ring.py

import os
from getpass import getpass
from langchain_ai21 import AI21LLM
from langchain_openai import OpenAI

from langchain.prompts import PromptTemplate
from langchain_core.output_parsers import StrOutputParser


LLM_GPT35 = "GPT3.5"
LLM_GPT4 = "GPT4"
LLM_GPT4o = "GPT4o"
LLM_LLAMA2 = "LLAMA2"
LLM_GEMINI = "Gemini"
LLM_AI21 = "AI21"

OPENAI_LLMS = [LLM_GPT35, LLM_GPT4, LLM_GPT4o]

PREF_ORDER_LLMS = (LLM_GEMINI, LLM_LLAMA2, LLM_GPT35, LLM_GPT4, LLM_GPT4o, AI21)


def requires_openai_key(llm_ver):

    return llm_ver in OPENAI_LLMS


def get_openai_api_key():

    # if openai_api_key_sb == None or len(openai_api_key_sb)==0:
    openai_api_key = os.environ.get("OPENAI_API_KEY")
    if openai_api_key == None:
        openai_api_key = str(open("../oaik", "r").readline())
    # else:
    #    openai_api_key = openai_api_key_sb

    return openai_api_key


def get_llamaapi_key():

    llamaapi_key = os.environ.get("LLAMAAPI_KEY")

    return llamaapi_key


def get_gemini_api_key():

    gemini_api_key = os.environ.get("GEMINIAPI_KEY")

    return gemini_api_key

def get_ai21_api_key():
    
    ai21_api_key = os.environ.get["AI21_API_KEY"]

    return ai21_api_key

def get_llm(llm_ver, temperature):

    if llm_ver == LLM_GPT35:
        llm = OpenAI(temperature=temperature, openai_api_key=get_openai_api_key())

    elif llm_ver == LLM_GPT4:
        from langchain_openai import ChatOpenAI

        llm = ChatOpenAI(
            model_name="gpt-4",
            openai_api_key=get_openai_api_key(),
            temperature=temperature,
        )
    elif llm_ver == LLM_GPT4o:
        from langchain_openai import ChatOpenAI

        llm = ChatOpenAI(
            model_name="gpt-4o",
            openai_api_key=get_openai_api_key(),
            temperature=temperature,
        )

    elif llm_ver == LLM_LLAMA2:
        from llamaapi import LlamaAPI
        import asyncio

        # Create a new event loop
        loop = asyncio.new_event_loop()

        # Set the event loop as the current event loop
        asyncio.set_event_loop(loop)

        llama = LlamaAPI(get_llamaapi_key())

        from langchain_experimental.llms import ChatLlamaAPI

        llm = ChatLlamaAPI(client=llama)

    elif llm_ver == LLM_GEMINI:

        from langchain_google_genai import ChatGoogleGenerativeAI

        llm = ChatGoogleGenerativeAI(
            model="gemini-pro", google_api_key=get_gemini_api_key()
        )

    return llm


GENERAL_QUERY_PROMPT_TEMPLATE = """Answer the following question. Provide succinct, yet scientifically accurate
    answers. Question: {question}

    Answer: """

GENERAL_QUERY_LIMITED_PROMPT_TEMPLATE = """You are a neuroscientist who is answering questions about the worm C. elegans. Provide succinct, yet scientifically accurate
    answers. If the question is not related to biology, physics or chemistry, then don't answer the question, but instead explain that you 
    can currently only answer questions related to C. elegans. Question: {question}

    Answer: """


def generate_response(input_text, llm_ver, temperature, only_celegans):

    template = (
        GENERAL_QUERY_LIMITED_PROMPT_TEMPLATE
        if only_celegans
        else GENERAL_QUERY_PROMPT_TEMPLATE
    )
    prompt = PromptTemplate(template=template, input_variables=["question"])

    llm = get_llm(llm_ver, temperature)

    llm_chain = prompt | llm | StrOutputParser()

    response = llm_chain.invoke(input_text)

    return response


def generate_panel_response(input_text, llm_panelists, llm_panel_chair, temperature):

    responses = {}

    for llm_ver in llm_panelists:

        prompt = PromptTemplate(
            template=GENERAL_QUERY_PROMPT_TEMPLATE, input_variables=["question"]
        )

        llm = get_llm(llm_ver, temperature)

        llm_chain = prompt | llm | StrOutputParser()

        responses[llm_ver] = llm_chain.invoke(input_text)

    panel_chair_prompt = """You are a neuroscientist chairing a panel discussion on the nematode C. elegans. A researcher has asked the following question:
    {question}
    and %i experts on the panel have give their answers. 
    """ % (
        len(llm_panelists)
    )

    for llm_ver in llm_panelists:
        panel_chair_prompt += """
The panelist named Dr. %s has provided the answer: %s
""" % (
            llm_ver,
            responses[llm_ver],
        )

    panel_chair_prompt += (
        """
Please generate a brief answer to the researcher's question based on their responses, pointing out where there is any inconsistency"""
        + """ in their answers, and using your own knowledge of C. elegans to try to resolve it."""
    )

    print(panel_chair_prompt)

    prompt = PromptTemplate(template=panel_chair_prompt, input_variables=["question"])

    llm = get_llm(llm_panel_chair, temperature)

    llm_chain = prompt | llm | StrOutputParser()

    response_chair = llm_chain.invoke(input_text)

    response = """**%s**: %s""" % (llm_panel_chair, response_chair)

    response += """

-----------------------------------
_Individual responses:_

"""
    for llm_ver in responses:
        response += """
_**%s**:_ _%s_
""" % (
            llm_ver,
            responses[llm_ver].strip().replace("\n", " "),
        )

    return response