For the overview of BOHR, please see https://github.com/giganticode/bohr
- This repository contains CLI and engine for running the BOHR pipeline:
- reading the BOHR config
- fetching the heuristics from remote BOHR repository needed for a specific task
- fetching the datasets
- applying heuristics to artifacts and combining their outputs
- preparing new dataset with the trained model;
Moreover, BOHR-runtime provides utilities for debugging heuristics and evaluating their effectiveness.
- Clone the repository.
- Inside the repository, run
poetry install. This will create a virtual environment and install the dependencies. - To run python interpreter within the virtual environment, use
poetry run ... - For example, to run the tests, execute:
poetry run pytest --doctest-modules --ignore=test-b2b
CI runs unit-tests and integration tests (two tiny scenarios). Data for the integration tests is downloaded from the unibz squirrel server using DVC via http: http://squirrel.inf.unibz.it:8180. Storage location on the server is /data/bohr_dvc_storage. CI has read-only access to the data.