Zero-Shot classification with Transformers

Classify text into arbitrary labels with a zero-shot classification pipeline.

1 min read Updated

Zero-shot classification lets you label text with categories the model was never explicitly trained on. This script loads a multilingual NLI model and classifies text against arbitrary labels. The script is below.

zero-shot-classification.py

python
from transformers import pipeline

classifier = pipeline("zero-shot-classification", 
                      model="BaptisteDoyen/camembert-base-xnli")

sequence = "L'équipe de France joue aujourd'hui au Parc des Princes"
candidate_labels = ["sport","politique","science"]
hypothesis_template = "Ce texte parle de {}."    

result = classifier(sequence, candidate_labels, hypothesis_template=hypothesis_template)     

print(result)

Search articles

Type to filter articles. Use the arrow keys to move through results and Enter to open one. Press Escape to close.