如何用Python创建自定义示例选择器：从零开始的实用指南

本文链接：https://blog.csdn.net/mmlihaio/article/details/142534780

# 引言

在自然语言处理任务中，选择合适的示例来帮助模型理解输入至关重要。示例选择器（Example Selector）正是为此而生的。本文将详细介绍如何创建自定义示例选择器，并通过实例展示其使用方法。

# 主要内容

## 示例选择器的基本接口

示例选择器的基类接口定义如下：

```python
from abc import ABC, abstractmethod
from typing import List, Dict, Any

class BaseExampleSelector(ABC):
    """Interface for selecting examples to include in prompts."""

    @abstractmethod
    def select_examples(self, input_variables: Dict[str, str]) -> List[dict]:
        """Select which examples to use based on the inputs."""
        
    @abstractmethod
    def add_example(self, example: Dict[str, str]) -> Any:
        """Add new example to store."""

不同的实现可以根据需要自定义select_examples和add_example方法。

创建自定义示例选择器

下面，我们将实现一个依据输入词长度选择示例的选择器：

from langchain_core.example_selectors.base import BaseExampleSelector

class CustomExampleSelector(BaseExampleSelector):
    def __init__(self, examples):
        self.examples = examples

    def add_example(self, example):
        self.examples.append(example)

    def select_examples(self, input_variables):
        new_word = input_variables["input"]
        new_word_length = len(new_word)

        best_match = None
        smallest_diff = float("inf")

        for example in self.examples:
            current_diff = abs(len(example["input"]) - new_word_length)

            if current_diff < smallest_diff:
                smallest_diff = current_diff
                best_match = example

        return [best_match]

示例列表

我们需要创建一些示例用于测试：

examples = [
    {"input": "hi", "output": "ciao"},
    {"input": "bye", "output": "arrivederci"},
    {"input": "soccer", "output": "calcio"},
]

使用示例选择器

example_selector = CustomExampleSelector(examples)
result = example_selector.select_examples({"input": "okay"})
print(result)  # 输出: [{'input': 'bye', 'output': 'arrivederci'}]

example_selector.add_example({"input": "hand", "output": "mano"})
result = example_selector.select_examples({"input": "okay"})
print(result)  # 输出: [{'input': 'hand', 'output': 'mano'}]

代码示例

将示例选择器与提示模板结合使用：

from langchain_core.prompts.few_shot import FewShotPromptTemplate
from langchain_core.prompts.prompt import PromptTemplate

example_prompt = PromptTemplate.from_template("Input: {input} -> Output: {output}")

prompt = FewShotPromptTemplate(
    example_selector=example_selector,
    example_prompt=example_prompt,
    suffix="Input: {input} -> Output:",
    prefix="Translate the following words from English to Italian:",
    input_variables=["input"],
)

print(prompt.format(input="word"))
# 输出:
# Translate the following words from English to Italian:
#
# Input: hand -> Output: mano
#
# Input: word -> Output: