EnumOutputParser

• Author: ranian963

• Design:

• Peer Review : JaeHo Kim

• This is a part of LangChain Open Tutorial

Overview

In this tutorial, we introduce how to use EnumOutputParser to extract valid Enum values from the output of an LLM.

EnumOutputParser is a tool that parses the output of a language model into one of the predefined enumeration (Enum) values , offering the following features:

• Enumeration Parsing: Converts the string output into a predefined Enum value.

• Type Safety: Ensures that the parsed result is always one of the defined Enum values.

• Flexibility: Automatically handles spaces and line breaks.

Table of Contents

• Overview

• Environment Setup

• Introduction to EnumOutputParser

• Example: Colors Enum Parser

References

• LangChain: Output Parsers

• LangChain: EnumOutputParser

• Enum in Python

• LangChain Open Tutorial

---

Environment Setup

Set up the environment. You may refer to Environment Setup for more details.

[Note]

• langchain-opentutorial is a package that provides a set of easy-to-use environment setup, useful functions and utilities for tutorials.

• You can checkout the langchain-opentutorial for more details.

%%capture --no-stderr
%pip install langchain-opentutorial

# Installing required libraries
from langchain_opentutorial import package

package.install(
    [
        "langsmith",
        "langchain",
        "langchain_openai"
    ],
    verbose=False,
    upgrade=False,
)

# Set environment variables
from langchain_opentutorial import set_env

set_env(
    {
        "OPENAI_API_KEY": "",
        "LANGCHAIN_API_KEY": "",
        "LANGCHAIN_TRACING_V2": "true",
        "LANGCHAIN_ENDPOINT": "https://api.smith.langchain.com",
        "LANGCHAIN_PROJECT": "07-EnumOutputParser",
    }
)

You can alternatively set OPENAI_API_KEY in .env file and load it.

[Note] This is not necessary if you've already set OPENAI_API_KEY in previous steps.

from dotenv import load_dotenv

load_dotenv(override=True)

Introduction to EnumOutputParser

EnumOutputParser is a tool that strictly parses an LLM's output into a defined enumeration (Enum). This ensures that the model output is always one of the enumerated values.

Use cases

• When you only want one valid choice from a set of possibilities.

• When you want to avoid typos and variations by using a clear Enum value.

In the following example, we define an Colors Enum and make the LLM return one of red/green/blue by parsing the output.

Example: Colors Enum Parser

The code below shows how to define the Colors(Enum) class and wrap it with EnumOutputParser, then integrate it into a prompt chain. Once the chain is executed, the LLM response is strictly parsed into one of the values in Colors.

# Import EnumOutputParser
from langchain.output_parsers.enum import EnumOutputParser

Define the Colors enumeration using the Enum class from Python's built-in enum module.

from enum import Enum

class Colors(Enum):
    RED = "Red"
    GREEN = "Green"
    BLUE = "Blue"

Now we create an EnumOutputParser object for parsing strings into the Colors enumeration.

# Instantiate EnumOutputParser
parser = EnumOutputParser(enum=Colors)

# You can view the format instructions that the parser expects.
print(parser.get_format_instructions())

Below is an example that constructs a simple chain using PromptTemplate and ChatOpenAI. If the LLM responds about which color the object "sky" is, the parser will convert that string into a Colors value.

from langchain_core.prompts import PromptTemplate
from langchain_openai import ChatOpenAI

# Prompt template: the parser's format instructions are added at the end.
prompt = (
    PromptTemplate.from_template(
        """Which color is this object?

Object: {object}

Instructions: {instructions}"""
    ).partial(instructions=parser.get_format_instructions())
)

# Entire chain: (prompt) -> (LLM) -> (Enum Parser)
chain = prompt | ChatOpenAI(temperature=0) | parser

Now let's run the chain.

response = chain.invoke({"object": "sky"})
print("Parsed Enum:", response)
print("Raw Enum Value:", response.value)