Function Calling AWS Bedrock Converse Agent¶

This notebook shows you how to use our AWS Bedrock Converse agent, powered by function calling capabilities.

Initial Setup¶

Let's start by importing some simple building blocks.

The main thing we need is:

1. AWS credentials with access to Bedrock and the Claude Haiku LLM
2. a place to keep conversation history
3. a definition for tools that our agent can use.

If you're opening this Notebook on colab, you will probably need to install LlamaIndex 🦙.

In [ ]:

%pip install llama-index-llms-bedrock-converse
%pip install llama-index-embeddings-huggingface

%pip install llama-index-llms-bedrock-converse %pip install llama-index-embeddings-huggingface

In [ ]:

!pip install llama-index

!pip install llama-index

In [ ]:

from llama_index.llms.bedrock_converse import BedrockConverse
from llama_index.core.tools import FunctionTool

import nest_asyncio

nest_asyncio.apply()

from llama_index.llms.bedrock_converse import BedrockConverse from llama_index.core.tools import FunctionTool import nest_asyncio nest_asyncio.apply()

Let's define some very simple calculator tools for our agent.

In [ ]:

def multiply(a: int, b: int) -> int:
    """Multiple two integers and returns the result integer"""
    return a * b


multiply_tool = FunctionTool.from_defaults(fn=multiply)

def multiply(a: int, b: int) -> int: """Multiple two integers and returns the result integer""" return a * b multiply_tool = FunctionTool.from_defaults(fn=multiply)

In [ ]:

def add(a: int, b: int) -> int:
    """Add two integers and returns the result integer"""
    return a + b


add_tool = FunctionTool.from_defaults(fn=add)

def add(a: int, b: int) -> int: """Add two integers and returns the result integer""" return a + b add_tool = FunctionTool.from_defaults(fn=add)

Make sure to set your AWS credentials, either the profile_name or the keys below.

In [ ]:

llm = BedrockConverse(
    model="anthropic.claude-3-haiku-20240307-v1:0",
    # NOTE replace with your own AWS credentials
    aws_access_key_id="AWS Access Key ID to use",
    aws_secret_access_key="AWS Secret Access Key to use",
    aws_session_token="AWS Session Token to use",
    region_name="AWS Region to use, eg. us-east-1",
)

llm = BedrockConverse( model="anthropic.claude-3-haiku-20240307-v1:0", # NOTE replace with your own AWS credentials aws_access_key_id="AWS Access Key ID to use", aws_secret_access_key="AWS Secret Access Key to use", aws_session_token="AWS Session Token to use", region_name="AWS Region to use, eg. us-east-1", )

Initialize AWS Bedrock Converse Agent¶

Here we initialize a simple AWS Bedrock Converse agent with calculator functions.

In [ ]:

from llama_index.core.agent import FunctionCallingAgent

agent = FunctionCallingAgent.from_tools(
    [multiply_tool, add_tool],
    llm=llm,
    verbose=True,
    allow_parallel_tool_calls=False,
)

from llama_index.core.agent import FunctionCallingAgent agent = FunctionCallingAgent.from_tools( [multiply_tool, add_tool], llm=llm, verbose=True, allow_parallel_tool_calls=False, )

Chat¶

In [ ]:

response = agent.chat("What is (121 + 2) * 5?")

response = agent.chat("What is (121 + 2) * 5?")

In [ ]:

# inspect sources
print(str(response))
print(response.sources)

# inspect sources print(str(response)) print(response.sources)

AWS Bedrock Converse Agent over RAG Pipeline¶

Build an AWS Bedrock Converse agent over a simple 10K document. We use both HuggingFace embeddings and BAAI/bge-small-en-v1.5 to construct the RAG pipeline, and pass it to the AWS Bedrock Converse agent as a tool.

In [ ]:

!mkdir -p 'data/10k/'
!curl -o 'data/10k/uber_2021.pdf' 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/10k/uber_2021.pdf'

!mkdir -p 'data/10k/' !curl -o 'data/10k/uber_2021.pdf' 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/10k/uber_2021.pdf'

In [ ]:

from llama_index.core.tools import QueryEngineTool, ToolMetadata
from llama_index.core import SimpleDirectoryReader, VectorStoreIndex
from llama_index.embeddings.huggingface import HuggingFaceEmbedding
from llama_index.llms.bedrock_converse import BedrockConverse

embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-small-en-v1.5")
query_llm = BedrockConverse(
    model="anthropic.claude-3-haiku-20240307-v1:0",
    # NOTE replace with your own AWS credentials
    aws_access_key_id="AWS Access Key ID to use",
    aws_secret_access_key="AWS Secret Access Key to use",
    aws_session_token="AWS Session Token to use",
    region_name="AWS Region to use, eg. us-east-1",
)

# load data
uber_docs = SimpleDirectoryReader(
    input_files=["./data/10k/uber_2021.pdf"]
).load_data()
# build index
uber_index = VectorStoreIndex.from_documents(
    uber_docs, embed_model=embed_model
)
uber_engine = uber_index.as_query_engine(similarity_top_k=3, llm=query_llm)
query_engine_tool = QueryEngineTool(
    query_engine=uber_engine,
    metadata=ToolMetadata(
        name="uber_10k",
        description=(
            "Provides information about Uber financials for year 2021. "
            "Use a detailed plain text question as input to the tool."
        ),
    ),
)

from llama_index.core.tools import QueryEngineTool, ToolMetadata from llama_index.core import SimpleDirectoryReader, VectorStoreIndex from llama_index.embeddings.huggingface import HuggingFaceEmbedding from llama_index.llms.bedrock_converse import BedrockConverse embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-small-en-v1.5") query_llm = BedrockConverse( model="anthropic.claude-3-haiku-20240307-v1:0", # NOTE replace with your own AWS credentials aws_access_key_id="AWS Access Key ID to use", aws_secret_access_key="AWS Secret Access Key to use", aws_session_token="AWS Session Token to use", region_name="AWS Region to use, eg. us-east-1", ) # load data uber_docs = SimpleDirectoryReader( input_files=["./data/10k/uber_2021.pdf"] ).load_data() # build index uber_index = VectorStoreIndex.from_documents( uber_docs, embed_model=embed_model ) uber_engine = uber_index.as_query_engine(similarity_top_k=3, llm=query_llm) query_engine_tool = QueryEngineTool( query_engine=uber_engine, metadata=ToolMetadata( name="uber_10k", description=( "Provides information about Uber financials for year 2021. " "Use a detailed plain text question as input to the tool." ), ), )

In [ ]:

from llama_index.core.agent import FunctionCallingAgent

agent = FunctionCallingAgent.from_tools(
    [query_engine_tool], llm=llm, verbose=True
)

from llama_index.core.agent import FunctionCallingAgent agent = FunctionCallingAgent.from_tools( [query_engine_tool], llm=llm, verbose=True )

In [ ]:

response = agent.chat(
    "Tell me both the risk factors and tailwinds for Uber? Do two parallel tool calls."
)

response = agent.chat( "Tell me both the risk factors and tailwinds for Uber? Do two parallel tool calls." )

In [ ]:

print(str(response))

print(str(response))