4 projects
bm25-vectorizer
BM25 Vectorizer (Scikit-learn Compatible)
json-repair-llm
JSON repair using multiple backends: LLMs and FSM-based processing with Pydantic models
weak-annotators
Weak annotators for information extraction (NER)
spacy-trankit
spacy wrapper for Trankit, a Transformer-based multilingual neural dependency parser with tokenization and NER