Format summary text with Spacy

#spacy #llm

Here is the solution proposed by Groq :

import spacy
from spacy import displacy
from spacy.util import minibatch, compounding

# Load the French language model
nlp = spacy.load("fr_core_news_sm")

# Process the text
doc = nlp("Leurs contrôles, dans le cadre du dispositif anti-inflation, ont permis de veiller à ce que les engagements des industriels et distributeurs soient tenus. ...")

# Extract the sentences
sentences = [sent for sent in doc.sents]

# Correct the orthography
orthography_corrected_sentences = []
for sentence in sentences:
    orthography_corrected_sentence = sentence.text
    orthography_corrected_sentence = orthography_corrected_sentence.replace("«", "\"").replace("»", "\"").replace("â", "a").replace("ê", "e").replace("ô", "o")
    orthography_corrected_sentences.append(orthography_corrected_sentence)

# Format the sentences as a bullet list
bullet_list = []
for sentence in orthography_corrected_sentences:
    bullet_list.append("- " + sentence.strip())

print("\n".join(bullet_list))