Add the retriever code

.env

@@ -1,2 +1,4 @@
+SUPABASE_URL = "https://jaqnaccpocxhsavlkdpt.supabase.co/"
+SUPABASE_KEY = "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJzdXBhYmFzZSIsInJlZiI6ImphcW5hY2Nwb2N4aHNhdmxrZHB0Iiwicm9sZSI6ImFub24iLCJpYXQiOjE3NTE4MDk2MjYsImV4cCI6MjA2NzM4NTYyNn0.5r37tq8gA_BEumvi614fbovfs6qlb6qxDX0gcB5mo3o"
 SERPAPI_KEY = "046aa74b206478f196088246569b77000619bcae1c5ccda5d0143ce274a932e1"
-OPENAI_API_KEY = "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"
+OPENAI_API_KEY = "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"

app.py

@@ -5,7 +5,7 @@ import inspect
 import pandas as pd
 from langgraph.prebuilt import ToolNode, tools_condition
 from langchain_core.messages import HumanMessage
-from tools import assistant, AgentState, tools
+from tools import assistant, AgentState, tools, retriever
 from langgraph.graph import StateGraph, START
 # (Keep Constants as is)
 # --- Constants ---
@@ -16,9 +16,11 @@ DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 class BasicAgent:
     def __init__(self):
         builder = StateGraph(AgentState)
+        builder.add_node("retriever", retriever)
         builder.add_node("assistant", assistant)
         builder.add_node("tools", ToolNode(tools))
-        builder.add_edge(START, "assistant")
+        builder.add_edge(START, "retriever")
+        builder.add_conditional_edges("retriever", "assistant")
         builder.add_conditional_edges("assistant", tools_condition)
         builder.add_edge("tools", "assistant")
         

pyproject.toml

@@ -6,7 +6,7 @@ authors = [
     {name = "sharath"}
 ]
 readme = "README.md"
-requires-python = ">=3.9"
+requires-python = ">=3.10"
 dependencies = [
 "gradio",
 "requests",
@@ -20,9 +20,14 @@ dependencies = [
 "langchain-openai",
 "serpapi",
 "python-dotenv",
-"google-search-results"
+"google-search-results",
+"langchain-huggingface",
+"langchain-community",
+"supabase",
+"sentence-transformers"
 ]
-
+[tool.poetry]
+package-mode = false
 [build-system]
 requires = ["poetry-core>=2.0.0,<3.0.0"]
 build-backend = "poetry.core.masonry.api"

requirements.txt

@@ -10,4 +10,8 @@ langchain
 langchain-openai
 serpapi
 python-dotenv
-google-search-results
+google-search-results
+langchain-huggingface
+langchain-community
+supabase
+sentence-transformers

tools.py

@@ -2,6 +2,9 @@ from typing import List, TypedDict, Annotated, Optional, Dict, Union
 from langchain_openai import ChatOpenAI
 from langchain_core.messages import SystemMessage, HumanMessage, AnyMessage
 from langgraph.graph.message import add_messages
+from langchain_community.vectorstores import SupabaseVectorStore
+from supabase.client import  create_client
+from langchain_huggingface import HuggingFaceEmbeddings
 from serpapi import GoogleSearch
 from dotenv import load_dotenv
 import os
@@ -47,6 +50,25 @@ llm = ChatOpenAI(model = "gpt-3.5-turbo")
 tools = [add, subtract, divide, web_search]
 llm_with_tools = llm.bind_tools(tools, parallel_tool_calls = False)
 
+def retriever(query: str) -> str:
+    """
+    Retrieve the answer fom vector database instead of searching if we found a user query similar to which is already found in the dataset
+    
+    """
+    supabase_url = os.environ.get("SUPABASE_URL")
+    supabase_key = os.environ.get("SUPABASE_KEY")
+    supabase = create_client(supabase_url, supabase_key)
+    embeddings = HuggingFaceEmbeddings(model_name = "sentence-transformers/all-mpnet-base-v2")
+    vector_store = SupabaseVectorStore(
+    embedding=embeddings,
+    client=supabase,
+    table_name="documents",
+    query_name="match_documents",
+)
+    
+    docs = vector_store.similarity_search(query = query, k = 1)
+    return docs[0].page_content.split("Final answer : ")[-1]
+
 def assistant(state: AgentState) -> Dict:
     
     system_message = """