Determination of model importance as an ensemble member

This is the data repository for the following paper:

Title: Beyond forecast leaderboards: Measuring individual model importance based on contribution to ensemble accuracy

Authors: Minsu Kim, Evan L. Ray, Nicholas G. Reich

Computational environment

We use conda for managing a Python environment. If you’re just getting started, you can use the following commands to create your environment.

Enter this command in your terminal to create a new conda environment for the project:

conda create -n importance python=3.11

Enter these commands in an R session:

library(reticulate)
library(qens)
reticulate::use_condaenv("importance")
qens::install_py_dependencies()

Whenever you’re using this package, you’ll need to tell reticulate to use the conda environment you created:

library(reticulate)
reticulate::use_condaenv("importance")