trueGPT.py

from dotenv import load_dotenv
from langchain.chains import RetrievalQA
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
from langchain.vectorstores import Chroma
from langchain.llms import GPT4All, LlamaCpp
import os
import argparse

load_dotenv()

embeddings_model_name = os.environ.get("EMBEDDINGS_MODEL_NAME")
persist_directory = os.environ.get('PERSIST_DIRECTORY')

model_type = os.environ.get('MODEL_TYPE')
model_path = os.environ.get('MODEL_PATH')
model_n_ctx = os.environ.get('MODEL_N_CTX')

from constants import CHROMA_SETTINGS

def get_response(user_input):
    embeddings = HuggingFaceEmbeddings(model_name=embeddings_model_name)
    db = Chroma(persist_directory=persist_directory, embedding_function=embeddings, client_settings=CHROMA_SETTINGS)
    retriever = db.as_retriever()
    # Activate/deactivate the streaming StdOut callback for LLMs
    callbacks = []
    # Prepare the LLM
    match model_type:
        case "LlamaCpp":
            llm = LlamaCpp(model_path=model_path, n_ctx=model_n_ctx, callbacks=callbacks, verbose=False)
        case "GPT4All":
            llm = GPT4All(model=model_path, n_ctx=model_n_ctx, backend='gptj', callbacks=callbacks, verbose=False)
        case _default:
            print(f"Model {model_type} not supported!")
            exit;

    qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=retriever, return_source_documents=False)

    # Get the answer from the chain
    res = qa(user_input)
    answer = res['result']

    return answer