zero_shot_classification

Zero-shot Text Classification

This example shows how to use a pre-trained BART model for zero-shot text classification.

python zero_shot_classification/zero_shot_classification.py

As candidate labels we use ["sport", "music", "politics", "business", "technology", "entertainment", "science", "health", "war"]. You can then enter a text and a list of labels and the model will predict the label that best fits the text, e.g.:

• I like to play football and tennis.
• The battery of my phone is dead.
• New research shows that the playing music can help you sleep better.
• Military forces are preparing to attack the enemy.

You can also use it for sentiment analysis (candidate_labes=["positive", "negative", "neutral"]):

• I absolutely loved the friendly service at the cafe!
• The movie was a complete waste of time and money.