CODEWITHSUNDEEP
rdplyrcount
hamjam223
hamjam223

Reputation: 15

Using R to count consecutive occurrences of a number and how long each occurrence lasts for?

I've got a large dataset (dt) which has a time columnn (where time is in seconds) and a column that records a 1 when some other variables meet a certain value and a 0 when they don't, e.g:

time (s) var
0        1   
0.3      1
0.6      0
0.9      0
1.2      1
1.5      1
1.8      0

Part 1) what I want to do is count each time 1 is repeated as a unique occurrence (more than twice) in a count column which would look like this:

time (s) var count
0        1   1
0.3      1   1
0.6      0   0
0.9      0   0
1.2      1   2
1.5      1   2
1.8      0   0

where each occurrence in the same bout would have the same number and where 0 occurs is there is no counting.

For Part 1 I have this so far, but I would like it to print each unique occurrence as a count in a column which it doesn't do:

with(rle(dt$var), sum(lengths[values] > 2))

Part 2) I also want to know the length of time each occurrence lasts. (I also have a replicate column which has value of 1 for each row)

I tried this to calculate Part 2 but it doesn't work... 

var_time <- dt %>%
  group_by(replicate) %>%
  mutate(var_time = cumsum(var != lag(var, default = ""))) %>%
  group_by(var, time) %>%
  summarise(start = min(time),
            end   = max(time),
            var = sum(var))

Upvotes: 1

Views: 1696

Answers (2)

akrun
akrun

Reputation: 887118

An option with data.table

library(data.table)
setDT(df1)[, count := rleid(var) * var][count != 0, 
     count := match(count, unique(count))][]
#     time var count
#1:  0.0   1     1
#2:  0.3   1     1
#3:  0.6   0     0
#4:  0.9   0     0
#5:  1.2   1     2
#6:  1.5   1     2
#7:  1.8   0     0

Or with base R using rle/inverse.rle

df1$count <- inverse.rle(within.list(rle(df1$var), 
      values[as.logical(values)] <- seq_along(values[as.logical(values)])))

data

df1 <- data.frame(time = c(0, 0.3, 0.6, 0.9, 1.2, 1.5, 1.8), var = c(1, 1, 0, 0, 1, 1, 0))

Upvotes: 1

Ronak Shah
Ronak Shah

Reputation: 388982

You can use rle to get answer to first part. 

dt$count <- with(rle(dt$var), rep(values * cumsum(values & lengths >= 2),lengths))

dt
#  time var count
#1  0.0   1     1
#2  0.3   1     1
#3  0.6   0     0
#4  0.9   0     0
#5  1.2   1     2
#6  1.5   1     2
#7  1.8   0     0

Upvotes: 2

Related Questions