A large-scale question answering dataset on real-world Schema.org data
The following OS packages are required to build and run Genie Toolkit:
The following commands can be used to install all the dependencies on common Linux distributions:
dnf -y install nodejs make gcc-c++ gettext zip # Fedora/RHEL
apt -y install nodejs build-essential make g++ gettext zip # Ubuntu/Debian
For GenieNLP, we recommend using Python 3.8, as it is the version we use for testing its builds.
Genie toolkit is the core of Schema2QA - it generates large, natural synthetic datasets given a database schema and example values. To replicate the results in the leader board, you need to use v0.8.0 of Genie. Install Genie with the following commands:
git clone https://github.com/stanford-oval/genie-toolkit
cd genie-toolkit
git checkout tags/v0.8.0
npm install
GenieNLP provides the infrastructure for training as well as AutoQA neural paraphraser. Install GenieNLP v0.6.0 from source as follows:
git clone git@github.com:stanford-oval/genienlp.git
cd genienlp
git checkout tags/v0.6.0
pip3 install -e .
Alternatively, if you do not plan to change the training or paraphrasing code, you can simply install GenieNLP using pip by running pip install genienlp==0.6.0.