Added a rag template for Kendra (#12470)

## Description
Adds a rag template for Amazon Kendra with Bedrock.

---------

Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
pull/12492/head
Piyush Jain 7 months ago committed by GitHub
parent eb903e211c
commit 689853902e
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

@ -1,4 +1,4 @@
# This file is automatically @generated by Poetry 1.5.1 and should not be changed by hand.
# This file is automatically @generated by Poetry 1.6.1 and should not be changed by hand.
[[package]]
name = "aiohttp"
@ -839,28 +839,6 @@ files = [
{file = "numpy-1.24.4.tar.gz", hash = "sha256:80f5e3a4e498641401868df4208b74581206afbee7cf7b8329daae82676d9463"},
]
[[package]]
name = "openai"
version = "0.28.1"
description = "Python client library for the OpenAI API"
optional = false
python-versions = ">=3.7.1"
files = [
{file = "openai-0.28.1-py3-none-any.whl", hash = "sha256:d18690f9e3d31eedb66b57b88c2165d760b24ea0a01f150dd3f068155088ce68"},
{file = "openai-0.28.1.tar.gz", hash = "sha256:4be1dad329a65b4ce1a660fe6d5431b438f429b5855c883435f0f7fcb6d2dcc8"},
]
[package.dependencies]
aiohttp = "*"
requests = ">=2.20"
tqdm = "*"
[package.extras]
datalib = ["numpy", "openpyxl (>=3.0.7)", "pandas (>=1.2.3)", "pandas-stubs (>=1.1.0.11)"]
dev = ["black (>=21.6b0,<22.0)", "pytest (==6.*)", "pytest-asyncio", "pytest-mock"]
embeddings = ["matplotlib", "numpy", "openpyxl (>=3.0.7)", "pandas (>=1.2.3)", "pandas-stubs (>=1.1.0.11)", "plotly", "scikit-learn (>=1.0.2)", "scipy", "tenacity (>=8.0.1)"]
wandb = ["numpy", "openpyxl (>=3.0.7)", "pandas (>=1.2.3)", "pandas-stubs (>=1.1.0.11)", "wandb"]
[[package]]
name = "packaging"
version = "23.2"
@ -1323,7 +1301,7 @@ files = [
]
[package.dependencies]
greenlet = {version = "!=0.4.17", markers = "platform_machine == \"win32\" or platform_machine == \"WIN32\" or platform_machine == \"AMD64\" or platform_machine == \"amd64\" or platform_machine == \"x86_64\" or platform_machine == \"ppc64le\" or platform_machine == \"aarch64\""}
greenlet = {version = "!=0.4.17", markers = "platform_machine == \"aarch64\" or platform_machine == \"ppc64le\" or platform_machine == \"x86_64\" or platform_machine == \"amd64\" or platform_machine == \"AMD64\" or platform_machine == \"win32\" or platform_machine == \"WIN32\""}
typing-extensions = ">=4.2.0"
[package.extras]
@ -1409,26 +1387,6 @@ requests = ">=2.26.0"
[package.extras]
blobfile = ["blobfile (>=2)"]
[[package]]
name = "tqdm"
version = "4.66.1"
description = "Fast, Extensible Progress Meter"
optional = false
python-versions = ">=3.7"
files = [
{file = "tqdm-4.66.1-py3-none-any.whl", hash = "sha256:d302b3c5b53d47bce91fea46679d9c3c6508cf6332229aa1e7d8653723793386"},
{file = "tqdm-4.66.1.tar.gz", hash = "sha256:d88e651f9db8d8551a62556d3cff9e3034274ca5d66e93197cf2490e2dcb69c7"},
]
[package.dependencies]
colorama = {version = "*", markers = "platform_system == \"Windows\""}
[package.extras]
dev = ["pytest (>=6)", "pytest-cov", "pytest-timeout", "pytest-xdist"]
notebook = ["ipywidgets (>=6)"]
slack = ["slack-sdk"]
telegram = ["requests"]
[[package]]
name = "typing-extensions"
version = "4.8.0"
@ -1578,4 +1536,4 @@ multidict = ">=4.0"
[metadata]
lock-version = "2.0"
python-versions = ">=3.8.1,<4.0"
content-hash = "d6fa65710b7733cde1cc8215452c856efcbfeb174481b97009239f8e6c5ed5c6"
content-hash = "0bb560b6b7caa0c0d7941ccdeac8d1e80e3fae6a6dade815f7ed442c0e5108ed"

@ -8,7 +8,6 @@ readme = "README.md"
[tool.poetry.dependencies]
python = ">=3.8.1,<4.0"
langchain = ">=0.0.313, <0.1"
openai = ">=0.28.1"
tiktoken = ">=0.5.1"
faiss-cpu = ">=1.7.4"
boto3 = ">=1.28.57"

@ -0,0 +1,21 @@
MIT License
Copyright (c) 2023 LangChain, Inc.
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

@ -0,0 +1,21 @@
# RAG AWS Kendra
[Amazon Kendra](https://aws.amazon.com/kendra/) is an intelligent search service powered by machine learning (ML).
Here we will use `Anthropic Claude` for text generation and `Amazon Kendra` for retrieving documents. Together, with these two services, this application uses a Retrieval chain to answer questions from your documents.
(See [this page](https://aws.amazon.com/blogs/machine-learning/quickly-build-high-accuracy-generative-ai-applications-on-enterprise-data-using-amazon-kendra-langchain-and-large-language-models/) for additional context on building RAG applications with Amazon Kendra.)
Code here uses the `boto3` library to connect with the Bedrock service. See [this page](https://boto3.amazonaws.com/v1/documentation/api/latest/guide/quickstart.html#configuration) for setting up and configuring boto3 to work with an AWS account.
## Kendra Index
You will need a Kendra Index setup before using this template. For setting up a sample index, you can use this [Cloudformation template](https://github.com/aws-samples/amazon-kendra-langchain-extensions/blob/main/kendra_retriever_samples/kendra-docs-index.yaml) to create the index. This template includes sample data containing AWS online documentation for Amazon Kendra, Amazon Lex, and Amazon SageMaker. Alternately, if you have an Amazon Kendra index and have indexed your own dataset, you can use that. Launching the stack requires about 30 minutes followed by about 15 minutes to synchronize it and ingest the data in the index. Therefore, wait for about 45 minutes after launching the stack. Note the Index ID and AWS Region on the stacks Outputs tab.
## Environment variables
The code assumes that you are working with the `default` AWS profile and `us-east-1` region. If not, specify these environment variables to reflect the correct region and AWS profile.
* `AWS_DEFAULT_REGION`
* `AWS_PROFILE`
This code also requires specifying the `KENDRA_INDEX_ID` env variable which should have the Index ID of the Kendra index. Note that the Index ID is a 36 character alphanumeric value that can be found in the index detail page.

@ -0,0 +1,6 @@
from rag_aws_kendra.chain import chain
if __name__ == "__main__":
query = "Does Kendra support table extraction?"
print(chain.invoke(query))

File diff suppressed because it is too large Load Diff

@ -0,0 +1,26 @@
[tool.poetry]
name = "rag_aws_kendra"
version = "0.0.1"
description = ""
authors = []
readme = "README.md"
[tool.poetry.dependencies]
python = ">=3.8.1,<4.0"
langchain = ">=0.0.313, <0.1"
tiktoken = ">=0.5.1"
boto3 = ">=1.28.57"
awscli = ">=1.29.57"
[tool.poetry.group.dev.dependencies]
langchain-cli = ">=0.0.4"
fastapi = "^0.104.0"
sse-starlette = "^1.6.5"
[tool.langserve]
export_module = "rag_aws_kendra.chain"
export_attr = "chain"
[build-system]
requires = ["poetry-core"]
build-backend = "poetry.core.masonry.api"

@ -0,0 +1,49 @@
import os
from langchain.llms.bedrock import Bedrock
from langchain.prompts import ChatPromptTemplate
from langchain.retrievers import AmazonKendraRetriever
from langchain.schema.output_parser import StrOutputParser
from langchain.schema.runnable import RunnableParallel, RunnablePassthrough
# Get region and profile from env
region = os.environ.get("AWS_DEFAULT_REGION", "us-east-1")
profile = os.environ.get("AWS_PROFILE", "default")
kendra_index = os.environ.get("KENDRA_INDEX_ID", None)
if not kendra_index:
raise ValueError(
"No value provided in env variable 'KENDRA_INDEX_ID'. "
"A Kendra index is required to run this application."
)
# Set LLM and embeddings
model = Bedrock(
model_id="anthropic.claude-v2",
region_name=region,
credentials_profile_name=profile,
model_kwargs={'max_tokens_to_sample': 200}
)
# Create Kendra retriever
retriever = AmazonKendraRetriever(
index_id=kendra_index,
top_k=5,
region_name=region
)
# RAG prompt
template = """Answer the question based only on the following context:
{context}
Question: {question}
"""
prompt = ChatPromptTemplate.from_template(template)
# RAG
chain = (
RunnableParallel({"context": retriever, "question": RunnablePassthrough()})
| prompt
| model
| StrOutputParser()
)
Loading…
Cancel
Save