Ariel Data Challenge 2023

This is the repository for the 3rd place solution in the 2023 Ariel Data Challenge.

Setup

download the data from the ADC 2023 challenge website, and place it in the data directory
- check the training_path variables in the notebooks and adjust if needed
python=3.11.3 was used
install the packages from the requirements.txt file
for spectral scores and taurex3, please see the notes below (used in final_inference.ipynb)

code in the base_code directory is from the Baseline github repository provided by UCL
code in tabnet directory is from a TabNet for Tensorflow 2.0 repository
for details on calculating the spectral scores, and setting up TauREx3, please see the Metrics section of the baseline repository

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
base_code		base_code
data		data
histories		histories
saved_objects		saved_objects
tabnet		tabnet
weights		weights
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
final_inference.ipynb		final_inference.ipynb
kfold_gauss_multivariate_training.ipynb		kfold_gauss_multivariate_training.ipynb
kfold_uniform_quantile_training.ipynb		kfold_uniform_quantile_training.ipynb
model.py		model.py
requirements.txt		requirements.txt
utils.py		utils.py