Custom evaluation metrics in LangChain

Introduction

Custom evaluation metrics help you measure how well your AI or language model is doing in ways that matter most to your project.

When you want to check if your AI answers are accurate for your specific topic.

When default scores don't show the full picture of your model's performance.

When you need to compare different AI models using your own rules.

When you want to track improvements based on your unique goals.

When you want to give feedback to your AI system in a way that fits your needs.

Syntax

LangChain

from langchain.evaluation import Evaluation

class MyMetric(Evaluation):
    def evaluate(self, prediction: str, reference: str) -> float:
        # Your custom logic here
        score = 0.0
        return score

Create a class that inherits from Evaluation.

Implement the evaluate method to return a numeric score.

Examples

This metric returns 1 if the prediction exactly matches the reference, otherwise 0.

LangChain

from langchain.evaluation import Evaluation

class ExactMatch(Evaluation):
    def evaluate(self, prediction: str, reference: str) -> float:
        return 1.0 if prediction == reference else 0.0

This metric scores higher when the prediction length is closer to the reference length.

LangChain

from langchain.evaluation import Evaluation

class LengthDifference(Evaluation):
    def evaluate(self, prediction: str, reference: str) -> float:
        return 1.0 / (1 + abs(len(prediction) - len(reference)))

Sample Program

This example defines a simple similarity metric that compares how many words overlap between prediction and reference. It then prints the similarity score.

LangChain

from langchain.evaluation import Evaluation

class SimpleSimilarity(Evaluation):
    def evaluate(self, prediction: str, reference: str) -> float:
        pred_words = set(prediction.lower().split())
        ref_words = set(reference.lower().split())
        common = pred_words.intersection(ref_words)
        total = pred_words.union(ref_words)
        return len(common) / len(total) if total else 0.0

# Example usage
metric = SimpleSimilarity()
pred = "The quick brown fox"
ref = "The quick fox jumps"
score = metric.evaluate(pred, ref)
print(f"Similarity score: {score:.2f}")

Important Notes

Custom metrics should return a number, usually between 0 and 1, where higher means better.

Keep your metric logic simple and fast for better performance.

Test your metric with different inputs to make sure it behaves as expected.

Summary

Custom evaluation metrics let you measure AI results in your own way.

Define a class inheriting from Evaluation and implement evaluate.

Use your metric to get scores that help improve your AI models.

Practice

1. What is the main purpose of creating a custom evaluation metric in Langchain?

easy

A. To speed up the AI model training process

B. To measure AI results in a way that fits your specific needs

C. To automatically fix errors in AI outputs

D. To replace the AI model with a simpler one

Solution

1. Step 1: Understand the role of evaluation metrics

   Evaluation metrics measure how well an AI model performs its task.

2. Step 2: Identify why custom metrics are used

   Custom metrics let you measure results in ways that standard metrics might not cover, fitting your unique needs.

3. Final Answer:

   To measure AI results in a way that fits your specific needs -> Option B

4. Quick Check:

   Custom metrics = tailored measurement [OK]

Hint: Custom metrics tailor scoring to your AI task [OK]

Common Mistakes:

• Thinking custom metrics speed training
• Believing they fix AI errors automatically
• Confusing metrics with model replacement

2. Which of the following is the correct way to start defining a custom evaluation metric class in Langchain?

easy

A. class MyMetric(Evaluation):

B. def MyMetric():

C. class MyMetric():

D. function MyMetric extends Evaluation {}

Solution

1. Step 1: Recall Langchain class inheritance syntax

   Custom metrics inherit from the Evaluation base class using Python class syntax.

2. Step 2: Identify correct class definition

   class MyMetric(Evaluation): correctly defines a class inheriting from Evaluation, matching Langchain patterns.

3. Final Answer:

   class MyMetric(Evaluation): -> Option A

4. Quick Check:

   Class inherits Evaluation = correct syntax [OK]

Hint: Use class inheritance with Evaluation base [OK]

Common Mistakes:

• Defining a function instead of a class
• Missing inheritance from Evaluation
• Using JavaScript syntax in Python

3. Given this custom metric class, what will metric.evaluate(['hello'], ['hello']) return?

class ExactMatch(Evaluation):
    def evaluate(self, predictions, references):
        return sum(p == r for p, r in zip(predictions, references)) / len(references)

medium

A. 1.0

B. 0.0

C. Error due to missing method

D. None

Solution

1. Step 1: Understand the evaluate method logic

   It compares each prediction to the reference and counts matches, then divides by total references.

2. Step 2: Apply inputs to the method

   With predictions=['hello'] and references=['hello'], the single pair matches, so sum is 1 and length is 1, result is 1/1 = 1.0.

3. Final Answer:

   1.0 -> Option A

4. Quick Check:

   Exact match count / total = 1.0 [OK]

Hint: Check if predictions equal references, then divide [OK]

Common Mistakes:

• Forgetting to divide by length
• Confusing sum with boolean values
• Expecting method to return a list

4. What is wrong with this custom metric class that causes an error?

class LengthDiff(Evaluation):
    def evaluate(self, predictions, references):
        return abs(len(predictions) - len(references)) / len(references)

medium

A. It returns a number instead of a score between 0 and 1

B. It does not implement the evaluate method

C. It uses abs() incorrectly causing a syntax error

D. It does not handle empty lists causing runtime error

Solution

1. Step 1: Analyze the evaluate method with empty references

   If references=[], len(references)=0 causes ZeroDivisionError in the division.

2. Step 2: Identify the runtime error cause

   The code divides by len(references) without checking if references is empty, causing runtime error.

3. Final Answer:

   It does not handle empty lists causing runtime error -> Option D

4. Quick Check:

   len(references)==0 -> ZeroDivisionError [OK]

Hint: Check how method handles empty input lists [OK]

Common Mistakes:

• Assuming abs() causes syntax error
• Thinking evaluate method is missing
• Ignoring empty list edge cases

5. You want to create a custom metric that scores AI answers higher if they contain more keywords from a reference list. Which approach fits best?

hard

A. Calculate the difference in length between prediction and reference

B. Check if prediction exactly matches the reference string

C. Count how many keywords appear in the prediction, divide by total keywords

D. Return a fixed score regardless of prediction content

Solution

1. Step 1: Understand the goal of keyword-based scoring

   The metric should reward predictions containing more keywords from the reference list.

2. Step 2: Identify the approach that measures keyword presence proportionally

   Counting keywords in prediction and dividing by total keywords gives a score reflecting keyword coverage.

3. Final Answer:

   Count how many keywords appear in the prediction, divide by total keywords -> Option C

4. Quick Check:

   Keyword coverage scoring = Count how many keywords appear in the prediction, divide by total keywords [OK]

Hint: Score by keyword matches divided by total keywords [OK]

Common Mistakes:

• Using exact match instead of keyword count
• Measuring length difference unrelated to keywords
• Returning fixed scores ignoring content