28 lines
1.5 KiB
R
28 lines
1.5 KiB
R
source("renv/activate.R")
|
|
library("tibble")
|
|
library("dplyr", warn.conflicts = F)
|
|
library("tidyr")
|
|
library("tibble")
|
|
options(scipen=999)
|
|
|
|
# Using mostly indeixng instead of tidyr because is faster
|
|
resampled_episodes <- read.csv(snakemake@input[[1]])
|
|
resampled_episodes["n_resamples"] <- 1 + (resampled_episodes["end_timestamp"] - resampled_episodes["start_timestamp"]) %/% 60001
|
|
resampled_episodes <- resampled_episodes %>% uncount(n_resamples, .id = "nrow")
|
|
|
|
resampled_episodes["nrow"] <- (resampled_episodes["nrow"] - 1) * 60000
|
|
resampled_episodes["start_timestamp"] <- resampled_episodes["start_timestamp"] + resampled_episodes["nrow"]
|
|
# Use +59999 because each resampled minute should not overlap with each other
|
|
resampled_episodes["end_timestamp"] <- pmin(resampled_episodes["start_timestamp"] + 59999, resampled_episodes["end_timestamp"])
|
|
resampled_episodes <- resampled_episodes %>% select(-nrow)
|
|
resampled_episodes <- resampled_episodes %>% uncount(2, .id = "end_flag")
|
|
|
|
resampled_episodes <- resampled_episodes %>% add_column(timestamp = NA_real_)
|
|
if(nrow(resampled_episodes) > 0){
|
|
resampled_episodes[resampled_episodes$end_flag ==1, "timestamp"] = resampled_episodes[resampled_episodes$end_flag ==1, "start_timestamp"]
|
|
resampled_episodes[resampled_episodes$end_flag ==2, "timestamp"] = resampled_episodes[resampled_episodes$end_flag ==2, "end_timestamp"]
|
|
}
|
|
resampled_episodes <- resampled_episodes %>% select(-end_flag)
|
|
|
|
write.csv(resampled_episodes, snakemake@output[[1]], row.names = FALSE)
|