app.py

import gradio as gr
import random
import time

from langchain import PromptTemplate
from langchain.llms import OpenAI
from langchain.chat_models import ChatOpenAI
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.vectorstores import Pinecone
from langchain.chains import LLMChain
from langchain.chains.retrieval_qa.base import RetrievalQA
from langchain.chains.question_answering import load_qa_chain
import pinecone

import os
os.environ["TOKENIZERS_PARALLELISM"] = "false"

#OPENAI_API_KEY = ""
OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY", "")
OPENAI_TEMP  = 1

PINECONE_KEY = os.environ.get("PINECONE_KEY", "")
PINECONE_ENV = os.environ.get("PINECONE_ENV", "asia-northeast1-gcp")
PINECONE_INDEX = os.environ.get("PINECONE_INDEX", "3gpp")

EMBEDDING_MODEL = os.environ.get("PINECONE_INDEX", "sentence-transformers/all-mpnet-base-v2")

# return top-k text chunks from vector store
TOP_K_DEFAULT = 10
TOP_K_MAX = 25


BUTTON_MIN_WIDTH = 210

STATUS_NOK = "404-MODEL UNREADY-critical"
STATUS_OK  = "200-MODEL LOADED-9cf"

FORK_BADGE = "Fork-HuggingFace Space-9cf"


def get_logo(inputs, logo) -> str:
    return f"""https://img.shields.io/badge/{inputs}?style=flat&logo={logo}&logoColor=white"""

def get_status(inputs) -> str:
    return f"""<img
    src   = "{get_logo(inputs, "openai")}";
    style = "margin: 0 auto;"
    >"""
    

KEY_INIT   = "Initialize Model"
KEY_SUBMIT = "Submit"
KEY_CLEAR  = "Clear"

MODEL_NULL = get_status(STATUS_NOK)
MODEL_DONE = get_status(STATUS_OK)

MODEL_WARNING = f"Please paste your OpenAI API Key from \
[openai.com](https://platform.openai.com/account/api-keys) and then **{KEY_INIT}**"


TAB_1 = "Chatbot"

FAVICON = './icon.svg'

LLM_LIST = ["gpt-3.5-turbo", "text-davinci-003"]


DOC_1 = '3GPP'
DOC_2 = 'HTTP2'

DOC_SUPPORTED = [DOC_1, DOC_2]
DOC_DEFAULT = [DOC_1]

webui_title = """
# OpenAI Chatbot Based on Vector Database
## Example of 3GPP
"""

dup_link = f'''<a href="https://huggingface.co/spaces/ShawnAI/3GPP-ChatBot?duplicate=true">
<img src="{get_logo(FORK_BADGE, "addthis")}"></a> '''

init_message = f"""Welcome to use 3GPP Chatbot, this demo toolkit is based on OpenAI with LangChain and Pinecone
    1. Insert your OpenAI API key and click  `{KEY_INIT}`
    2. Insert your Question and click  `{KEY_SUBMIT}`
"""



#----------------------------------------------------------------------------------------------------------
#----------------------------------------------------------------------------------------------------------

def init_model(api_key, emb_name, db_api_key, db_env, db_index):
    try:
        if (api_key and api_key.startswith("sk-") and len(api_key) > 50) and \
        (emb_name and db_api_key and db_env and db_index):
            
            embeddings = HuggingFaceEmbeddings(model_name=emb_name)

            pinecone.init(api_key     = db_api_key,
                          environment = db_env)

            #llm = OpenAI(temperature=OPENAI_TEMP, model_name="gpt-3.5-turbo-0301")

            
            llm_dict = {}
            for llm_name in LLM_LIST:
                if llm_name == "gpt-3.5-turbo":
                    llm_dict[llm_name] = ChatOpenAI(model_name=llm_name,
                                                    temperature = OPENAI_TEMP,
                                                    openai_api_key = api_key)
                else:
                    llm_dict[llm_name] = OpenAI(model_name=llm_name,
                                                temperature = OPENAI_TEMP,
                                                openai_api_key = api_key)

            '''
            ChatOpenAI(model_name="gpt-3.5-turbo",
                             temperature = OPENAI_TEMP,
                             openai_api_key = api_key)
            chain_1 = load_qa_chain(llm, chain_type="stuff")
            
            #LLMChain(llm=llm, prompt=condense_question_prompt)

            chain_2 = LLMChain(llm = llm,
                             prompt = PromptTemplate(template='{question}',
                                                     input_variables=['question']),
                             output_key = 'output_text')
                             '''
            
            db = Pinecone.from_existing_index(index_name = db_index,
                                              embedding  = embeddings)

            return api_key, MODEL_DONE, llm_dict, None, db, None
        else:
            return None,MODEL_NULL,None,None,None,None
    except Exception as e:
        print(e)
        return None,MODEL_NULL,None,None,None,None


def get_chat_history(inputs) -> str:
    res = []
    for human, ai in inputs:
        res.append(f"Human: {human}\nAI: {ai}")
    return "\n".join(res)

def remove_duplicates(documents):
    seen_content = set()
    unique_documents = []
    for doc in documents:
        if doc.page_content not in seen_content:
            seen_content.add(doc.page_content)
            unique_documents.append(doc)
    return unique_documents

def doc_similarity(query, db, top_k):
    docsearch = db.as_retriever(search_kwargs={'k':top_k})
    docs = docsearch.get_relevant_documents(query)
    udocs = remove_duplicates(docs)
    return udocs

def user(user_message, history):
    return "", history+[[user_message, None]]

def bot(box_message, ref_message,
        llm_dropdown, llm_dict, doc_list,
        db, top_k):

    # bot_message = random.choice(["Yes", "No"])
    # 0 is user question, 1 is bot response
    question = box_message[-1][0]
    history  = box_message[:-1]
    
    if (not llm_dict) or (not doc_check) or (not db):
        box_message[-1][1] = MODEL_WARNING
        return box_message, "", ""

    if not ref_message:
        ref_message = question
        details = f"Q:  {question}"
    else:
        details = f"Q:  {question}\nR: {ref_message}"
        
        
    llm = llm_dict[llm_dropdown]
    
    print(llm)
    print(doc_list)
    
    if DOC_1 in doc_list:
        chain = load_qa_chain(llm, chain_type="stuff")
        docs = doc_similarity(ref_message, db, top_k)
        delta_top_k = top_k - len(docs)

        if delta_top_k > 0:
            docs = doc_similarity(ref_message, db, top_k+delta_top_k)
            
    else:
        chain = LLMChain(llm = llm,
                         prompt = PromptTemplate(template='{question}',
                                                input_variables=['question']),
                         output_key = 'output_text')
        docs = []

    all_output = chain({"input_documents": docs,
                        "question": question,
                        "chat_history": get_chat_history(history)})
    
    bot_message = all_output['output_text']


    source = "".join([f"""<details> <summary>{doc.metadata["source"]}</summary>
{doc.page_content}

</details>""" for i, doc in enumerate(docs)])

    #print(source)

    box_message[-1][1] = bot_message
    return box_message, "", [[details, bot_message + source]]

#----------------------------------------------------------------------------------------------------------
#----------------------------------------------------------------------------------------------------------

with gr.Blocks(
    title = TAB_1,
    theme = "Base",
    css = """.bigbox {
    min-height:250px;
}
""") as demo:
    llm = gr.State()
    chain_2 = gr.State() # not inuse
    vector_db = gr.State()
    gr.Markdown(webui_title)
    gr.HTML(dup_link)
    gr.Markdown(init_message)
    
    with gr.Row():
        with gr.Column(scale=10):
            llm_api_textbox = gr.Textbox(
                label = "OpenAI API Key",
                # show_label = False,
                value = OPENAI_API_KEY,
                placeholder = "Paste Your OpenAI API Key (sk-...) and Hit ENTER",
                lines=1,
                type='password')
            
        with gr.Column(scale=1, min_width=BUTTON_MIN_WIDTH):
            
            init = gr.Button(KEY_INIT) #.style(full_width=False)
            model_statusbox = gr.HTML(MODEL_NULL)
    
    with gr.Tab(TAB_1):
        with gr.Row():
            with gr.Column(scale=10):
                chatbot = gr.Chatbot(elem_classes="bigbox")
            #with gr.Column(scale=1):
            with gr.Column(scale=1, min_width=BUTTON_MIN_WIDTH):
                doc_check = gr.CheckboxGroup(choices = DOC_SUPPORTED,
                                             value   = DOC_DEFAULT,
                                             label   = "Reference Docs",
                                             interactive=True)
                llm_dropdown = gr.Dropdown(LLM_LIST,
                                           value=LLM_LIST[0],
                                           multiselect=False,
                                           interactive=True,
                                           label="LLM Selection",
                                           )
        with gr.Row():
            with gr.Column(scale=10):
                query = gr.Textbox(label="Question:",
                                   lines=2)
                ref = gr.Textbox(label="Reference(optional):")

            with gr.Column(scale=1, min_width=BUTTON_MIN_WIDTH):

                clear = gr.Button(KEY_CLEAR)
                submit = gr.Button(KEY_SUBMIT,variant="primary")
                

    with gr.Tab("Details"):
        top_k = gr.Slider(1,
                          TOP_K_MAX,
                          value=TOP_K_DEFAULT,
                          step=1,
                          label="Vector similarity top_k",
                          interactive=True)
        detail_panel = gr.Chatbot(label="Related Docs")
    
    with gr.Tab("Database"):
        with gr.Row():
            emb_textbox = gr.Textbox(
                label = "Embedding Model",
                # show_label = False,
                value = EMBEDDING_MODEL,
                placeholder = "Paste Your Embedding Model Repo on HuggingFace",
                lines=1,
                interactive=True,
                type='email')
        with gr.Accordion("Pinecone Database for "+DOC_1):
            with gr.Row():
                db_api_textbox = gr.Textbox(
                    label = "Pinecone API Key",
                    # show_label = False,
                    value = PINECONE_KEY,
                    placeholder = "Paste Your Pinecone API Key (xx-xx-xx-xx-xx) and Hit ENTER",
                    lines=1,
                    interactive=True,
                    type='password')
            with gr.Row():
                db_env_textbox = gr.Textbox(
                    label = "Pinecone Environment",
                    # show_label = False,
                    value = PINECONE_ENV,
                    placeholder = "Paste Your Pinecone Environment (xx-xx-xx) and Hit ENTER",
                    lines=1,
                    interactive=True,
                    type='email')
                db_index_textbox = gr.Textbox(
                    label = "Pinecone Index",
                    # show_label = False,
                    value = PINECONE_INDEX,
                    placeholder = "Paste Your Pinecone Index (xxxx) and Hit ENTER",
                    lines=1,
                    interactive=True,
                    type='email')

    init_input  = [llm_api_textbox, emb_textbox, db_api_textbox, db_env_textbox, db_index_textbox]
    init_output = [llm_api_textbox, model_statusbox,
                   llm, chain_2,
                   vector_db, chatbot]
                
    llm_api_textbox.submit(init_model, init_input, init_output)
    init.click(init_model, init_input, init_output)
    
    submit.click(user,
                 [query, chatbot],
                 [query, chatbot],
                 queue=False).then(
        bot,
        [chatbot, ref,
         llm_dropdown, llm, doc_check,
         vector_db, top_k],
        [chatbot, ref, detail_panel]
    )
    
    clear.click(lambda: (None,None,None), None, [query, ref, chatbot], queue=False)

#----------------------------------------------------------------------------------------------------------
#----------------------------------------------------------------------------------------------------------
    
if __name__ == "__main__":
    demo.launch(share        = False,
                inbrowser    = True,
                favicon_path = FAVICON)