app.py

from langchain.chains import ConversationalRetrievalChain
from langchain.chains.question_answering import load_qa_chain
from langchain.chains import RetrievalQA
from langchain.memory import ConversationBufferMemory
from langchain.memory import ConversationTokenBufferMemory
from langchain.llms import HuggingFacePipeline
# from langchain import PromptTemplate
from langchain.prompts import PromptTemplate
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
from langchain.vectorstores import Chroma
from langchain.embeddings import HuggingFaceBgeEmbeddings
from langchain.document_loaders import (
    CSVLoader,
    DirectoryLoader,
    GitLoader,
    NotebookLoader,
    OnlinePDFLoader,
    PythonLoader,
    TextLoader,
    UnstructuredFileLoader,
    UnstructuredHTMLLoader,
    UnstructuredPDFLoader,
    UnstructuredWordDocumentLoader,
    WebBaseLoader,
    PyPDFLoader,
    UnstructuredMarkdownLoader,
    UnstructuredEPubLoader,
    UnstructuredHTMLLoader,
    UnstructuredPowerPointLoader,
    UnstructuredODTLoader,
    NotebookLoader,
    UnstructuredFileLoader
)
from transformers import (
    AutoModelForCausalLM,
    AutoTokenizer,
    StoppingCriteria,
    StoppingCriteriaList,
    pipeline,
    GenerationConfig,
    TextStreamer,
    pipeline
)
from langchain.llms import HuggingFaceHub
import torch
from transformers import BitsAndBytesConfig
import os
from langchain.llms import CTransformers
import streamlit as st
from langchain.document_loaders.base import BaseLoader
from langchain.schema import Document
import gradio as gr
import tempfile
import timeit

FILE_LOADER_MAPPING = {
    "csv": (CSVLoader, {"encoding": "utf-8"}),
    "doc": (UnstructuredWordDocumentLoader, {}),
    "docx": (UnstructuredWordDocumentLoader, {}),
    "epub": (UnstructuredEPubLoader, {}),
    "html": (UnstructuredHTMLLoader, {}),
    "md": (UnstructuredMarkdownLoader, {}),
    "odt": (UnstructuredODTLoader, {}),
    "pdf": (PyPDFLoader, {}),
    "ppt": (UnstructuredPowerPointLoader, {}),
    "pptx": (UnstructuredPowerPointLoader, {}),
    "txt": (TextLoader, {"encoding": "utf8"}),
    "ipynb": (NotebookLoader, {}),
    "py": (PythonLoader, {}),
    # Add more mappings for other file extensions and loaders as needed
}

def load_model():
    config = {'max_new_tokens': 1024,
              'repetition_penalty': 1.1,
              'temperature': 0.1,
              'top_k': 50,
              'top_p': 0.9,
              'stream': True,
              'threads': int(os.cpu_count() / 2)
            }
    
    llm = CTransformers(
        model = "TheBloke/zephyr-7B-beta-GGUF",
        model_file = "zephyr-7b-beta.Q4_0.gguf",
        callbacks=[StreamingStdOutCallbackHandler()],
        lib="avx2", #for CPU use
        **config
        # model_type=model_type,
        # max_new_tokens=max_new_tokens,  # type: ignore
        # temperature=temperature,  # type: ignore
    )
    return llm