partition_survival_data.R

# Idea thanks to https://scottclowe.com/2016-03-19-stratified-regression-partitions/
# Adapted for right censored survival outcomes by applying the above procedure
# separately for censored and uncensored observations.

# Copyright 2019 Steve Yadlowsky

# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:

# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.

# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.

library(dplyr)

partition_sorted_data_frame = function(frame, k) {
  n = nrow(frame)
  left_overs = n %% k
  fold_order = do.call(c, lapply(1:n %/% k,
                                 function(x) {sample.int(k, k)}))
  return(c(fold_order, sample.int(k, left_overs)))
}

partition_survival_data = function(surv_obj, k) {
  survival_frame = data.frame(t_ = surv_obj[,1],
                              e_ = surv_obj[,2],
                              index=1:nrow(surv_obj))
  censored_obs = survival_frame %>% dplyr::filter(e_ == 0)
  if (nrow(censored_obs) > 0) {
    censored_obs = censored_obs %>% dplyr::arrange(t_)
    censored_fold_ids = partition_sorted_data_frame(censored_obs,k)
    censored_indices = censored_obs %>% pull(index)
  } else {
    censored_fold_ids = NULL
    censored_indices = NULL
  }
  event_obs = survival_frame %>% dplyr::filter(e_ == 1)
  if (nrow(event_obs) > 0) {
    event_obs = event_obs %>% dplyr::arrange(t_)
    event_fold_ids = partition_sorted_data_frame(event_obs, k)
    event_indices = event_obs %>% pull(index)
  } else {
    event_fold_ids = NULL
    event_indices = NULL
  }
  fold_order = c(censored_fold_ids,
                 event_fold_ids)
  indices = c(censored_indices,
              event_indices)
  return(fold_order[order(indices)])
}