grp){ + data.frame(group = .grp$group[1L] + , unique = unique(.grp$id) + ) + }) + )> > head(uniqueIDs, 25)group unique 1.1 1 15 1.2 1 282 1.3 1 361 1.4 1 497 1.5 1 553 1.6 1 583 1.7 1 618 1.8 1 729 1.9 1 817 1.10 1 842 2.1 2 135 2.2 2 253 2.3 2 290 2.4 2 351 2.5 2 418 2.6 2 457 2.7 2 518 2.8 2 594 2.9 2 686 2.10 2 804 3.1 3 208 3.2 3 327 3.3 3 442 3.4 3 516 3.5 3 536 On Thu, Feb 16, 2012 at 11:15 AM, Matt Spitzer <matthewjspitzer at gmail.com> wrote:> Dear R experts, > I am having difficulty using loops productively and would like to please > ask for advice. ?I have a dataframe of ids and groups. ?I would like to > break down the dataframe into groups, find the unique sets of ids, then > reassemble. ?My thought was to use a loop, but I have been unable to finish > this loop in a logical way. ?I would like to find the unique ids for group > 1, group 2, etc., and rbind these back together. ?However, I am unclear how > to do this because in other attempts my final product is always a part of > the last run loop. ?My way of working around this has been to write.csv and > then re read at the end, which is so clumsy. ?Previously, I have used a > store matrix for individual cells. 1. Is there a better way to approach > this? ?2. How can I combine parts of matrices to other parts created in > prior loops? > I have created a primitive example below. ?Each of the groups varies in > number, so my repetitive example below is not accurate. ?In my real data, > ids repeat often within groups. > Thank you so much, Matt > > example <- data.frame(id=rep( > ( abs(round(rnorm(50,mean=500,sd=250),digits=0))) > ,3), group=rep(1:15,10)) > example <-example[with(example,order(id,group)),] > > for (i in 1:15) { > ai <- example[example[,2]==i,][!duplicated ( example[example[,2]==i,][,1] > ),] > write.csv(ai, paste('a',i,'.csv',sep="")) > } > b1<-read.csv('a1.csv') > b2<-read.csv('a2.csv') > b3<-read.csv('a3.csv') > b4<-read.csv('a4.csv') > b5<-read.csv('a5.csv') > b6<-read.csv('a6.csv') > b7<-read.csv('a7.csv') > b8<-read.csv('a8.csv') > b9<-read.csv('a9.csv') > b10<-read.csv('a10.csv') > b11<-read.csv('a11.csv') > b12<-read.csv('a12.csv') > b13<-read.csv('a13.csv') > > unis2 <- > rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind(rbind > (b1,b2),b3),b4),b5),b6),b7),b8),b9),b10),b11),b12),b13) > > ? ? ? ?[[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > 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.-- Jim Holtman Data Munger Guru What is the problem that you are trying to solve? Tell me what you want to do, not how you want to do it.