如何定义自定义评估器
关键概念
自定义评估器只是接收一个数据集示例和应用程序输出结果,并返回一个或多个指标的函数。 这些函数可以直接传递给 evaluate() / aevaluate()。
基础示例
- Python
- TypeScript
需要 langsmith>=0.2.0
from langsmith import evaluate
def correct(outputs: dict, reference_outputs: dict) -> bool:
"""Check if the answer exactly matches the expected answer."""
return outputs["answer"] == reference_outputs["answer"]
def dummy_app(inputs: dict) -> dict:
return {"answer": "hmm i'm not sure", "reasoning": "i didn't understand the question"}
results = evaluate(
dummy_app,
data="dataset_name",
evaluators=[correct]
)
需要 langsmith>=0.2.9
import type { EvaluationResult } from "langsmith/evaluation";
const correct = async ({ outputs, referenceOutputs }: {
outputs: Record<string, any>;
referenceOutputs?: Record<string, any>;
}): Promise<EvaluationResult> => {
const score = outputs?.answer === referenceOutputs?.answer;
return { key: "correct", score };
}
评估器参数
自定义评估器函数必须具有特定的参数名称。它们可以接受以下参数的任意子集:
run: Run: 应用程序在给定示例上生成的完整 Run 对象。example: Example: 完整数据集 示例,包括示例输入、输出(如果可用)和元数据(如果可用)。inputs: dict: 对应数据集中单个示例的输入字典。outputs: dict: 应用程序在给定inputs上生成的输出字典。reference_outputs/referenceOutputs: dict: 与示例相关联的参考输出的字典(如果可用)。
对于大多数用例,您只需要 inputs、outputs 和 reference_outputs。run 和 example 仅在实际应用的输入和输出之外需要额外的跟踪或示例元数据时才有用。
当使用 JS/TS 时,这些参数应作为单个对象参数的一部分传入。
评估器输出
自定义评估器应返回以下类型之一:
Python 和 JS/TS
dict:{"score" | "value": ..., "key": ...}形式的字典允许您自定义指标类型(数值型使用 \"score\",分类型使用 \"value\")和指标名称。如果您想将整数记录为分类指标,这将非常有用。
仅限 Python
int | float | bool: 这被解释为一个可以平均、排序等的连续指标。函数名称用作该指标的名称。str: 这被解释为分类指标。函数名称用作该指标的名称。list[dict]: 使用单个函数返回多个指标。
附加示例
- Python
- TypeScript
需要 langsmith>=0.2.0
from langsmith import evaluate, wrappers
from langsmith.schemas import Run, Example
from openai import AsyncOpenAI
# Assumes you've installed pydantic.
from pydantic import BaseModel
# We can still pass in Run and Example objects if we'd like
def correct_old_signature(run: Run, example: Example) -> dict:
"""Check if the answer exactly matches the expected answer."""
return {"key": "correct", "score": run.outputs["answer"] == example.outputs["answer"]}
# Just evaluate actual outputs
def concision(outputs: dict) -> int:
"""Score how concise the answer is. 1 is the most concise, 5 is the least concise."""
return min(len(outputs["answer"]) // 1000, 4) + 1
# Use an LLM-as-a-judge
oai_client = wrappers.wrap_openai(AsyncOpenAI())
async def valid_reasoning(inputs: dict, outputs: dict) -> bool:
"""Use an LLM to judge if the reasoning and the answer are consistent."""
instructions = """\
Given the following question, answer, and reasoning, determine if the reasoning for the \
answer is logically valid and consistent with question and the answer."""
class Response(BaseModel):
reasoning_is_valid: bool
msg = f"Question: {inputs['question']}\nAnswer: {outputs['answer']}\nReasoning: {outputs['reasoning']}"
response = await oai_client.beta.chat.completions.parse(
model="gpt-4o-mini",
messages=[{"role": "system", "content": instructions,}, {"role": "user", "content": msg}],
response_format=Response
)
return response.choices[0].message.parsed.reasoning_is_valid
def dummy_app(inputs: dict) -> dict:
return {"answer": "hmm i'm not sure", "reasoning": "i didn't understand the question"}
results = evaluate(
dummy_app,
data="dataset_name",
evaluators=[correct_old_signature, concision, valid_reasoning]
)
import { Client } from "langsmith";
import { evaluate } from "langsmith/evaluation";
import { Run, Example } from "langsmith/schemas";
import OpenAI from "openai";
// Type definitions
interface AppInputs {
question: string;
}
interface AppOutputs {
answer: string;
reasoning: string;
}
interface Response {
reasoning_is_valid: boolean;
}
// Old signature evaluator
function correctOldSignature(run: Run, example: Example) {
return {
key: "correct",
score: run.outputs?.["answer"] === example.outputs?.["answer"],
};
}
// Output-only evaluator
function concision({ outputs }: { outputs: AppOutputs }) {
return {
key: "concision",
score: Math.min(Math.floor(outputs.answer.length / 1000), 4) + 1,
};
}
// LLM-as-judge evaluator
const openai = new OpenAI();
async function validReasoning({
inputs,
outputs
}: {
inputs: AppInputs;
outputs: AppOutputs;
}) {
const instructions = ` Given the following question, answer, and reasoning, determine if the reasoning for the answer is logically valid and consistent with question and the answer.`;
const msg = `Question: ${inputs.question}
Answer: ${outputs.answer}\nReasoning: ${outputs.reasoning}`;
const response = await openai.chat.completions.create({
model: "gpt-4",
messages: [
{ role: "system", content: instructions },
{ role: "user", content: msg }
],
response_format: { type: "json_object" },
functions: [{
name: "parse_response",
parameters: {
type: "object",
properties: {
reasoning_is_valid: {
type: "boolean",
description: "Whether the reasoning is valid"
}
},
required: ["reasoning_is_valid"]
}
}]
});
const parsed = JSON.parse(response.choices[0].message.content ?? "{}") as Response;
return {
key: "valid_reasoning",
score: parsed.reasoning_is_valid ? 1 : 0
};
}
// Example application
function dummyApp(inputs: AppInputs): AppOutputs {
return {
answer: "hmm i'm not sure",
reasoning: "i didn't understand the question"
};
}
const results = await evaluate(dummyApp, {
data: "dataset_name",
evaluators: [correctOldSignature, concision, validReasoning],
client: new Client()
});
相关
- 评估聚合实验结果: 定义汇总评估器,用于计算整个实验的指标。
- 运行评估以比较两个实验: 定义成对评估器,通过比较两个(或更多)实验相互计算指标。