OpenAI functions agent in LangChain - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the OpenAI functions agent from LangChain.

LangChain

from langchain_community.agents import [1]

Drag options to blanks, or click blank then click option'

AAgentExecutor

BOpenAIChatAgent

COpenAIFunctionsAgent

DTool

Attempts:

3 left

💡 Hint

Common Mistakes

Importing OpenAIChatAgent instead of OpenAIFunctionsAgent

Using AgentExecutor which is a different class

Importing Tool which is unrelated here

The OpenAIFunctionsAgent is the correct import to create an agent that uses OpenAI functions in LangChain.

2fill in blank

medium

Complete the code to create an OpenAI chat model with temperature 0.

LangChain

from langchain_openai import ChatOpenAI
chat = ChatOpenAI(temperature=[1])

Drag options to blanks, or click blank then click option'

A0

B1.0

C0.5

D-1

Attempts:

3 left

💡 Hint

Common Mistakes

Using temperature 1.0 which is very random

Using negative temperature which is invalid

Using 0.5 which is moderate randomness

Setting temperature=0 makes the model deterministic, which is common for function calling agents.

3fill in blank

hard

Fix the error in the code to create an OpenAI functions agent with the chat model and tools.

LangChain

agent = OpenAIFunctionsAgent.from_llm_and_tools([1], tools)

Drag options to blanks, or click blank then click option'

Allm

Bmodel

Copenai

Dchat

Attempts:

3 left

💡 Hint

Common Mistakes

Passing 'llm' which is undefined here

Passing 'model' or 'openai' which are not defined variables

The variable 'chat' is the ChatOpenAI instance passed to the agent constructor.

4fill in blank

hard

Fill both blanks to create a list of tools and initialize the agent executor.

LangChain

tools = [[1]]
agent_executor = AgentExecutor(agent=agent, tools=[2])

Drag options to blanks, or click blank then click option'

Atool

Btools

Cagent

Dexecutor

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'tools' inside the list instead of 'tool'

Passing 'agent' instead of 'tools' to AgentExecutor

The tools list contains 'tool' and the AgentExecutor uses the 'tools' list as argument.

5fill in blank

hard

Fill all three blanks to run the agent executor with input and print the output.

LangChain

input_text = [1]
output = agent_executor.run([2])
print([3])

Drag options to blanks, or click blank then click option'

A"What is the weather today?"

Binput_text

Coutput

D"Hello world"

Attempts:

3 left

💡 Hint

Common Mistakes

Passing a variable name instead of a string literal for input_text

Printing input_text instead of output

Passing output instead of input_text to run()

We assign a string to input_text, run the agent with input_text, and print the output variable.