ChatBees
  • 👋Welcome
  • Concepts
    • 📖Architecture
    • 📖Security
    • 📖Serverless RAG
    • 📖Namespace and Collection
    • 📖Access Control
    • 🚀Ticket AI Agent
  • Ticket AI Agent
    • Installation
  • WEB APP REFERENCES
    • 🔑Sign-in and Sign-out
    • ⛓️Manage Connectors
    • 🌏Manage Collections
      • 💿Data Sources
        • 📖Configure Periodic Import
      • ❓Chat with collection
      • ⏱️View Q/A history
      • 📖Publish a Collection
    • 🏛️Manage Users
    • 💰[Flex] Billing and Payment
    • 💰[Enterprise] Billing and Payment
    • 📈Account Usage
    • 🗝️API Keys
    • 🖥️Generated Code Sample
  • ChatBots
    • 🪄AI Chat for Confluence
    • 🪄ChatBees Slack Bot
    • 🪄ChatBees Website ChatBot
    • 🪄Pnyx Discord Bot
  • Snowflake Native App
  • API References
    • 📖API Key
    • 📖Collection Operations
      • 📖Create Collection
      • 📖Configure Collection
      • 📖List Collections
      • 📖Delete Collection
    • 📖Document Operations
      • 📖Upload Document
      • 📖Summarize Document
      • 📖Get Document Outlines and FAQs
      • 📖Ask
      • 📖Chat
      • 📖Search
      • 📖Personalize Response
      • 📖List Documents
      • 📖Delete Document
    • 📖Crawl Operations
      • 📖Create Crawl
      • 📖Get Crawl
      • 📖Index Crawl
      • 📖Delete Crawl
    • 📖Ingest Data Sources
      • 📖Create Ingestion
      • 📖Get Ingestion
      • 📖Index Ingestion
      • 📖Delete Ingestion
Powered by GitBook
On this page
  1. API References
  2. Document Operations

Ask

You can ask questions within a collection. ask() method returns a plain-text answer to your question, as well as a list of most relevance references used to derive the answer.

By default, ask will get answers from all relevant documents in the collection. If you want to ask a single document, you could limit the document when ask.

POST /docs/ask HTTP/1.1
Api-Key: my_api_key
Content-Type: application/json
Host: my_account_id.us-west-2.aws.chatbees.ai

{
  "namespace_name": "string",
  "collection_name": "string",
  "question": "string",
  // Optional top_k, default is 5.
  "top_k": int,
  // Optional doc_name. if specified, question will be answered with the
  // context only from the doc.
  "doc_name": "string" or null,
  // The chat history
  history_messages: [(question1, answer1), ...]
}

Response:
{
  "answer": "string",
  "refs": [
    {"doc_name": "string", "page_num": int, "sample_text": "string"},
    ...
  ]
}
import chatbees as cb

# Configure API key
cb.init(api_key="my_api_key", account_id="my_account_id")

# get answer from all docs in the collection
answer, refs = cb.collection('llm_research').ask('what is a transformer?')

# get answer from a specific doc in the collection
answer, refs = cb.collection('llm_research').ask(question='what is a transformer?', doc_name='file.pdf')
PreviousGet Document Outlines and FAQsNextChat

Last updated 10 months ago

📖
📖