RAG Processor¶

class rag.rag_processor.RAGProcessor(model: str, max_token: int, stream: bool, temperature: float, top_p: float, top_k: int, client)¶

Class implementing the RAG process

Parameters:

  • model (str) – LLM Model used for generating chatbot answers

  • max_token (int)

  • stream (bool)

  • temperature (float)

  • top_p (float)

  • top_k (int)

  • client

create_openai_message(context_docs, query)¶

Format the message to send to the OpenAI API.

Parameters:

  • context_docs (str) – Context matching the query according to the retrieval process

  • query (str) – User input question

Returns:

Contains the message in the correct format to send to the OpenAI API

Return type:

list of dict

create_openai_stream(messages)¶

Create a stream to communicate with OpenAI.

Parameters:

messages (dict)

Return type:

chat.completion

async embed(text_input: EmbeddingRequest)¶

Get the embedding of an embedding request.

Parameters:

text_input (EmbeddingRequest)

Returns:

The requested text embedding

Return type:

dict

generate(openai_stream, source_url)¶

Generate the answer using LLM.

Parameters:

  • openai_stream

  • source_url

async process(request: RAGRequest)¶

Execute the RAG process and query the LLM model.

Parameters:

request (RAGRequest) – User input question

Returns:

LLM generated answer to the question

Return type:

str

async retrieve(request: RAGRequest, language: str = None, k: int = 0)¶

Retrieve context documents related to the user input question.

Only supports retrieval of 1 document at the moment (set in /config/config.yaml).

Todo

multi-doc retrieval later

Parameters:

  • request (RAGRequest) – User input question

  • language (str) – Question and context documents language

  • k (int, default 0) – Number of context documents to return (need to be revised, current logic in the code is confusing)