app.py

import openai
import random
import time
import gradio as gr
import os
from langchain.embeddings.openai import OpenAIEmbeddings
from langchain.vectorstores import DeepLake
from langchain.chat_models import ChatOpenAI
from langchain.chains import ConversationalRetrievalChain
from langchain.document_loaders import TextLoader
from langchain.text_splitter import CharacterTextSplitter
from langchain.document_loaders import PyPDFDirectoryLoader
from langchain.memory import ConversationBufferMemory
from langchain.llms import OpenAI

def set_api_key(key):
 os.environ["OPENAI_API_KEY"] = key
 return f"Your API Key has been set to: {key}"

def reset_api_key():
 os.environ["OPENAI_API_KEY"] = ""
 return "Your API Key has been reset"

def get_api_key():
 api_key = os.getenv("OPENAI_API_KEY")
 return api_key

def set_model(model):
 os.environ["OPENAI_MODEL"] = model
 return f"{model} selected"

def get_model():
 model = os.getenv("OPENAI_MODEL")
 return model 

def upload_file(files):
 file_paths = [file.name for file in files]
 return file_paths

def create_vectorstore(files):
 pdf_dir = files.name
 pdf_loader = PyPDFDirectoryLoader(pdf_dir)
 pdf_docs = pdf_loader.load_and_split()
 text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
 texts = text_splitter.split_documents(pdf_docs)
 embeddings = OpenAIEmbeddings()
 db = DeepLake.from_documents(texts, dataset_path="./documentation_db", embedding=embeddings, overwrite=True)
 return "Vectorstore Successfully Created"

def respond(message, chat_history):

 # Get embeddings
 embeddings = OpenAIEmbeddings()
 
 #Connect to existing vectorstore
 db = DeepLake(dataset_path="./documentation_db", embedding_function=embeddings, read_only=True)
 #Set retriever settings
 retriever = db.as_retriever(search_kwargs={"distance_metric":'cos',
 "fetch_k":10,
 "maximal_marginal_relevance":True,
 "k":10})

 if len(chat_history) != 0: 
 chat_history = [(chat_history[0][0], chat_history[0][1])]

 model = get_model()
 # Create ChatOpenAI and ConversationalRetrievalChain
 model = ChatOpenAI(model=model)
 qa = ConversationalRetrievalChain.from_llm(model, retriever)

 bot_message = qa({"question": message, "chat_history": chat_history})
 chat_history = [(message, bot_message["answer"])]
 time.sleep(1)
 return "", chat_history

with gr.Blocks() as demo:
 #create chat history 
 chat_history = []
 with gr.Row():
 #create textbox for API input
 api_input = gr.Textbox(label = "API Key",
 placeholder = "Please provide your OpenAI API key here.")
 #create textbox to validate API
 api_key_status = gr.Textbox(label = "API Key Status",
 placeholder = "Your API Key has not be set yet. Please enter your key.",
 interactive = False)
 #create button to submit API key
 api_submit_button = gr.Button("Submit")
 #set api_submit_button functionality
 api_submit_button.click(set_api_key, inputs=api_input, outputs=api_key_status)
 #create button to reset API key
 api_reset_button = gr.Button("Clear API Key from session")
 #set api_reset_button functionality
 api_reset_button.click(reset_api_key, outputs=api_key_status)
 
 with gr.Row():
 with gr.Column():
 #create dropdown to select model (gpt-3.5-turbo or gpt4)
 model_selection = gr.Dropdown(
 ["gpt-3.5-turbo", "gpt-4"], label="Model Selection", info="Please ensure you provide the API Key that corresponds to the Model you select!"
 )
 #create button to submit model selection
 model_submit_button = gr.Button("Submit Model Selection")
 model_status = gr.Textbox(label = "Selected Model", interactive = False, lines=4)
 #set model_submit_button functionality
 model_submit_button.click(set_model, inputs=model_selection, outputs=model_status)
 
 file_output = gr.File(label = "Uploaded files - Please note these files are persistent and will not be automatically deleted")
 upload_button = gr.UploadButton("Click to Upload a PDF File", file_types=["pdf"], file_count="multiple")
 upload_button.upload(upload_file, upload_button, file_output)
 create_vectorstore_button = gr.Button("Click to create the vectorstore for your uploaded documents")
 db_output = gr.Textbox(label = "Vectorstore Status")
 create_vectorstore_button.click(create_vectorstore, inputs=file_output, outputs = db_output)
 
 chatbot = gr.Chatbot(label="ChatGPT Powered Grant Writing Assistant")
 msg = gr.Textbox(label="User Prompt", placeholder="Your Query Here")
 clear = gr.Button("Clear")
 
 msg.submit(respond, inputs = [msg, chatbot], outputs = [msg, chatbot])
 clear.click(lambda: None, None, chatbot, queue=False)
 
 
demo.launch()