We must register to the challenge and create a team with all of us (info here!)
To run the code, download train/
and test/
folders into project root (two options):
- From the synapse if you are logged in, here
- In artemisa are the
train/
andtest/
folders. (/home/artemisa/DREAM-FINRISK/
)
All code must be included in src/
folder. In this folder, we have an utils/
repository to stored those function and/or scripts for general stuffs (such as data preprocessing, etc.).
Each new model/pipeline must be have a specific folder in src/
(see baseline as example). The new models/pipelines have to import data from train/
and make predict using test/
data. The predicted results must be saved on output/
as .csv
(see here for more information)
After download train/
and test/
folders, you can import both subsets with the following code:
source("src/utils/importPseq.r")
train <- pseq(subset = "train")
test <- pseq(subset = "test")