Repository files navigation

Configurations for maintaining the Annif projects with YKL vocabulary used atFinto AI service.

The projects are trained and evaluated using aDVC (Data Version Control) pipeline defined indvc.yaml.

The pipeline takes care of

1. installing Annif in a venv,
2. loading the vocabulary,
3. training the projects,
4. evaluating the projects.

When the necessary vocabulary and training corpora are in place the pipeline can be run using the command

dvc repro

For more information about using DVC with Annif projects see theDVC exercise of Annif tutorial.