9. Step 6: Using the Model for Prediction

Section 9 of 10-~ 12 min read-Synced from Cuantum content

Finally, we’ll use the trained model to predict sentiment for new reviews.

Code Example: Predicting Sentiment

# New reviews for predictionreviews = [    "The movie was absolutely fantastic! A must-watch.",    "I regret watching this film. It was a waste of time.",    "The movie was just okay, nothing special."] # Tokenize new reviewsinputs = tokenizer(reviews, truncation=True, padding=True, return_tensors="pt") # Get predictionsoutputs = model(**inputs)predictions = outputs.logits.argmax(dim=-1) # Map predictions to labelslabels = ["Negative", "Positive"]for review, prediction in zip(reviews, predictions):    print(f"Review: {review}")    print(f"Predicted Sentiment: {labels[prediction]}")