Classificazione zero-shot e QNLI

Natural Language Processing (NLP) in Python

Fouad Trad

Machine Learning Engineer

Classificazione zero-shot

  • Permette al modello di assegnare testi a etichette mai viste
  • Usa il linguaggio naturale per prevedere l’output
  • Utile per:
    • Tag dei contenuti
    • Supporto clienti
    • Filtrare articoli di news

Immagine con il testo "The national football team won the cup yesterday" da classificare in tre categorie: sports, technology e health.

Natural Language Processing (NLP) in Python

Pipeline di classificazione zero-shot

from transformers import pipeline

zero_shot_classifier = pipeline(
task="zero-shot-classification",
model="MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli" )
text = "The national football team won the cup yesterday."
candidate_labels = ["sports", "technology", "health"]
result = zero_shot_classifier(text, candidate_labels)
print(result)
{'sequence': 'The national football team won the cup yesterday.',
 'labels': ['sports', 'technology', 'health'], 
 'scores': [0.9948731064796448, 0.0029330444522202015, 0.002193822991102934]}
Natural Language Processing (NLP) in Python

Question Natural Language Inference (QNLI)

  • Verifica se la risposta a una domanda è trovabile in un passaggio
  • Utile per:
    • Ricerca documenti
    • Chatbot
    • Information retrieval

Immagine che mostra che QNLI riceve un passaggio e una domanda e restituisce un punteggio.

Natural Language Processing (NLP) in Python

Pipeline QNLI

from transformers import pipeline

qnli_pipeline = pipeline( task="text-classification", model="cross-encoder/qnli-electra-base" )
passage = "Penguins are found primarily in the Southern Hemisphere."
question = "Where do penguins live?"
result = qnli_pipeline({"text": question, "text_pair": passage})
print(result)
{'label': 'LABEL_0', 'score': 0.9951545000076294}
Natural Language Processing (NLP) in Python

Pipeline QNLI

from transformers import pipeline
qnli_pipeline = pipeline(
    task="text-classification", 
    model="cross-encoder/qnli-electra-base"
    )
passage = "Penguins are found primarily in the Southern Hemisphere."
question = "What is the capital of Paris?"
result = qnli_pipeline({"text": question, "text_pair": passage})
print(result)
{'label': 'LABEL_0', 'score': 0.008907231502234936}
Natural Language Processing (NLP) in Python

Passons à la pratique !

Natural Language Processing (NLP) in Python

Preparing Video For Download...