amazon personalize langchain extensionsダウンロード - amazon personalize langchain extensionsソースコードのダウンロード

Amazon Personalize Langchain 拡張機能

このリポジトリは、Langchain で動作する一連のユーティリティクラスを提供します。現在、Amazon Personalize キャンペーン/レコメンダーを操作するためのユーティリティクラスAmazonPersonalizeと、Amazon Personalize から推奨事項を取得してデフォルトのプロンプト (ユーザーがオーバーライドできる) を実行するAmazonPersonalizeChainカスタムチェーンビルドがあります。

インストール中

リポジトリのクローンを作成する

git clone https://github.com/aws-samples/amazon-personalize-langchain-extensions.git

リポジトリディレクトリに移動

 cd amazon-personalize-langchain-extensions

クラスをインストールする

pip install .

使用法

[ユースケース-1] Amazon Personalize Client をセットアップし、結果を要約するために Personalize Chain を呼び出す

 from aws_langchain import AmazonPersonalize

recommender_arn = "<insert_arn>"

client = AmazonPersonalize ( credentials_profile_name = "default" , region_name = "us-west-2" , recommender_arn = recommender_arn )
client . get_recommendations ( user_id = "1" )

[ユースケース-2] Amazon Personalize Client をセットアップし、結果を要約するために Personalize Chain を呼び出す

 from aws_langchain import AmazonPersonalize
from aws_langchain import AmazonPersonalizeChain
from langchain . llms . bedrock import Bedrock

recommender_arn = "<insert_arn>"

bedrock_llm = Bedrock ( model_id = "anthropic.claude-v2" , region_name = "us-west-2" )
client = AmazonPersonalize ( credentials_profile_name = "default" , region_name = "us-west-2" , recommender_arn = recommender_arn )
# Create personalize chain
# Use return_direct=True if you do not want summary
chain = AmazonPersonalizeChain . from_llm (
    llm = bedrock_llm , 
    client = client ,
    return_direct = False 
)
response = chain ({ 'user_id' : '1' })
print ( response )

[ユースケース-3] 独自のプロンプトを使用して Amazon Personalize Chain を呼び出す

 from langchain . prompts . prompt import PromptTemplate
from aws_langchain import AmazonPersonalize
from aws_langchain import AmazonPersonalizeChain
from langchain . llms . bedrock import Bedrock

RANDOM_PROMPT_QUERY = """
You are a skilled publicist. Write a high-converting marketing email advertising several movies available in a video-on-demand streaming platform next week, 
    given the movie and user information below. Your email will leverage the power of storytelling and persuasive language. 
    The movies to recommend and their information is contained in the <movie> tag. 
    All movies in the <movie> tag must be recommended. Give a summary of the movies and why the human should watch them. 
    Put the email between <email> tags.

    <movie>
    {result} 
    </movie>

    Assistant:
    """

RANDOM_PROMPT = PromptTemplate ( input_variables = [ "result" ], template = RANDOM_PROMPT_QUERY )


recommender_arn = "<insert_arn>"

bedrock_llm = Bedrock ( model_id = "anthropic.claude-v2" , region_name = "us-west-2" )
client = AmazonPersonalize ( credentials_profile_name = "default" , region_name = "us-west-2" , recommender_arn = recommender_arn )

chain = AmazonPersonalizeChain . from_llm ( llm = bedrock_llm , client = client , return_direct = False , prompt_template = RANDOM_PROMPT )
chain . run ({ 'user_id' : '1' , 'item_id' : '234' })

[ユースケース-4] シーケンシャルチェーンで Amazon Personalize を呼び出す

 from langchain . chains import SequentialChain
from langchain . chains import LLMChain
from aws_langchain import AmazonPersonalize
from aws_langchain import AmazonPersonalizeChain
from langchain . llms . bedrock import Bedrock
from langchain . prompts . prompt import PromptTemplate

RANDOM_PROMPT_QUERY_2 = """
You are a skilled publicist. Write a high-converting marketing email advertising several movies available in a video-on-demand streaming platform next week, 
    given the movie and user information below. Your email will leverage the power of storytelling and persuasive language. 
    You want the email to impress the user, so make it appealing to them.
    The movies to recommend and their information is contained in the <movie> tag. 
    All movies in the <movie> tag must be recommended. Give a summary of the movies and why the human should watch them. 
    Put the email between <email> tags.

    <movie>
    {result}
    </movie>

    Assistant:
    """

recommender_arn = "<insert_arn>"

bedrock_llm = Bedrock ( model_id = "anthropic.claude-v2" , region_name = "us-west-2" )
client = AmazonPersonalize ( credentials_profile_name = "default" , region_name = "us-west-2" , recommender_arn = recommender_arn )

RANDOM_PROMPT_2 = PromptTemplate ( input_variables = [ "result" ], template = RANDOM_PROMPT_QUERY_2 )
personalize_chain_instance = AmazonPersonalizeChain . from_llm ( llm = bedrock_llm , client = client , return_direct = True )
random_chain_instance = LLMChain ( llm = bedrock_llm , prompt = RANDOM_PROMPT_2 )
overall_chain = SequentialChain ( chains = [ personalize_chain_instance , random_chain_instance ], input_variables = [ "user_id" ], verbose = True )
overall_chain . run ({ 'user_id' : '1' , 'item_id' : '234' })

[ユースケース-5] レスポンスのメタデータを使用して Amazon Personalize を呼び出す

 from aws_langchain import AmazonPersonalize
from aws_langchain import AmazonPersonalizeChain
from langchain . llms . bedrock import Bedrock

recommender_arn = "<insert_arn>"
metadata_column_list = [ "METADATA_COL1" ]
metadataMap = { "ITEMS" : metadata_column_list }

bedrock_llm = Bedrock ( model_id = "anthropic.claude-v2" , region_name = "us-west-2" )
client = AmazonPersonalize ( credentials_profile_name = "default" , region_name = "us-west-2" , recommender_arn = recommender_arn )
# Create personalize chain
# Use return_direct=True if you do not want summary
chain = AmazonPersonalizeChain . from_llm (
    llm = bedrock_llm , 
    client = client ,
    return_direct = False 
)
response = chain ({ 'user_id' : '1' , 'metadata_columns' : metadataMap })
print ( response )