jurispacy-tokenizer 1.2.0

Flair tokenizer adapted to French court decisions using spacy tokenization

JuriSpacyTokenizer

Description

Tokenizer(s) used in our NLP projects. Built using Flair and spaCy

Installation

pip install jurispacy-tokenizer
python -m spacy download fr_core_news_sm-3.6.0

Usage

Tokenize strings

You can use this library to tokenize a string into a list of strings representing tokens:

from jurispacy_tokenizer import JuriSpacyTokenizer

tokenizer = JuriSpacyTokenizer()
text = "M.Paul et Jean-Pierre sont heureux."

tokens = tokenizer.tokenize(text)

for token in tokens:
    print(token)

This should ouptut:

M.
Paul
et
Jean-Pierre
sont
heureux
.

Tokenize longer text into sentences

You can also parse longer text to create Flair Sentence objects:

from jurispacy_tokenizer import JuriSpacyTokenizer

tokenizer = JuriSpacyTokenizer()

text = """Bonjour tout le monde! Je m'appelle Amaury.

Je travaille avec Paul."""

sentences = tokenizer.get_tokenized_sentences(text)

for s in sentences:
    print(s)

This should output:

Sentence[5]: "Bonjour tout le monde!"
Sentence[5]: "Je m'appelle Amaury."
Sentence[5]: "Je travaille avec Paul."