How to Create AI Agents Locally with Custom Data

 This video is a step-by-step tutorial to locally create AI agents with your own data with LlamaIndex and Ollama.

import json

from llama_index.core.tools import QueryEngineTool, ToolMetadata

from llama_index.core.agent import ReActAgent

from llama_index.llms.ollama import Ollama

from llama_index.core import Settings

from llama_index.core import (

    SimpleDirectoryReader,

    VectorStoreIndex,

    StorageContext,

    load_index_from_storage,

)

llm = Ollama(model="llama3:latest", request_timeout=120.0)

Settings.llm = llm

try:

    storage_context = StorageContext.from_defaults(persist_dir="./storage/oracle")

    dba_index = load_index_from_storage(storage_context)

    index_loaded = True

except:

    index_loaded = False

print (index_loaded)    

oracle_docs = SimpleDirectoryReader(input_files=["oracledocs.pdf"]).load_data()

oracle_index = VectorStoreIndex.from_documents(oracle_docs)

# persist index

oracle_index.storage_context.persist(persist_dir="./storage/oracle")

oracle_engine = oracle_index.as_query_engine(similarity_top_k=3)

query_engine_tools = [

    QueryEngineTool(

        query_engine=oracle_engine,

        metadata=ToolMetadata(

            name="oracledocs",

            description=(

                "Provides information about Oracle. "

                "Use a detailed plain text question as input to the tool."

            ),

        ),

    ),

]

agent = ReActAgent.from_tools(query_engine_tools,llm=llm,verbose=True)

response = agent.chat("What is Oracle?")

print(str(response))

response = agent.chat(

    "What backup options are available in Oracle database?"

)

print(str(response))