Describe how to represent a UD token as a tabular data structure to allow outputing matched words in a TSV

Improve scripts for profile computation: searching from (serialized) indexed trees, and using ranges to compact occurrences found

Add types to represent indexed sentences and documents, effectively absorbing part of the work that was done during the search

Add a script to serialize syntax trees computed from the .conllu files, copying part of the Conllu.Type types in the process to get a Generic instance for Serialization and to slightly improve structure by indexing features by name with a Map

Lift some constraint on the monad in  and use that to pass the config around as a Reader in the new profile extractor script

A tiny script used to add the manually annotated label to a bunch of texts being prepared in a JSONL (to be used as input for prodigy)

Add predictor for MultiBERT model, improve the trainer for the same model and expose the predicted score in the model's output

Separate train script into one to train directly one classifier and one to train one binary classifier (accept/reject) for each discursive function

Add support for labeled data in input tsv and make the paralallel between conversions to and from prodigy's jsonl more obvious

Temporary hack to be able to extract paragraph subcorpora as well article ones (TODO: expose something similar to readNamedTsv returning Either (Vector ParagraphRecord) (Vector ArticleRecord))