Transition-Based Dependency Parser (Python)

This project implements a transition-based dependency parser that parses English sentences into CoNLL-style dependency graphs using a trained TransitionParser. It supports both file input and stdin input, and leverages spaCy to initialize sentences for parsing.

---

Requirements

• Python 3.12 (or compatible)

• pip packages:

  • numpy
  • scipy
  • scikit-learn >= 1.2
  • joblib
  • spacy
  • nltk

• SpaCy model: en_core_web_sm (automatically downloaded if missing)

---

Usage

From a text file

python parse_sentences.py <model_file> <input_file> > output.conll

python parse_sentences.py mymodel.joblib my_text.txt > parsed.conll