- Instructions are in the notebook
legal_recursive_retrieval.ipynb - Create a venv if desired
python3 -m venv venvsource venv/bin/activate
- Run
pip install -r requirements.txtto install the necessary packages.
- If you are providing a workspace, it should be empty and should only contain documents used build the lexical graph.
- As of the notebook's creation date, any developers interested in the Reducto API must reach out to founders to get a key for the demo.
- However, the free demo on the website allows up to 10 pages of text to be analyzed
- The API response JSONs can be copied and saved as
<pdf filename without extension>.jsonunderdocs_structure/reducto/.
- In our example,
bnm_compliance.pdfin/docswas saved asbnm_compliance.jsonin/docs_structure/reducto/. - This already has been done for you so you can run the notebook as-is – follow the steps above if you want to try with your own documents.