Python scripts for data analysis, mostly work in progress.
Python 3
Install dependencies:
pip3 install -r requirements.txt
Change into the nwbib directory:
cd nwbib
Load sample NWBib data from the Lobid API:
python3 nwbib_subjects_load.py
Run classification experiment:
python3 nwbib_subjects_process.py
Run bulk classification (first run takes some time):
python3 nwbib_subjects_bulk.py
Run a pipeline with cross-validation and hyperparameter optimization:
python3 nwbib_subjects_pipeline.py
Run experiments based on paragraph vectors:
python3 nwbib_doc2vec.py