R: group_by() column1 and perform rle() on column2

[davidss101]

Hello Everyone,

I was wondering if people could help with this problem. I am trying to use group_by() and rle() in R to group according to column1 and count the consecutive number of times there is either a one or a zero within each group from column2. I would like to return both the "values" and "lengths" data from using rle().

This is what my data frame (df1) looks like:

I used this code to generate what I thought might work based on some suggestions I found on the web.

df2 <- df1 %>% group_by(column1) %>% reframe(data.frame(unclass(rle(df1$column2))))

This is the output I get for df2:

It seems that my code is not grouping the data effectively according to column1. Any suggestions would be appreciated. Thanks.

[davidss101]

I'm sorry I forgot to mention that this is what I am looking for:

[davidss101]

Hi Everyone,

Since I was ultimately interested in only the values in column1 which were "1", I used this code that I adapted from another website. Unfortunately, I forget now what that site was.

df2 <- df1 %>% group_by(column1) %>% mutate(sequence = data.table::rleid(column2 == 1),) %>% filter(column2 == 1) %>% group_by(column1, sequence) %>% summarise(length = n())

Thanks.

[williamlai2]

Hi, the issues here a specific to the package code.

You might get more help on your problem at: https://community.rstudio.com/

[psychelzh]

Or a better place to go is https://stackoverflow.com/.