This repository serves two purposes. First, it is the working repository for the weighted (causal) partial dependence plot (WPDP) first introduced by Lazzari, Alvarez, and Ruggieri (2022) [1]. The goal is to formalize further the WPDP under the risk of sample selection bias. The current repository is, thus, work in progress.
Second, this repository presents the general pipeline of said paper. It does so to meet the goals of the NoBIAS ITN testbed. The original code used for the paper is kept in a separate, private repository due to the paper's use of proprietary data. The code presented here (denoted with the prefix lar for Lazzari, Alvarez, and Ruggieri) provides an almost exact implementation of the paper's general pipeline.
[1] M. Lazzari, J.M. Alvarez, and S. Ruggieri. Predicting and Explaining Employee Turnover Intention. International Journal of Data Science and Analytics, May 2022, DOI: 10.1007/s41060-022-00329-w.