1

here is my questions: I got data with 3000 obs. and 5000 features, the 3000 obs. has a numeric names like 100.1,100.3,100.5,100.7. I changed the names into a integer variables by segs <-as.integer(names), then I want to use segs as a factor to sum all of the 3000 features. The length of the segs is 300 so the final data frame is 300 by 5000. I know tapply could be used to get the sum by factor for one variable but I have to use for to get all of the 5000 features summed. It is really time-consuming, so I want to know if there is a clear way in R to solve those problems or if there are some packages to solve this kind of problem.

This is the dirty code and df0 is the data while df is what I want:

df <- data.frame()
 for(i in 2:ncol(df0)-1){
    temp <- tapply(df0[,i],df2$segs,sum)
    df <- cbind(df,temp)
}

Thanks!

=====

Thanks, Roland, a demo data is shown as follows:

set.seed(42)    
df0 <- data.frame(
X = rnorm(100,10,10),
Y = rnorm(100), 
Z = rnorm(100))
df0$seq <- as.integer(df0$X)
yufree
  • 433
  • 3
  • 9
  • This would be very easy and extremely efficient with package data.table. Provide [a reproducible example](http://stackoverflow.com/a/5963610/1412059) and I might show you how. Btw. you are making the [cardinal sin](http://www.burns-stat.com/pages/Tutor/R_inferno.pdf) of growing an object in a loop. – Roland Jun 13 '14 at 07:22
  • @Roland Thanks a lot for your suggestion, I simulate simple data with a seed in the question. `df0` is 100 by 3, `df` will be 35 by 3. – yufree Jun 13 '14 at 07:44

1 Answers1

2

Try this...

set.seed(42)    
df0 <- data.frame(
    X = rnorm(100,10,10),
    Y = rnorm(100), 
    Z = rnorm(100))
df0$seq <- as.integer(df0$X)

library(data.table)
dt = data.table(df0)
dt[,lapply(.SD, sum), by=seq ]

    seq           X            Y           Z
 1:  23 164.8144774  1.293768670 -3.74807730
 2:   4   8.9247301  1.909529066 -0.06277254
 3:  13  40.2090180 -2.036599633  0.88836392
 4:  16 147.8571697 -2.571487358 -1.35542918
 5:  14  72.1640142  0.432493959 -1.49983832
 6:   8  42.8498355 -0.582031919 -1.35989852
 7:  25  75.9995653  0.896369560 -1.08024329
 8:   9  27.5244048  0.833429855 -1.19363017
 9:  30  30.1842371  0.188193035 -0.64574372
10:  32  32.8664539  0.108072728  2.03697217
11:  -3  -7.5714175 -0.899304085 -1.27286230
12:   7  29.6254908 -0.929790177  2.75906514

27:  12  50.2535374 -0.620793351 -3.80900436
28:  24  24.4410126 -0.433169033 -0.02671746
29: -19 -19.9309008 -0.533492330 -1.01759612
30:  11  11.8523056 -1.071782384  0.96954501
31:  19  38.5407490 -0.751408534 -4.81312992
32:   0  -0.9642319  1.453325156  2.20977601
33:  -1  -4.3685646 -0.834654913 -0.24624546
34:  18  18.2177311 -1.594588162  0.27369527
35:  -4  -4.5921400  0.586487537  0.86256338
vrajs5
  • 4,066
  • 1
  • 27
  • 44