本文速读：
-
自定义样例选择器
-
长度选样例择器
-
MMR样例选择器
-
n-gram重叠度样例选择器
-
相似度样例选择器
在上一篇【LangChain系列 8】Prompt模版------少样本prompt模版(二)中，介绍动态少样本prompt模版的时候，根据输入的内容，样例选择器 从所有样例中动态地选择部分样例，这部分样例与输入内容更加相关，这样语言模型能更好地理解prompt，从而给出更好的回答。
本文将介绍五种样例选择器的用法：
-
自定义样例选择器
-
长度样例选择器
-
MMR样例选择器
-
n-gram重叠度样例选择器
-
相似度样例选择器
01 自定义样例选择器
自定义样例选择器是一种常见的操作，因为业务逻辑千变万化，通过自定义样例选择器可以更加灵活地选择选择样例。
一个样例选择器至少要实现两个方法：
-
add_example方法：接收一个样例，然后把它传递给样例选择器。
-
select_examples方法：接收用户输入变量，然后返回一个样例列表给 少样本prompt模版 使用。
下面我们来实现一个随机选择两个样例的选择器。
python
from langchain.prompts.example_selector.base import BaseExampleSelector
from typing import Dict, List
import numpy as np
class CustomExampleSelector(BaseExampleSelector):
def __init__(self, examples: List[Dict[str, str]]):
self.examples = examples
def add_example(self, example: Dict[str, str]) -> None:
"""Add new example to store for a key."""
self.examples.append(example)
def select_examples(self, input_variables: Dict[str, str]) -> List[dict]:
"""Select which examples to use based on the inputs."""
return np.random.choice(self.examples, size=2, replace=False)定义好选择器后，就可以使用它了。
css
examples = [ {"foo": "1"}, {"foo": "2"}, {"foo": "3"}]
# Initialize example selector.
example_selector = CustomExampleSelector(examples)
# Select examples
example_selector.select_examples({"foo": "foo"})
# -> array([{'foo': '2'}, {'foo': '3'}], dtype=object)
# Add new example to the set of examples
example_selector.add_example({"foo": "4"})
example_selector.examples
# -> [{'foo': '1'}, {'foo': '2'}, {'foo': '3'}, {'foo': '4'}]
# Select examples
example_selector.select_examples({"foo": "foo"})
# -> array([{'foo': '1'}, {'foo': '4'}], dtype=object)02 长度样例选择器
顾名思义，长度样例选择器就是根据样例的长度来选择样例，这适用于prompt过长会超过上下文窗口长度的情况。如果用户输入内容比较长，那么就会选择更少的样例，如果用户输入内容比较短，那么就会选择更多的样例。
LengthBasedExampleSelector是LangChain提供的长度选择器。
python
from langchain.prompts import PromptTemplate
from langchain.prompts import FewShotPromptTemplate
from langchain.prompts.example_selector import LengthBasedExampleSelector
# Examples of a pretend task of creating antonyms.
examples = [
{"input": "happy", "output": "sad"},
{"input": "tall", "output": "short"},
{"input": "energetic", "output": "lethargic"},
{"input": "sunny", "output": "gloomy"},
{"input": "windy", "output": "calm"},
example_prompt = PromptTemplate(
input_variables=["input", "output"],
template="Input: {input}\nOutput: {output}",
)
example_selector = LengthBasedExampleSelector(
# The examples it has available to choose from.
examples=examples,
# The PromptTemplate being used to format the examples.
example_prompt=example_prompt,
# The maximum length that the formatted examples should be.
# Length is measured by the get_text_length function below.
max_length=25,
# The function used to get the length of a string, which is used
# to determine which examples to include. It is commented out because
# it is provided as a default value if none is specified.
# get_text_length: Callable[[str], int] = lambda x: len(re.split("\n| ", x))
)
dynamic_prompt = FewShotPromptTemplate(
# We provide an ExampleSelector instead of examples.
example_selector=example_selector,
example_prompt=example_prompt,
prefix="Give the antonym of every input",
suffix="Input: {adjective}\nOutput:",
input_variables=["adjective"],
)用户输入内容比较短时，选择了所有样例。
yaml
# An example with small input, so it selects all examples.
print(dynamic_prompt.format(adjective="big"))
Give the antonym of every input
Input: happy
Output: sad
Input: tall
Output: short
Input: energetic
Output: lethargic
Input: sunny
Output: gloomy
Input: windy
Output: calm
Input: big
Output:用户输入较长时，只选择了一个样例。
ini
# An example with long input, so it selects only one example.
long_string = "big and huge and massive and large and gigantic and tall and much much much much much bigger than everything else"
print(dynamic_prompt.format(adjective=long_string))
Give the antonym of every input
Input: happy
Output: sad
Input: big and huge and massive and large and gigantic and tall and much much much much much bigger than everything else
Output:同时，你可以动态增加样例。
yaml
# You can add an example to an example selector as well.
new_example = {"input": "big", "output": "small"}
dynamic_prompt.example_selector.add_example(new_example)
print(dynamic_prompt.format(adjective="enthusiastic"))
Give the antonym of every input
Input: happy
Output: sad
Input: tall
Output: short
Input: energetic
Output: lethargic
Input: sunny
Output: gloomy
Input: windy
Output: calm
Input: big
Output: small
Input: enthusiastic
Output:03 MMR样例选择器
MMR(maximal marginal relevance)样例选择器的意思是：选择一组样例，既保证这些样例与用户输入是相似的，同时也要保证样例的多样性。它是如何做到的呢？主要从两个方面实现的：
-
相似度：通过embeddings计算样本和用户输入余弦相似度，从样本中选择相似度高的，从而保持选择的样本和用户输入是相似的。
-
多样性：当选择一个新样本加入时，如果它与已选择的样本很相似，那么会做一个惩罚计算，从而保证了多样性。
ini
from langchain.prompts.example_selector import (
MaxMarginalRelevanceExampleSelector,
SemanticSimilarityExampleSelector,
)
from langchain.vectorstores import FAISS
from langchain.embeddings import OpenAIEmbeddings
from langchain.prompts import FewShotPromptTemplate, PromptTemplate
example_prompt = PromptTemplate(
input_variables=["input", "output"],
template="Input: {input}\nOutput: {output}",
)
# Examples of a pretend task of creating antonyms.
examples = [
{"input": "happy", "output": "sad"},
{"input": "tall", "output": "short"},
{"input": "energetic", "output": "lethargic"},
{"input": "sunny", "output": "gloomy"},
{"input": "windy", "output": "calm"},
]
example_selector = MaxMarginalRelevanceExampleSelector.from_examples(
# The list of examples available to select from.
examples,
# The embedding class used to produce embeddings which are used to measure semantic similarity.
OpenAIEmbeddings(),
# The VectorStore class that is used to store the embeddings and do a similarity search over.
FAISS,
# The number of examples to produce.
k=2,
)
mmr_prompt = FewShotPromptTemplate(
# We provide an ExampleSelector instead of examples.
example_selector=example_selector,
example_prompt=example_prompt,
prefix="Give the antonym of every input",
suffix="Input: {adjective}\nOutput:",
input_variables=["adjective"],
)
# Input is a feeling, so should select the happy/sad example as the first one
print(mmr_prompt.format(adjective="worried"))执行代码，输出结果：
yaml
Give the antonym of every input
Input: happy
Output: sad
Input: windy
Output: calm
Input: worried
Output:04 相似度样例选择器
与MMR样例选择器不同，相似度选择器仅仅根据相似度来选择样例，LangChain提供了SemanticSimilarityExampleSelector可以直接使用，继续用MMR的样例，但选择器用 相似度样例选择器 时，我们看一下选择的样例有什么不同？
ini
from langchain.prompts import PromptTemplate
from langchain.prompts.example_selector.ngram_overlap import NGramOverlapExampleSelector
from langchain.prompts import FewShotPromptTemplate, PromptTemplate
example_prompt = PromptTemplate(
input_variables=["input", "output"],
template="Input: {input}\nOutput: {output}",
)
# Examples of a pretend task of creating antonyms.
examples = [
{"input": "happy", "output": "sad"},
{"input": "tall", "output": "short"},
{"input": "energetic", "output": "lethargic"},
{"input": "sunny", "output": "gloomy"},
{"input": "windy", "output": "calm"},
]
example_prompt = PromptTemplate(
input_variables=["input", "output"],
template="Input: {input}\nOutput: {output}",
)
example_selector = NGramOverlapExampleSelector(
# The examples it has available to choose from.
examples=examples,
# The PromptTemplate being used to format the examples.
example_prompt=example_prompt,
# The threshold, at which selector stops.
# It is set to -1.0 by default.
threshold=-1.0,
# For negative threshold:
# Selector sorts examples by ngram overlap score, and excludes none.
# For threshold greater than 1.0:
# Selector excludes all examples, and returns an empty list.
# For threshold equal to 0.0:
# Selector sorts examples by ngram overlap score,
# and excludes those with no ngram overlap with input.
)
dynamic_prompt = FewShotPromptTemplate(
# We provide an ExampleSelector instead of examples.
example_selector=example_selector,
example_prompt=example_prompt,
prefix="Give the Spanish translation of every input",
suffix="Input: {sentence}\nOutput:",
input_variables=["sentence"],
)
# An example input with large ngram overlap with "Spot can run."
# and no overlap with "My dog barks."
print(dynamic_prompt.format(sentence="Spot can run fast."))执行代码，输出结果：
yaml
Give the Spanish translation of every input
Input: Spot can run.
Output: Spot puede correr.
Input: See Spot run.
Output: Ver correr a Spot.
Input: My dog barks.
Output: Mi perro ladra.
Input: Spot can run fast.
Output:由于阈值默认是-1.0，所以会选择所有样例，并排序；当然，你也可以动态增加样例：
ini
# You can add examples to NGramOverlapExampleSelector as well.
new_example = {"input": "Spot plays fetch.", "output": "Spot juega a buscar."}
example_selector.add_example(new_example)
print(dynamic_prompt.format(sentence="Spot can run fast."))把阈值设置为0时：
yaml
# You can set a threshold at which examples are excluded.
# For example, setting threshold equal to 0.0
# excludes examples with no ngram overlaps with input.
# Since "My dog barks." has no ngram overlaps with "Spot can run fast."
# it is excluded.
example_selector.threshold = 0.0
print(dynamic_prompt.format(sentence="Spot can run fast."))
Give the Spanish translation of every input
Input: Spot can run.
Output: Spot puede correr.
Input: See Spot run.
Output: Ver correr a Spot.
Input: Spot plays fetch.
Output: Spot juega a buscar.
Input: Spot can run fast.
Output:当阈值设置为0.09时：
yaml
# Setting small nonzero threshold
example_selector.threshold = 0.09
print(dynamic_prompt.format(sentence="Spot can play fetch."))
Give the Spanish translation of every input
Input: Spot can run.
Output: Spot puede correr.
Input: Spot plays fetch.
Output: Spot juega a buscar.
Input: Spot can play fetch.
Output:当阈值设置为大于1时：
ini
# Setting threshold greater than 1.0
example_selector.threshold = 1.0 + 1e-9
print(dynamic_prompt.format(sentence="Spot can play fetch."))
Give the Spanish translation of every input
Input: Spot can play fetch.
Output:此时所有样例都会被排除。
本文小结
本文主要介绍了几种样例选择器的用法和区别，在不同的业务场景，我们可以选择合适的样例选择器来提高少样本prompt的质量。