Spaces:
Sleeping
Sleeping
File size: 3,753 Bytes
ef24768 af98e35 e379c49 d24775b a59ef83 ea89a3e ef24768 c84aaa1 ef24768 c84aaa1 fb4b26a 1299579 fb4b26a c84aaa1 1299579 ef24768 1299579 ef24768 1299579 ef24768 e379c49 ef24768 65913cc ef24768 cbd1f0b 1299579 ef24768 1299579 ef24768 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 |
from pymongo import MongoClient
# error since Jan 2024, from langchain.embeddings.openai import OpenAIEmbeddings
from langchain.embeddings import OpenAIEmbeddings
# error since Jan 2024, from langchain.vectorstores import MongoDBAtlasVectorSearch
from langchain_community.vectorstores import MongoDBAtlasVectorSearch
# error since Jan 2024, from langchain.document_loaders import DirectoryLoader
from langchain_community.document_loaders import DirectoryLoader
# error since Jan 2024, from langchain.llms import OpenAI
from langchain_community.llms import OpenAI
from langchain.chains import RetrievalQA
import gradio as gr
from gradio.themes.base import Base
#import key_param
import os
def query_data(query,openai_api_key,mongo_uri):
#openai_api_key = os.getenv("OPENAI_API_KEY")
#mongo_uri = os.getenv("MONGO_URI")
client = MongoClient(mongo_uri)
dbName = "langchain_demo"
collectionName = "collection_of_text_blobs"
collection = client[dbName][collectionName]
# Define the text embedding model
embeddings = OpenAIEmbeddings(openai_api_key=openai_api_key)
# Initialize the Vector Store
vectorStore = MongoDBAtlasVectorSearch( collection, embeddings, index_name="default" )
# Convert question to vector using OpenAI embeddings
# Perform Atlas Vector Search using Langchain's vectorStore
# similarity_search returns MongoDB documents most similar to the query
docs = vectorStore.similarity_search(query, K=1)
as_output = docs[0].page_content
# Leveraging Atlas Vector Search paired with Langchain's QARetriever
# Define the LLM that we want to use -- note that this is the Language Generation Model and NOT an Embedding Model
# If it's not specified (for example like in the code below),
# then the default OpenAI model used in LangChain is OpenAI GPT-3.5-turbo, as of August 30, 2023
llm = OpenAI(openai_api_key=openai_api_key, temperature=0, model_name='gpt-4-1106-preview')
# Get VectorStoreRetriever: Specifically, Retriever for MongoDB VectorStore.
# Implements _get_relevant_documents which retrieves documents relevant to a query.
retriever = vectorStore.as_retriever()
# Load "stuff" documents chain. Stuff documents chain takes a list of documents,
# inserts them all into a prompt and passes that prompt to an LLM.
qa = RetrievalQA.from_chain_type(llm, chain_type="stuff", retriever=retriever, model_name='gpt-4-1106-preview')
# Execute the chain
retriever_output = qa.run(query)
# Return Atlas Vector Search output, and output generated using RAG Architecture
return as_output, retriever_output
# Create a web interface for the app, using Gradio
with gr.Blocks(theme=Base(), title="Question Answering App using Vector Search + RAG") as demo:
gr.Markdown(
"""
# Question Answering App using Atlas Vector Search + RAG Architecture
""")
openai_api_key = gr.Textbox(label = "OpenAI 3.5 API Key", value = "sk-", lines = 1)
mongo_uri = gr.Textbox(label = "Mongo URI", value = "mongodb+srv://", lines = 1)
textbox = gr.Textbox(label="Enter your Question:")
with gr.Row():
button = gr.Button("Submit", variant="primary")
with gr.Column():
output1 = gr.Textbox(lines=1, max_lines=10, label="Output with just Atlas Vector Search (returns text field as is):")
output2 = gr.Textbox(lines=1, max_lines=10, label="Output generated by chaining Atlas Vector Search to Langchain's RetrieverQA + OpenAI LLM:")
# Call query_data function upon clicking the Submit button
button.click(query_data,
inputs=[textbox, openai_api_key, mongo_uri],
outputs=[output1, output2]
)
demo.launch()
|