📌OpenAI Function Calling with LCEL (+pydantic)

• We use Pydantic to create schemas & create the openai functions descriptions
  • OpenAI Function Calling - general idea
  • we’ve used just big blobs of json

setup

import os
import openai
 
from dotenv import load_dotenv, find_dotenv
_ = load_dotenv(find_dotenv()) # read local .env file
openai.api_key = os.environ['OPENAI_API_KEY']
 
from typing import List
from pydantic import BaseModel, Field

pydantic & openai function definition

• Pydantic syntax
• OpenAI Function Calling - general idea
  • rn we’re just repeating what we did previously (making the functions for openai) but just with pydantic → just creating the schema

class WeatherSearch(BaseModel):
    """Call this with an airport code to get the weather at that airport"""
    airport_code: str = Field(description="airport code to get weather for")
    
from langchain.utils.openai_functions import convert_pydantic_to_openai_function
weather_function = convert_pydantic_to_openai_function(WeatherSearch)

{'name': 'WeatherSearch',
 'description': 'Call this with an airport code to get the weather at that airport',
 'parameters': {'title': 'WeatherSearch',
  'description': 'Call this with an airport code to get the weather at that airport',
  'type': 'object',
  'properties': {'airport_code': {'title': 'Airport Code',
    'description': 'airport code to get weather for',
    'type': 'string'}},
  'required': ['airport_code']}}

• you need a function description string """""" or else u cannot convert
  • only the function description is mandatory

with LCEL

from langchain_chat_models import ChatOpenAI
model = ChatOpenAI()

diff ways

• we can pass in the functions as arguments

model.invoke("what is the weather in SF today?", functions=[weather_function])

AIMessage(content='', additional_kwargs={'function_call': {'name': 'WeatherSearch', 'arguments': '{"airport_code":"SFO"}'}})

• we can bind the function invocation to the model
  • Bind
  • u also have to reassign to a new variable because Runnables are immutable

model = model.bind(functions=[weather_function])
model.invoke("what is the weather in SF today?")

AIMessage(content='', additional_kwargs={'function_call': {'name': 'WeatherSearch', 'arguments': '{"airport_code":"SFO"}'}})

we can still force the model to use a function

• OpenAI Function Calling - general idea → function call options
  • auto, none, specific function (ex. function_call={"name": "get_current_weather"})

model_with_forced_function = model.bind(
	functions=[weather_function], 
	function_call={"name":"WeatherSearch"}
)
model_with_forced_function.invoke("hi!")

AIMessage(content='', additional_kwargs={'function_call': {'name': 'WeatherSearch', 'arguments': '{"airport_code":"SFO"}'}})

• still calls the weather search

in a chain

from langchain.prompts import ChatPromptTemplate
 
prompt = ChatPromptTemplate.from_messages([
    ("system", "You are a helpful assistant"),
    ("user", "{input}")
])
chain = prompt | model_with_function
chain.invoke({"input": "what is the weather in sf?"})

multiple functions

• Even better, we can pass a set of function and let the LLM decide which to use based on the question context

class ArtistSearch(BaseModel):
    """Call this to get the names of songs by a particular artist"""
    artist_name: str = Field(description="name of artist to look up")
    n: int = Field(description="number of results")
class WeatherSearch(BaseModel):
    """Call this with an airport code to get the weather at that airport"""
    airport_code: str = Field(description="airport code to get weather for")
functions = [
    convert_pydantic_to_openai_function(WeatherSearch),
    convert_pydantic_to_openai_function(ArtistSearch),
]
model_with_functions = model.bind(functions=functions)
model_with_functions.invoke("what is the weather in sf?")
# AIMessage(content='', additional_kwargs={'function_call': {'name': 'WeatherSearch', 'arguments': '{"airport_code":"SFO"}'}})
model_with_functions.invoke("what are three songs by taylor swift?")
# AIMessage(content='', additional_kwargs={'function_call': {'name': 'ArtistSearch', 'arguments': '{"artist_name":"Taylor Swift","n":3}'}})
model_with_functions.invoke("hi!")
# AIMessage(content='Hello! How can I assist you today?')