Introduction

We want to benchmark algorithms for private data synthesis, particularly for spatio-temporal data synthesis. This can have profund impact on multiple areas, especially in public health

Usage

Running using Docker

Build the image

docker build -t evaluation_container .

Run the container

docker run -v dataset:/app/dataset/ -t -i dataset_eval

The dataset folder should contain your target dataset with the name "dataset.csv".

The script trains and evaluates TabNet, Random Forest and MLP on all possible combinations of categorical data.

Contributors to the code/project (random order)

1. Mikhail Dmitrienko
2. Sheshank Shankar
3. Soham Dasgupta
4. Ishaan Singh
5. Rohan Sukumaran
6. Abhishek Singh
7. Praneeth Vepakomma
8. Hrishikesh Kamath