query_data.py

from langchain.prompts.prompt import PromptTemplate
from langchain.llms import LlamaCpp
from langchain.chains import ConversationalRetrievalChain
from langchain.memory import ConversationBufferMemory
from huggingface_hub import hf_hub_download

import psutil
import os

#_template = """Given the following conversation and a follow up question, rephrase the follow up question to be a standalone question.
#You can assume the question about the uploaded document.

#Chat History:
#{chat_history}
#Follow Up Input: {question}
#Standalone question:"""
#CONDENSE_QUESTION_PROMPT = PromptTemplate.from_template(_template)

#template = """You are an AI assistant for answering questions about the uploaded document.
#You are given the following extracted parts of a long document and a question. Provide a conversational answer.
#If you don't know the answer, just say "Hmm, I'm not sure." Don't try to make up an answer.
#If the question is not about the uploaded document, politely inform them that you are tuned to only answer questions about the uploaded document.
#Question: {question}

#Answer in Markdown:"""
##QA_PROMPT = PromptTemplate(template=template, input_variables=["question", "context"])
#QA_PROMPT = PromptTemplate(template=template, input_variables=["question"])

#=========
#{context}
#=========


def get_chain(vectorstore):
    if not os.path.exists("ggml-vic7b-q5_1.bin"):
        hf_hub_download(repo_id="eachadea/ggml-vicuna-7b-1.1", filename="ggml-vic7b-q5_1.bin", local_dir=".")
    llm = LlamaCpp(model_path="ggml-vic7b-q5_1.bin", n_ctx=2048, n_threads=psutil.cpu_count(logical=False)/2)
    qa_chain = ConversationalRetrievalChain.from_llm(
        llm,
        vectorstore.as_retriever(),
#        condense_question_prompt=CONDENSE_QUESTION_PROMPT,
    )
    return qa_chain