You can setup this project on Linux, Windows, or MacOS. Additionally, you can also use Google Colab.
Here is an example about using the project in Google Colab: Feature Augmentation with Webtables
This is how to setup the project on your machine:
conda create -n Bench python=3.8
conda activate Bench
git clone https://github.com/LUH-DBS/DataDiscoveryBenchmark.git
cd DataDiscoveryBenchmark
git pull origin main
python -m pip install .
If python -m pip install . does not work, try this command:
python -m pip install --use-feature=in-tree-build .
You can run the feature augmentation task with:
python datadiscoverybench/feature_augmentation/augmentation.py
On Mac OS, you need to run the following command additionally:
brew install libomp