Currently I am using Google Discovery engine regular search but I want to use this multi-turn search which saves the context of my previous queries but I am unable to find any documentation regarding how to implement it in python, link: https://cloud.google.com/generative-ai-app-builder/docs/multi-turn-search
My current Implementation:
def genappbuilder_search(
search_query: str = "",
project_id: str = project_id,
location: str = location,
search_engine_id: str = search_engine_id,
serving_config_id: str = serving_config_id,
) -> List[discoveryengine.SearchResponse.SearchResult]:
client = discoveryengine.SearchServiceClient()
serving_config = client.serving_config_path(
project=project_id,
location=location,
data_store=search_engine_id,
serving_config=serving_config_id,
)
request = discoveryengine.SearchRequest(
serving_config=serving_config, query=search_query, page_size=5
)
response = client.search(request)
return response.results
You can use the conversation notebook in this repository to use Multi-Turn (Conversational) Search in Python.
https://github.com/rasalt/snippets_es/
Official code samples should be added to the documentation soon.