vector_store_retriever.py

import gradio as gr
from langchain.document_loaders import TextLoader
from langchain.vectorstores import Chroma
from langchain.chains import RetrievalQA
from langchain.embeddings import HuggingFaceInstructEmbeddings 
from langchain.agents import Tool

# Initialize the HuggingFaceInstructEmbeddings
hf = HuggingFaceInstructEmbeddings(
  model_name="hkunlp/instructor-large",
  embed_instruction="Represent the document for retrieval: ",
  query_instruction="Represent the query for retrieval: "
)

# Example texts for the vector store
texts=["The meaning of life is to love","The meaning of vacation is to relax","Roses are red.","Hack the planet!"]

# Create a Chroma vector store from the example texts
db = Chroma.from_texts(texts, hf, collection_name="my-collection")

# Create a RetrievalQA chain
llm = LLM.from_model("lgaalves/gpt2-dolly")  # Replace with the appropriate LLM model
docsearcher = RetrievalQA.from_chain_type(
  llm=llm, 
  chain_type="stuff",  # Replace with the appropriate chain type
  return_source_documents=False,
  retriever=db.as_retriever(search_type="similarity", search_kwargs={"k": 1})
)

class VectorStoreRetrieverTool(Tool):
    name = "vectorstore_retriever"
    description = "This tool uses LangChain's RetrievalQA to find relevant answers from a vector store based on a given query."

    inputs = ["text"]
    outputs = ["text"]

    def __call__(self, query: str):
        # Run the query through the RetrievalQA chain
        response = docsearcher.run(query)
        return response

# Create the Gradio interface using the HuggingFaceTool
tool = gr.Interface(
    VectorStoreRetrieverTool(),
    live=True,
    title="LangChain-Application: Vectorstore-Retriever",
    description="This tool uses LangChain's RetrievalQA to find relevant answers from a vector store based on a given query.",
)

# Launch the Gradio interface
tool.launch()