- Python
- Linux (should work on macOS too, but not tested, can't guarantee about Windows, it might not work)
- Requirements:
pip install -r requirements.txt
Note: if you want to get LDAvis visualizations, you need to install pyLDAvis
from source, because its dependencies are outdated.
First run files to make LDA models:
python3 make_fake_models.py
python3 make_real_models.py
Then you can get the visualizations of metrics:
python3 calculate_metrics_fake.py
python3 calculate_metrics_real.py
To get wordclouds run
python3 make_wcs.py
To get topic visualizations first install pyLDAvis
, then
pip3 install jupyter
python -m notebook
Then open make_ldavis.ipynb
on jupyter notebook, and run cells one by one.