1.Dataset preparation

For SUPPORT and SYNTHETIC dataset, please refer to auton survival's setting

The links are provided: https://github.com/autonlab/auton-survival

For MIMIC-III and MIMIC-IV, please refer to the github repo for preprocessing steps:

https://github.com/YerevaNN/mimic3-benchmarks

we extract time-series data for 17 clinical variables from the first 48 hours of each ICU admission

2.Environment setting

conda env create -f environment.yml

3.Run model

use the DDPSurv Branch

python baseline.py

add arguments by your own requirements