Building a search completion system
Search completion, or auto-complete, is a common NLP application used in search engines and messaging apps. The goal is to suggest possible completions based on a user's partial input. Your task is to use Hugging Face's "text-generation" pipeline to implement a basic auto-complete system that generates relevant completions from the user's query.
Deze oefening maakt deel uit van de cursus
Natural Language Processing (NLP) in Python
Oefeninstructies
- Create an
autocompletepipeline with the"distilgpt2"model. - Generate five search query
suggestionsfor the givenprompt, limiting each to a maximum of eight tokens.
Praktische interactieve oefening
Probeer deze oefening eens door deze voorbeeldcode in te vullen.
from transformers import pipeline
# Create the pipeline
autocomplete = pipeline(task="____", model="____")
prompt = "Best books to read for"
# Generate search query completions
suggestions = ____
for suggestion in suggestions:
print(suggestion['generated_text'])