Home

Awesome

Autoevals

Autoevals is a tool to quickly and easily evaluate AI model outputs.

It bundles together a variety of automatic evaluation methods including:

Autoevals is developed by the team at Braintrust.

Autoevals uses model-graded evaluation for a variety of subjective tasks including fact checking, safety, and more. Many of these evaluations are adapted from OpenAI's excellent evals project but are implemented so you can flexibly run them on individual examples, tweak the prompts, and debug their outputs.

You can also create your own model-graded evaluations with Autoevals. It's easy to add custom prompts, parse outputs, and manage exceptions.

Installation

Autoevals is distributed as a Python library on PyPI and Node.js library on NPM.

Python

pip install autoevals

Node.js

npm install autoevals

Example

Use Autoevals to model-grade an example LLM completion using the factuality prompt. By default, Autoevals uses your OPENAI_API_KEY environment variable to authenticate with OpenAI's API.

Python

from autoevals.llm import *

# Create a new LLM-based evaluator
evaluator = Factuality()

# Evaluate an example LLM completion
input = "Which country has the highest population?"
output = "People's Republic of China"
expected = "China"

result = evaluator(output, expected, input=input)

# The evaluator returns a score from [0,1] and includes the raw outputs from the evaluator
print(f"Factuality score: {result.score}")
print(f"Factuality metadata: {result.metadata['rationale']}")

Custom Client

If you need to use a custom OpenAI client, you can initialize the library with a custom client.

import openai
from autoevals import init
from autoevals.oai import LLMClient

openai_client = openai.OpenAI(base_url="https://api.openai.com/v1/")

class CustomClient(LLMClient):
    openai=openai_client  # you can also pass in openai module and we will instantiate it for you
    embed = openai.embeddings.create
    moderation = openai.moderations.create
    RateLimitError = openai.RateLimitError

    def complete(self, **kwargs):
        # make adjustments as needed
        return self.openai.chat.completions.create(**kwargs)

# Autoevals will now use your custom client
client = init(client=CustomClient)

If you only need to use a custom client for a specific evaluator, you can pass in the client to the evaluator.

evaluator = Factuality(client=CustomClient)

Node.js

import { Factuality } from "autoevals";

(async () => {
  const input = "Which country has the highest population?";
  const output = "People's Republic of China";
  const expected = "China";

  const result = await Factuality({ output, expected, input });
  console.log(`Factuality score: ${result.score}`);
  console.log(`Factuality metadata: ${result.metadata.rationale}`);
})();

Using Braintrust with Autoevals

Once you grade an output using Autoevals, it's convenient to use Braintrust to log and compare your evaluation results.

Python

from autoevals.llm import *
import braintrust

# Create a new LLM-based evaluator
evaluator = Factuality()

# Set up an example LLM completion
input = "Which country has the highest population?"
output = "People's Republic of China"
expected = "China"

# Set up a BrainTrust experiment to log our eval to
experiment = braintrust.init(
    project="Autoevals", api_key="YOUR_BRAINTRUST_API_KEY"
)

# Start a span and run our evaluator
with experiment.start_span() as span:
    result = evaluator(output, expected, input=input)

    # The evaluator returns a score from [0,1] and includes the raw outputs from the evaluator
    print(f"Factuality score: {result.score}")
    print(f"Factuality metadata: {result.metadata['rationale']}")

    span.log(
        inputs={"query": input},
        output=output,
        expected=expected,
        scores={
            "factuality": result.score,
        },
        metadata={
            "factuality": result.metadata,
        },
    )

print(experiment.summarize())

Node.js

Create a file named example.eval.js (it must end with .eval.js or .eval.js):

import { Eval } from "braintrust";
import { Factuality } from "autoevals";

Eval("Autoevals", {
  data: () => [
    {
      input: "Which country has the highest population?",
      expected: "China",
    },
  ],
  task: () => "People's Republic of China",
  scores: [Factuality],
});

Then, run

npx braintrust run example.eval.js

Supported Evaluation Methods

LLM-as-a-Judge

RAG

Composite

Embeddings

Heuristic

Statistical

Custom Evaluation Prompts

Autoevals supports custom evaluation prompts for model-graded evaluation. To use them, simply pass in a prompt and scoring mechanism:

Python

from autoevals import LLMClassifier

# Define a prompt prefix for a LLMClassifier (returns just one answer)
prompt_prefix = """
You are a technical project manager who helps software engineers generate better titles for their GitHub issues.
You will look at the issue description, and pick which of two titles better describes it.

I'm going to provide you with the issue description, and two possible titles.

Issue Description: {{input}}

1: {{output}}
2: {{expected}}
"""

# Define the scoring mechanism
# 1 if the generated answer is better than the expected answer
# 0 otherwise
output_scores = {"1": 1, "2": 0}

evaluator = LLMClassifier(
    name="TitleQuality",
    prompt_template=prompt_prefix,
    choice_scores=output_scores,
    use_cot=True,
)

# Evaluate an example LLM completion
page_content = """
As suggested by Nicolo, we should standardize the error responses coming from GoTrue, postgres, and realtime (and any other/future APIs) so that it's better DX when writing a client,
We can make this change on the servers themselves, but since postgrest and gotrue are fully/partially external may be harder to change, it might be an option to transform the errors within the client libraries/supabase-js, could be messy?
Nicolo also dropped this as a reference: http://spec.openapis.org/oas/v3.0.3#openapi-specification"""
output = (
    "Standardize error responses from GoTrue, Postgres, and Realtime APIs for better DX"
)
expected = "Standardize Error Responses across APIs"

response = evaluator(output, expected, input=page_content)

print(f"Score: {response.score}")
print(f"Metadata: {response.metadata}")

Node.js

import { LLMClassifierFromTemplate } from "autoevals";

(async () => {
  const promptTemplate = `You are a technical project manager who helps software engineers generate better titles for their GitHub issues.
You will look at the issue description, and pick which of two titles better describes it.

I'm going to provide you with the issue description, and two possible titles.

Issue Description: {{input}}

1: {{output}}
2: {{expected}}`;

  const choiceScores = { 1: 1, 2: 0 };

  const evaluator =
    LLMClassifierFromTemplate <
    { input: string } >
    {
      name: "TitleQuality",
      promptTemplate,
      choiceScores,
      useCoT: true,
    };

  const input = `As suggested by Nicolo, we should standardize the error responses coming from GoTrue, postgres, and realtime (and any other/future APIs) so that it's better DX when writing a client,
We can make this change on the servers themselves, but since postgrest and gotrue are fully/partially external may be harder to change, it might be an option to transform the errors within the client libraries/supabase-js, could be messy?
Nicolo also dropped this as a reference: http://spec.openapis.org/oas/v3.0.3#openapi-specification`;
  const output = `Standardize error responses from GoTrue, Postgres, and Realtime APIs for better DX`;
  const expected = `Standardize Error Responses across APIs`;

  const response = await evaluator({ input, output, expected });

  console.log("Score", response.score);
  console.log("Metadata", response.metadata);
})();

Creating custom scorers

You can also create your own scoring functions that do not use LLMs. For example, to test whether the word 'banana' is in the output, you can use the following:

Python

from autoevals import Score


def banana_scorer(output, expected, input):
    return Score(name="banana_scorer", score=1 if "banana" in output else 0)


input = "What is 1 banana + 2 bananas?"
output = "3"
expected = "3 bananas"

result = banana_scorer(output, expected, input)

print(f"Banana score: {result.score}")

Node.js

import { Score } from "autoevals";

const bananaScorer = ({
  output,
  expected,
  input,
}: {
  output: string;
  expected: string;
  input: string;
}): Score => {
  return { name: "banana_scorer", score: output.includes("banana") ? 1 : 0 };
};

(async () => {
  const input = "What is 1 banana + 2 bananas?";
  const output = "3";
  const expected = "3 bananas";

  const result = bananaScorer({ output, expected, input });
  console.log(`Banana score: ${result.score}`);
})();

Why does this library exist?

There is nothing particularly novel about the evaluation methods in this library. They are all well-known and well-documented. However, there are a few things that are particularly difficult when evaluating in practice:

Documentation

The full docs are available here.