File size: 3,498 Bytes
c561ff2 a74827b c561ff2 e4c7fbd | 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 89 90 91 92 93 94 95 96 97 98 99 100 | import os
import chromadb
import gradio as gr
from dotenv import load_dotenv
from llama_index.core import VectorStoreIndex
from llama_index.core.agent import ReActAgent
from llama_index.core.tools import QueryEngineTool
from llama_index.embeddings.huggingface import HuggingFaceEmbedding
from llama_index.llms.gemini import Gemini
from llama_index.core.workflow import Context
from llama_index.vector_stores.chroma import ChromaVectorStore
load_dotenv()
agent = None
conversation_context = None
async def initialize_agent():
"""Initialize the agent once"""
global agent, conversation_context
if agent is not None:
return agent, conversation_context
llm = Gemini(
model="models/gemini-flash-latest",
api_key=os.getenv("GEMINI_API"),
temperature=0.3,
)
embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-small-en-v1.5")
db = chromadb.PersistentClient(path="./product_db")
chroma_collection = db.get_collection(name="product_catalog")
vector_store = ChromaVectorStore(chroma_collection=chroma_collection)
index = VectorStoreIndex.from_vector_store(
vector_store=vector_store,
embed_model=embed_model
)
query_engine = index.as_query_engine(llm=llm)
query_tool = QueryEngineTool.from_defaults(
query_engine=query_engine,
name="ProductInfoTool",
description="A tool to retrieve information about camping products, including their stock availability.",
)
agent = ReActAgent(
llm=llm,
tools=[query_tool],
verbose=False,
system_prompt="""You are a friendly and knowledgeable camping gear expert.
Your goal is to find the perfect product for the user and tell them about it in a helpful, conversational way.
Use the `ProductInfoTool` to find the best match for the user's query.
In your final response to the user, you MUST include the following three pieces of information:
1. The full product name.
2. A brief, one-sentence reason why it's a good choice for them.
3. The exact stock status (e.g., '15 available' or 'out of stock').
If the tool cannot find a suitable product, just say: 'I'm sorry, I couldn't find a product that matches your request.'
Remember conversation context and refer back to previous messages when appropriate."""
)
conversation_context = Context(agent)
return agent, conversation_context
async def chat_with_agent(message, history):
"""Handle chat messages with the agent"""
global conversation_context, agent
try:
agent, ctx = await initialize_agent()
response = await agent.run(message, ctx=conversation_context)
return str(response.response)
except Exception as e:
if "index out of range" in str(e):
conversation_context = Context(agent)
response = await agent.run(message, ctx=conversation_context)
return str(response.response)
return f"Sorry, I encountered an error: {str(e)}"
def main():
"""Launch the simple Gradio interface"""
demo = gr.ChatInterface(
fn=chat_with_agent,
title="🏕️ Tory - The Camping Gear Assistant",
description="Ask me about camping products and I'll help you find the perfect gear!",
examples=[
"I need a lightweight tent for 2 people",
"What sleeping bags do you have?",
"Show me available camping stoves"
],
)
demo.launch()
if __name__ == "__main__":
main() |