LouisZango / cfrnet-reproduction

Reproducing Shalit et al.'s Individual Treatment Effect model. This is a deep neural net that can be applied to various problems in causal inference.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

CFRnet Reproduction

For the completion of Deep Learning CS4240 - University of Technology Delft.

Original paper at proceedings.mlr.press

A blog describing the goal of this repository can be found here.

This repository was originally hosted on GitLab. You can find the entire history of this project on our GitLab repository.

The used data are the IHDP-1000 and Jobs datasets. These can be retrieved via https://www.fredjo.com/.

Contributors

This project has been created by

Usage

Install all requirements with

pip install requirements.txt

Training the CFRnet can then be done by running

python cfrnet/main.py

An output folder with the results will be created. This folder and all other configuration parameters can be specified in cfrnet/config.py.

About

Reproducing Shalit et al.'s Individual Treatment Effect model. This is a deep neural net that can be applied to various problems in causal inference.

License:MIT License


Languages

Language:Python 100.0%