Posts in category: Amazon AWS

Today, we are excited to announce that the DBRX model, an open, general-purpose large language model (LLM) developed by Databricks, is available for customers through Amazon SageMaker JumpStart to deploy with one click for running inference. The DBRX LLM employs a fine-grained mixture-of-experts (MoE) architecture, pre-trained on 12 trillion tokens of carefully curated data and a maximum context length of 32,000 tokens.

You can try out this model with SageMaker JumpStart, a machine learning (ML) hub that provides access to algorithms and models so you can quickly get started with ML. In this post, we walk through how to discover and deploy the DBRX model.

What is the DBRX model

DBRX is a sophisticated decoder-only LLM built on transformer architecture. It employs a fine-grained MoE architecture, incorporating 132 billion total parameters, with 36 billion of these parameters being active for any given input.

The model underwent pre-training using a dataset consisting of 12 trillion tokens of text and code. In contrast to other open MoE models like Mixtral and Grok-1, DBRX features a fine-grained approach, using a higher quantity of smaller experts for optimized performance. Compared to other MoE models, DBRX has 16 experts and chooses 4.

The model is made available under the Databricks Open Model license, for use without restrictions.

What is SageMaker JumpStart

SageMaker JumpStart is a fully managed platform that offers state-of-the-art foundation models for various use cases such as content writing, code generation, question answering, copywriting, summarization, classification, and information retrieval. It provides a collection of pre-trained models that you can deploy quickly and with ease, accelerating the development and deployment of ML applications. One of the key components of SageMaker JumpStart is the Model Hub, which offers a vast catalog of pre-trained models, such as DBRX, for a variety of tasks.

You can now discover and deploy DBRX models with a few clicks in Amazon SageMaker Studio or programmatically through the SageMaker Python SDK, enabling you to derive model performance and MLOps controls with Amazon SageMaker features such as Amazon SageMaker Pipelines, Amazon SageMaker Debugger, or container logs. The model is deployed in an AWS secure environment and under your VPC controls, helping provide data security.

Discover models in SageMaker JumpStart

You can access the DBRX model through SageMaker JumpStart in the SageMaker Studio UI and the SageMaker Python SDK. In this section, we go over how to discover the models in SageMaker Studio.

SageMaker Studio is an integrated development environment (IDE) that provides a single web-based visual interface where you can access purpose-built tools to perform all ML development steps, from preparing data to building, training, and deploying your ML models. For more details on how to get started and set up SageMaker Studio, refer to Amazon SageMaker Studio.

In SageMaker Studio, you can access SageMaker JumpStart by choosing JumpStart in the navigation pane.

From the SageMaker JumpStart landing page, you can search for “DBRX” in the search box. The search results will list DBRX Instruct and DBRX Base.

You can choose the model card to view details about the model such as license, data used to train, and how to use the model. You will also find the Deploy button to deploy the model and create an endpoint.

Deploy the model in SageMaker JumpStart

Deployment starts when you choose the Deploy button. After deployment finishes, you will see that an endpoint is created. You can test the endpoint by passing a sample inference request payload or by selecting the testing option using the SDK. When you select the option to use the SDK, you will see example code that you can use in the notebook editor of your choice in SageMaker Studio.

DBRX Base

To deploy using the SDK, we start by selecting the DBRX Base model, specified by the model_id with value huggingface-llm-dbrx-base. You can deploy any of the selected models on SageMaker with the following code. Similarly, you can deploy DBRX Instruct using its own model ID.

from sagemaker.jumpstart.model import JumpStartModel

accept_eula = True

model = JumpStartModel(model_id="huggingface-llm-dbrx-base")
predictor = model.deploy(accept_eula=accept_eula)

This deploys the model on SageMaker with default configurations, including the default instance type and default VPC configurations. You can change these configurations by specifying non-default values in JumpStartModel. The Eula value must be explicitly defined as True in order to accept the end-user license agreement (EULA). Also make sure you have the account-level service limit for using ml.p4d.24xlarge or ml.pde.24xlarge for endpoint usage as one or more instances. You can follow the instructions here in order to request a service quota increase.

After it’s deployed, you can run inference against the deployed endpoint through the SageMaker predictor:

payload = {
    "inputs": "Hello!",
    "parameters": {
        "max_new_tokens": 10,
    },
}
predictor.predict(payload)

Example prompts

You can interact with the DBRX Base model like any standard text generation model, where the model processes an input sequence and outputs predicted next words in the sequence. In this section, we provide some example prompts and sample output.

Code generation

Using the preceding example, we can use code generation prompts as follows:

payload = { 
      "inputs": "Write a function to read a CSV file in Python using pandas library:", 
      "parameters": { 
          "max_new_tokens": 30, }, } 
           response = predictor.predict(payload)["generated_text"].strip() 
           print(response)

The following is the output:

import pandas as pd 
df = pd.read_csv("file_name.csv") 
#The above code will import pandas library and then read the CSV file using read_csv

Sentiment analysis

You can perform sentiment analysis using a prompt like the following with DBRX:

payload = {
"inputs": """
Tweet: "I am so excited for the weekend!"
Sentiment: Positive

Tweet: "Why does traffic have to be so terrible?"
Sentiment: Negative

Tweet: "Just saw a great movie, would recommend it."
Sentiment: Positive

Tweet: "According to the weather report, it will be cloudy today."
Sentiment: Neutral

Tweet: "This restaurant is absolutely terrible."
Sentiment: Negative

Tweet: "I love spending time with my family."
Sentiment:""",
"parameters": {
"max_new_tokens": 2,
},
}
response = predictor.predict(payload)["generated_text"].strip()
print(response)

The following is the output:

Positive

Question answering

You can use a question answering prompt like the following with DBRX:

# Question answering
payload = {
    "inputs": "Respond to the question: How did the development of transportation systems, such as railroads and steamships, impact global trade and cultural exchange?",
    "parameters": {
        "max_new_tokens": 225,
    },
}
response = predictor.predict(payload)["generated_text"].strip()
print(response)

The following is the output:

The development of transportation systems, such as railroads and steamships, impacted global trade and cultural exchange in a number of ways. 
The documents provided show that the development of these systems had a profound effect on the way people and goods were able to move around the world. 
One of the most significant impacts of the development of transportation systems was the way it facilitated global trade. 
The documents show that the development of railroads and steamships made it possible for goods to be transported more quickly and efficiently than ever before. 
This allowed for a greater exchange of goods between different parts of the world, which in turn led to a greater exchange of ideas and cultures. 
Another impact of the development of transportation systems was the way it facilitated cultural exchange. The documents show that the development of railroads and steamships made it possible for people to travel more easily and quickly than ever before. 
This allowed for a greater exchange of ideas and cultures between different parts of the world. Overall, the development of transportation systems, such as railroads and steamships, had a profound impact on global trade and cultural exchange.

DBRX Instruct

The instruction-tuned version of DBRX accepts formatted instructions where conversation roles must start with a prompt from the user and alternate between user instructions and the assistant (DBRX-instruct). The instruction format must be strictly respected, otherwise the model will generate suboptimal outputs. The template to build a prompt for the Instruct model is defined as follows:

<|im_start|>system
{system_message} <|im_end|>
<|im_start|>user
{human_message} <|im_end|>
<|im_start|>assistantn

<|im_start|> and <|im_end|> are special tokens for beginning of string (BOS) and end of string (EOS). The model can contain multiple conversation turns between system, user, and assistant, allowing for the incorporation of few-shot examples to enhance the model’s responses.

The following code shows how you can format the prompt in instruction format:

from typing import Dict, List

def format_instructions(instructions: List[Dict[str, str]]) -> List[str]:
    """Format instructions where conversation roles must alternate system/user/assistant/user/assistant/..."""
    prompt: List[str] = []
    for instruction in instructions:
        if instruction["role"] == "system":
            prompt.extend(["<|im_start|>systemn", (instruction["content"]).strip(), " <|im_end|>n"])
        elif instruction["role"] == "user":
            prompt.extend(["<|im_start|>usern", (instruction["content"]).strip(), " <|im_end|>n"])
        else:
            raise ValueError(f"Invalid role: {instruction['role']}. Role must be either 'user' or 'system'.")
    prompt.extend(["<|im_start|>assistantn"])
    return "".join(prompt)

def print_instructions(prompt: str, response: str) -> None:
    bold, unbold = '33[1m', '33[0m'
    print(f"{bold}> Input{unbold}n{prompt}nn{bold}> Output{unbold}n{response['generated_text'].strip()}n")

Knowledge retrieval

You can use the following prompt for knowledge retrieval:

instructions = [
{"role": "system", "content": "You are an expert in suggesting diet plans optimized for different sports or fitness activities, taking into account energy requirements, and recovery requirements."},
{"role": "user", "content": "I am trying to build muscle mass and increase my strength. What kind of high-protein, calorie-dense diet plan would you recommend for me?"}
]
prompt = format_instructions(instructions)
payload = {
"inputs": prompt,
"parameters": {"max_new_tokens": 700, "temperature": 0.5, "do_sample": True}
}
response=predictor.predict(payload)
print_instructions(prompt, response)

The following is the output:

> Input
<|im_start|>system
You are an expert in suggesting diet plans optimized for different sports or fitness activities, taking into account energy requirements, and recovery requirements. <|im_end|>
<|im_start|>user
I am trying to build muscle mass and increase my strength. What kind of high-protein, calorie-dense diet plan would you recommend for me? <|im_end|>
<|im_start|>assistant

> Output
Absolutely, I'd be happy to help you with that. Given your goal of building muscle mass and increasing strength, you'll want a diet that's high in protein and provides sufficient calories to support muscle growth and recovery. Here's a basic outline of what your diet plan could look like:

1. **Breakfast:**
   - 3 whole eggs
   - 3 egg whites
   - 1 cup of oatmeal
   - 1 scoop of whey protein
   - 1 banana

2. **Mid-Morning Snack:**
   - 1.5 ounces of almonds
   - 1 apple

3. **Lunch:**
   - 8 ounces of chicken breast
   - 1 cup of brown rice
   - 1 cup of mixed vegetables

4. **Afternoon Snack:**
   - 1.5 cups of Greek yogurt
   - 1 cup of mixed berries

5. **Dinner:**
   - 8 ounces of salmon or lean beef
   - 1 cup of quinoa
   - 1.5 cups of mixed vegetables

6. **Evening Snack:**
   - 1 scoop of casein protein
   - 1 tablespoon of almond butter

This is a high-protein, calorie-dense diet that includes a variety of nutrient-rich foods. It's important to note that everyone's nutritional needs are different, so you may need to adjust portion sizes or food choices based on your specific needs and preferences.

Also, remember to stay hydrated by drinking plenty of water throughout the day. Regular exercise, particularly strength training, is crucial to achieving your goals. It's also important to get enough sleep to aid in muscle recovery and growth.

Before starting any new diet or exercise program, it's a good idea to consult with a healthcare professional or a registered dietitian. They can provide personalized advice based on your health history and current lifestyle.

Code generation

DBRX models demonstrate benchmarked strengths for coding tasks. For example, see the following code:

instructions = [ 
{"role": "system", "content": "You are an expert in coding and programming. Your role is to provide examples, explanations, and insights related to programming languages, algorithms, data structures, and software development practices."}, 
{"role": "user", "content": "Write a Python script to read a CSV file containing stock prices and plot the closing prices over time using Matplotlib."} 
] 
prompt = format_instructions(instructions) 
payload = { "inputs": prompt, "parameters": {"max_new_tokens": 315, "temperature": 0.3, "do_sample": True} } 
response=predictor.predict(payload) 
print_instructions(prompt, response)

The following is the output:

> Input
<|im_start|>system
You are an expert in coding and programming. Your role is to provide examples, explanations, and insights related to programming languages, algorithms, data structures, and software development practices. <|im_end|>
<|im_start|>user
Write a Python script to read a CSV file containing stock prices and plot the closing prices over time using Matplotlib. <|im_end|>
<|im_start|>assistant

> Output
Sure, I'd be happy to help you with that. Here's a simple Python script that uses the `pandas` library to read a CSV file and `matplotlib` to plot the closing prices over time.

```python
import pandas as pd
import matplotlib.pyplot as plt

# Load the data from a CSV file
df = pd.read_csv('stock_prices.csv')

# Plot the closing prices over time
plt.figure(figsize=(12,5))
plt.plot(df['Date'], df['Close'])
plt.xlabel('Date')
plt.ylabel('Price')
plt.title('Stock Prices Over Time')
plt.show()
```

In this script, replace `'stock_prices.csv'` with the path to your CSV file. The file should have columns named 'Date' and 'Close' for this script to work correctly. If your file uses different column names, you'll need to adjust the script accordingly.

The `pandas` library is a powerful tool for data manipulation and analysis. It can read data from a variety of sources, including CSV files, and represent the data in a tabular format. The `matplotlib` library is a popular choice for creating static, animated, and interactive visualizations in Python.

This script first loads the data from the CSV file into a pandas DataFrame. It then plots the 'Close' column against the 'Date' column using matplotlib's `plot()` function. The `figure()` function is used to specify the size of the plot, and `show()` is used to display the plot.

Mathematics and reasoning

The DBRX models also report strengths in mathematic accuracy. For example, see the following code:

instructions = [
  {"role": "system", "content": "You are an expert in mathematics and reasoning. Your role is to provide examples, explanations, and insights related to mathematical concepts, problem-solving techniques, and logical reasoning."},
  {"role": "user", "content": "A car rental company charges $40 per day plus $0.25 per mile driven. If I rented a car for 3 days and drove 250 miles, what would be the total cost? Break down the calculation steps before giving the final answer."}
]
prompt = format_instructions(instructions)
payload = {
"inputs": prompt,
"parameters": {"max_new_tokens": 400, "temperature": 0.2, "do_sample": True}
}
response=predictor.predict(payload)
print_instructions(prompt, response)

DBRX can provide comprehension as shown in the following output with the math logic:

> Input
<|im_start|>system
You are an expert in mathematics and reasoning. Your role is to provide examples, explanations, and insights related to mathematical concepts, problem-solving techniques, and logical reasoning. <|im_end|>
<|im_start|>user
A car rental company charges $40 per day plus $0.25 per mile driven. If I rented a car for 3 days and drove 250 miles, what would be the total cost? Break down the calculation steps before giving the final answer. <|im_end|>
<|im_start|>assistant

> Output
Sure, I'd be happy to help with that! The problem you've presented involves both fixed costs (the daily rate) and variable costs (the cost per mile). Let's break it down:

1. First, let's calculate the total cost of the fixed daily rate. The car rental company charges $40 per day, and you rented the car for 3 days. So, the total cost for the daily rate is $40/day * 3 days = $120.

2. Next, let's calculate the total cost of the variable rate, which is the cost per mile driven. The company charges $0.25 per mile, and you drove 250 miles. So, the total cost for the mileage is $0.25/mile * 250 miles = $62.50.

3. Finally, to find the total cost of renting the car, we add the total cost of the daily rate and the total cost of the mileage. So, the total cost is $120 + $62.50 = $182.50.

So, the total cost of renting the car for 3 days and driving 250 miles would be $182.50.

Clean up

After you’re done running the notebook, make sure to delete all resources that you created in the process so your billing is stopped. Use the following code:

predictor.delete_model()
predictor.delete_endpoint()

Conclusion

In this post, we showed you how to get started with DBRX in SageMaker Studio and deploy the model for inference. Because foundation models are pre-trained, they can help lower training and infrastructure costs and enable customization for your use case. Visit SageMaker JumpStart in SageMaker Studio now to get started.

Resources

• SageMaker JumpStart documentation
• SageMaker JumpStart foundation models documentation
• SageMaker JumpStart product detail page
• SageMaker JumpStart model catalog

About the Authors

Shikhar Kwatra is an AI/ML Specialist Solutions Architect at Amazon Web Services, working with a leading Global System Integrator. He has earned the title of one of the Youngest Indian Master Inventors with over 400 patents in the AI/ML and IoT domains. He has over 8 years of industry experience from startups to large-scale enterprises, from IoT Research Engineer, Data Scientist, to Data & AI Architect. Shikhar aids in architecting, building, and maintaining cost-efficient, scalable cloud environments for organizations and supports GSI partners in building strategic industry

Niithiyn Vijeaswaran is a Solutions Architect at AWS. His area of focus is generative AI and AWS AI Accelerators. He holds a Bachelor’s degree in Computer Science and Bioinformatics. Niithiyn works closely with the Generative AI GTM team to enable AWS customers on multiple fronts and accelerate their adoption of generative AI. He’s an avid fan of the Dallas Mavericks and enjoys collecting sneakers.

Sebastian Bustillo is a Solutions Architect at AWS. He focuses on AI/ML technologies with a profound passion for generative AI and compute accelerators. At AWS, he helps customers unlock business value through generative AI. When he’s not at work, he enjoys brewing a perfect cup of specialty coffee and exploring the world with his wife.

Armando Diaz is a Solutions Architect at AWS. He focuses on generative AI, AI/ML, and data analytics. At AWS, Armando helps customers integrating cutting-edge generative AI capabilities into their systems, fostering innovation and competitive advantage. When he’s not at work, he enjoys spending time with his wife and family, hiking, and traveling the world.

Read More

At AWS re:Invent 2023, we announced the general availability of Knowledge Bases for Amazon Bedrock. With Knowledge Bases for Amazon Bedrock, you can securely connect foundation models (FMs) in Amazon Bedrock to your company data for fully managed Retrieval Augmented Generation (RAG).

In previous posts, we covered new capabilities like hybrid search support, metadata filtering to improve retrieval accuracy, and how Knowledge Bases for Amazon Bedrock manages the end-to-end RAG workflow.

Today, we’re introducing the new capability to chat with your document with zero setup in Knowledge Bases for Amazon Bedrock. With this new capability, you can securely ask questions on single documents, without the overhead of setting up a vector database or ingesting data, making it effortless for businesses to use their enterprise data. You only need to provide a relevant data file as input and choose your FM to get started.

But before we jump into the details of this feature, let’s start with the basics and understand what RAG is, its benefits, and how this new capability enables content retrieval and generation for temporal needs.

What is Retrieval Augmented Generation?

FM-powered artificial intelligence (AI) assistants have limitations, such as providing outdated information or struggling with context outside their training data. RAG addresses these issues by allowing FMs to cross-reference authoritative knowledge sources before generating responses.

With RAG, when a user asks a question, the system retrieves relevant context from a curated knowledge base, such as company documentation. It provides this context to the FM, which uses it to generate a more informed and precise response. RAG helps overcome FM limitations by augmenting its capabilities with an organization’s proprietary knowledge, enabling chatbots and AI assistants to provide up-to-date, context-specific information tailored to business needs without retraining the entire FM. At AWS, we recognize RAG’s potential and have worked to simplify its adoption through Knowledge Bases for Amazon Bedrock, providing a fully managed RAG experience.

Short-term and instant information needs

Although a knowledge base does all the heavy lifting and serves as a persistent large store of enterprise knowledge, you might require temporary access to data for specific tasks or analysis within isolated user sessions. Traditional RAG approaches are not optimized for these short-term, session-based data access scenarios.

Businesses incur charges for data storage and management. This may make RAG less cost-effective for organizations with highly dynamic or ephemeral information requirements, especially when data is only needed for specific, isolated tasks or analyses.

Ask questions on a single document with zero setup

This new capability to chat with your document within Knowledge Bases for Amazon Bedrock addresses the aforementioned challenges. It provides a zero-setup method to use your single document for content retrieval and generation-related tasks, along with the FMs provided by Amazon Bedrock. With this new capability, you can ask questions of your data without the overhead of setting up a vector database or ingesting data, making it effortless to use your enterprise data.

You can now interact with your documents in real time without prior data ingestion or database configuration. You don’t need to take any further data readiness steps before querying the data.

This zero-setup approach makes it straightforward to use your enterprise information assets with generative AI using Amazon Bedrock.

Use cases and benefits

Consider a recruiting firm that needs to analyze resumes and match candidates with suitable job opportunities based on their experience and skills. Previously, you would have to set up a knowledge base, invoking a data ingestion workflow to make sure only authorized recruiters can access the data. Additionally, you would need to manage cleanup when the data was no longer required for a session or candidate. In the end, you would pay more for the vector database storage and management than for the actual FM usage. This new feature in Knowledge Bases for Amazon Bedrock enables recruiters to quickly and ephemerally analyze resumes and match candidates with suitable job opportunities based on the candidate’s experience and skill set.

For another example, consider a product manager at a technology company who needs to quickly analyze customer feedback and support tickets to identify common issues and areas for improvement. With this new capability, you can simply upload a document to extract insights in no time. For example, you could ask “What are the requirements for the mobile app?” or “What are the common pain points mentioned by customers regarding our onboarding process?” This feature empowers you to rapidly synthesize this information without the hassle of data preparation or any management overhead. You can also request summaries or key takeaways, such as “What are the highlights from this requirements document?”

The benefits of this feature extend beyond cost savings and operational efficiency. By eliminating the need for vector databases and data ingestion, this new capability within Knowledge Bases for Amazon Bedrock helps secure your proprietary data, making it accessible only within the context of isolated user sessions.

Now that we’ve covered the feature benefits and the use cases it enables, let’s dive into how you can start using this new feature from Knowledge Bases for Amazon Bedrock.

Chat with your document in Knowledge Bases for Amazon Bedrock

You have multiple options to begin using this feature:

• The Amazon Bedrock console
• The Amazon Bedrock RetrieveAndGenerate API (SDK)

Let’s see how we can get started using the Amazon Bedrock console:

1. On the Amazon Bedrock console, under Orchestration in the navigation pane, choose Knowledge bases.
2. Choose Chat with your document.
   
3. Under Model, choose Select model.
   
4. Choose your model. For this example, we use the Claude 3 Sonnet model (we are only supporting Sonnet at the time of the launch).
5. Choose Apply.
   
6. Under Data, you can upload the document you want to chat with or point to the Amazon Simple Storage Service (Amazon S3) bucket location that contains your file. For this post, we upload a document from our computer.

The supported file formats are PDF, MD (Markdown), TXT, DOCX, HTML, CSV, XLS, and XLSX. Make that the file size does not exceed 10 MB and contains no more than 20,000 tokens. A token is considered to be a unit of text, such as a word, sub-word, number, or symbol, that is processed as a single entity. Due to the preset ingestion token limit, it is recommended to use a file under 10MB. However, a text-heavy file, that is much smaller than 10MB, can potentially breach the token limit.

You’re now ready to chat with your document.

As shown in the following screenshot, you can chat with your document in real time.

To customize your prompt, enter your prompt under System prompt.

Similarly, you can use the AWS SDK through the retrieve_and_generate API in major coding languages. In the following example, we use the AWS SDK for Python (Boto3):

import boto3

bedrock_client = boto3.client(service_name='bedrock-agent-runtime')
model_id = "your_model_id_here"    # Replace with your modelID
document_uri = "your_s3_uri_here"  # Replace with your S3 URI

def retrieveAndGenerate(input_text, sourceType, model_id, document_s3_uri=None, data=None):
    region = 'us-west-2'  
    model_arn = f'arn:aws:bedrock:{region}::foundation-model/{model_id}'

    if sourceType == "S3":
        return bedrock_client.retrieve_and_generate(
            input={'text': input_text},
            retrieveAndGenerateConfiguration={
                'type': 'EXTERNAL_SOURCES',
                'externalSourcesConfiguration': {
                    'modelArn': model_arn,
                    'sources': [
                        {
                            "sourceType": sourceType,
                            "s3Location": {
                                "uri": document_s3_uri  
                            }
                        }
                    ]
                }
            }
        )
        
    else:
        return bedrock_client.retrieve_and_generate(
            input={'text': input_text},
            retrieveAndGenerateConfiguration={
                'type': 'EXTERNAL_SOURCES',
                'externalSourcesConfiguration': {
                    'modelArn': model_arn,
                    'sources': [
                        {
                            "sourceType": sourceType,
                            "byteContent": {
                                "identifier": "testFile.txt",
                                "contentType": "text/plain",
                                "data": data  
                            }
                        }
                    ]
                }
            }
        )

response = retrieveAndGenerate(
                                input_text="What is the main topic of this document?",
                                sourceType="S3", 
                                model_id=model_id,
                                document_s3_uri=document_uri
                              )
                    
print(response['output']['text'])

Conclusion

In this post, we covered how Knowledge Bases for Amazon Bedrock now simplifies asking questions on a single document. We explored the core concepts behind RAG, the challenges this new feature addresses, and the various use cases it enables across different roles and industries. We also demonstrated how to configure and use this capability through the Amazon Bedrock console and the AWS SDK, showcasing the simplicity and flexibility of this feature, which provides a zero-setup solution to gather information from a single document, without setting up a vector database.

To further explore the capabilities of Knowledge Bases for Amazon Bedrock, refer to the following resources:

• Knowledge bases for Amazon Bedrock
• Getting started with Amazon Bedrock, RAG, and Vector database in Python
• Vector Embeddings and RAG Demystified: Leveraging Amazon Bedrock, Aurora, and LangChain (Part 1 and Part 2)

Share and learn with our generative AI community at community.aws.

About the authors

Suman Debnath is a Principal Developer Advocate for Machine Learning at Amazon Web Services. He frequently speaks at AI/ML conferences, events, and meetups around the world. He is passionate about large-scale distributed systems and is an avid fan of Python.

Sebastian Munera is a Software Engineer in the Amazon Bedrock Knowledge Bases team at AWS where he focuses on building customer solutions that leverage Generative AI and RAG applications. He has previously worked on building Generative AI-based solutions for customers to streamline their processes and Low code/No code applications. In his spare time he enjoys running, lifting and tinkering with technology.

Read More

Speaker diarization, an essential process in audio analysis, segments an audio file based on speaker identity. This post delves into integrating Hugging Face’s PyAnnote for speaker diarization with Amazon SageMaker asynchronous endpoints.

We provide a comprehensive guide on how to deploy speaker segmentation and clustering solutions using SageMaker on the AWS Cloud. You can use this solution for applications dealing with multi-speaker (over 100) audio recordings.

Solution overview

Amazon Transcribe is the go-to service for speaker diarization in AWS. However, for non-supported languages, you can use other models (in our case, PyAnnote) that will be deployed in SageMaker for inference. For short audio files where the inference takes up to 60 seconds, you can use real-time inference. For longer than 60 seconds, asynchronous inference should be used. The added benefit of asynchronous inference is the cost savings by auto scaling the instance count to zero when there are no requests to process.

Hugging Face is a popular open source hub for machine learning (ML) models. AWS and Hugging Face have a partnership that allows a seamless integration through SageMaker with a set of AWS Deep Learning Containers (DLCs) for training and inference in PyTorch or TensorFlow, and Hugging Face estimators and predictors for the SageMaker Python SDK. SageMaker features and capabilities help developers and data scientists get started with natural language processing (NLP) on AWS with ease.

The integration for this solution involves using Hugging Face’s pre-trained speaker diarization model using the PyAnnote library. PyAnnote is an open source toolkit written in Python for speaker diarization. This model, trained on the sample audio dataset, enables effective speaker partitioning in audio files. The model is deployed on SageMaker as an asynchronous endpoint setup, providing efficient and scalable processing of diarization tasks.

The following diagram illustrates the solution architecture.

For this post, we use the following audio file.

Stereo or multi-channel audio files are automatically downmixed to mono by averaging the channels. Audio files sampled at a different rate are resampled to 16kHz automatically upon loading.

Prerequisites

Complete the following prerequisites:

1. Create a SageMaker domain.
2. Make sure your AWS Identity and Access Management (IAM) user has the necessary access permissions for creating a SageMaker role.
3. Make sure the AWS account has a service quota for hosting a SageMaker endpoint for an ml.g5.2xlarge instance.

Create a model function for accessing PyAnnote speaker diarization from Hugging Face

You can use the Hugging Face Hub to access the desired pre-trained PyAnnote speaker diarization model. You use the same script for downloading the model file when creating the SageMaker endpoint.

See the following code:

from PyAnnote.audio import Pipeline

def model_fn(model_dir):
# Load the model from the specified model directory
model = Pipeline.from_pretrained(
"PyAnnote/speaker-diarization-3.1",
use_auth_token="Replace-with-the-Hugging-face-auth-token")
return model

Prepare essential files like inference.py, which contains the inference code:

%%writefile model/code/inference.py
from PyAnnote.audio import Pipeline
import subprocess
import boto3
from urllib.parse import urlparse
import pandas as pd
from io import StringIO
import os
import torch

def model_fn(model_dir):
    # Load the model from the specified model directory
    model = Pipeline.from_pretrained(
        "PyAnnote/speaker-diarization-3.1",
        use_auth_token="hf_oBxxxxxxxxxxxx)
    return model 


def diarization_from_s3(model, s3_file, language=None):
    s3 = boto3.client("s3")
    o = urlparse(s3_file, allow_fragments=False)
    bucket = o.netloc
    key = o.path.lstrip("/")
    s3.download_file(bucket, key, "tmp.wav")
    result = model("tmp.wav")
    data = {} 
    for turn, _, speaker in result.itertracks(yield_label=True):
        data[turn] = (turn.start, turn.end, speaker)
    data_df = pd.DataFrame(data.values(), columns=["start", "end", "speaker"])
    print(data_df.shape)
    result = data_df.to_json(orient="split")
    return result


def predict_fn(data, model):
    s3_file = data.pop("s3_file")
    language = data.pop("language", None)
    result = diarization_from_s3(model, s3_file, language)
    return {
        "diarization_from_s3": result
    }

Prepare a requirements.txt file, which contains the required Python libraries necessary to run the inference:

with open("model/code/requirements.txt", "w") as f:
    f.write("transformers==4.25.1n")
    f.write("boto3n")
    f.write("PyAnnote.audion")
    f.write("soundfilen")
    f.write("librosan")
    f.write("onnxruntimen")
    f.write("wgetn")
    f.write("pandas")

Lastly, compress the inference.py and requirements.txt files and save it as model.tar.gz:

!tar zcvf model.tar.gz *

Configure a SageMaker model

Define a SageMaker model resource by specifying the image URI, model data location in Amazon Simple Storage Service (S3), and SageMaker role:

import sagemaker
import boto3

sess = sagemaker.Session()

sagemaker_session_bucket = None
if sagemaker_session_bucket is None and sess is not None:
    sagemaker_session_bucket = sess.default_bucket()

try:
    role = sagemaker.get_execution_role()
except ValueError:
    iam = boto3.client("iam")
    role = iam.get_role(RoleName="sagemaker_execution_role")["Role"]["Arn"]

sess = sagemaker.Session(default_bucket=sagemaker_session_bucket)

print(f"sagemaker role arn: {role}")
print(f"sagemaker bucket: {sess.default_bucket()}")
print(f"sagemaker session region: {sess.boto_region_name}")

Upload the model to Amazon S3

Upload the zipped PyAnnote Hugging Face model file to an S3 bucket:

s3_location = f"s3://{sagemaker_session_bucket}/whisper/model/model.tar.gz"
!aws s3 cp model.tar.gz $s3_location

Create a SageMaker asynchronous endpoint

Configure an asynchronous endpoint for deploying the model on SageMaker using the provided asynchronous inference configuration:

from sagemaker.huggingface.model import HuggingFaceModel
from sagemaker.async_inference.async_inference_config import AsyncInferenceConfig
from sagemaker.s3 import s3_path_join
from sagemaker.utils import name_from_base

async_endpoint_name = name_from_base("custom-asyc")

# create Hugging Face Model Class
huggingface_model = HuggingFaceModel(
    model_data=s3_location,  # path to your model and script
    role=role,  # iam role with permissions to create an Endpoint
    transformers_version="4.17",  # transformers version used
    pytorch_version="1.10",  # pytorch version used
    py_version="py38",  # python version used
)

# create async endpoint configuration
async_config = AsyncInferenceConfig(
    output_path=s3_path_join(
        "s3://", sagemaker_session_bucket, "async_inference/output"
    ),  # Where our results will be stored
    # Add nofitication SNS if needed
    notification_config={
        # "SuccessTopic": "PUT YOUR SUCCESS SNS TOPIC ARN",
        # "ErrorTopic": "PUT YOUR ERROR SNS TOPIC ARN",
    },  #  Notification configuration
)

env = {"MODEL_SERVER_WORKERS": "2"}

# deploy the endpoint endpoint
async_predictor = huggingface_model.deploy(
    initial_instance_count=1,
    instance_type="ml.xx",
    async_inference_config=async_config,
    endpoint_name=async_endpoint_name,
    env=env,
)

Test the endpoint

Evaluate the endpoint functionality by sending an audio file for diarization and retrieving the JSON output stored in the specified S3 output path:

# Replace with a path to audio object in S3
from sagemaker.async_inference import WaiterConfig
res = async_predictor.predict_async(data=data)
print(f"Response output path: {res.output_path}")
print("Start Polling to get response:")

config = WaiterConfig(
  max_attempts=10, #  number of attempts
  delay=10#  time in seconds to wait between attempts
  )
res.get_result(config)
#import waiterconfig

To deploy this solution at scale, we suggest using AWS Lambda, Amazon Simple Notification Service (Amazon SNS), or Amazon Simple Queue Service (Amazon SQS). These services are designed for scalability, event-driven architectures, and efficient resource utilization. They can help decouple the asynchronous inference process from the result processing, allowing you to scale each component independently and handle bursts of inference requests more effectively.

Results

Model output is stored at s3://sagemaker-xxxx /async_inference/output/. The output shows that the audio recording has been segmented into three columns:

• Start (start time in seconds)
• End (end time in seconds)
• Speaker (speaker label)

The following code shows an example of our results:

[0.9762308998, 8.9049235993, "SPEAKER_01"]

[9.533106961, 12.1646859083, "SPEAKER_01"]

[13.1324278438, 13.9303904924, "SPEAKER_00"]

[14.3548387097, 26.1884550085, "SPEAKER_00"]

[27.2410865874, 28.2258064516, "SPEAKER_01"]

[28.3446519525, 31.298811545, "SPEAKER_01"]

Clean up

You can set a scaling policy to zero by setting MinCapacity to 0; asynchronous inference lets you auto scale to zero with no requests. You don’t need to delete the endpoint, it scales from zero when needed again, reducing costs when not in use. See the following code:

# Common class representing application autoscaling for SageMaker 
client = boto3.client('application-autoscaling') 

# This is the format in which application autoscaling references the endpoint
resource_id='endpoint/' + <endpoint_name> + '/variant/' + <'variant1'> 

# Define and register your endpoint variant
response = client.register_scalable_target(
    ServiceNamespace='sagemaker', 
    ResourceId=resource_id,
    ScalableDimension='sagemaker:variant:DesiredInstanceCount', # The number of EC2 instances for your Amazon SageMaker model endpoint variant.
    MinCapacity=0,
    MaxCapacity=5
)

async_predictor.delete_endpoint(async_endpoint_name)

Read More

Organizations across industries are using automatic text summarization to more efficiently handle vast amounts of information and make better decisions. In the financial sector, investment banks condense earnings reports down to key takeaways to rapidly analyze quarterly performance. Media companies use summarization to monitor news and social media so journalists can quickly write stories on developing issues. Government agencies summarize lengthy policy documents and reports to help policymakers strategize and prioritize goals.

By creating condensed versions of long, complex documents, summarization technology enables users to focus on the most salient content. This leads to better comprehension and retention of critical information. The time savings allow stakeholders to review more material in less time, gaining a broader perspective. With enhanced understanding and more synthesized insights, organizations can make better informed strategic decisions, accelerate research, improve productivity, and increase their impact. The transformative power of advanced summarization capabilities will only continue growing as more industries adopt artificial intelligence (AI) to harness overflowing information streams.

In this post, we explore leading approaches for evaluating summarization accuracy objectively, including ROUGE metrics, METEOR, and BERTScore. Understanding the strengths and weaknesses of these techniques can help guide selection and improvement efforts. The overall goal of this post is to demystify summarization evaluation to help teams better benchmark performance on this critical capability as they seek to maximize value.

Types of summarization

Summarization can generally be divided into two main types: extractive summarization and abstractive summarization. Both approaches aim to condense long pieces of text into shorter forms, capturing the most critical information or essence of the original content, but they do so in fundamentally different ways.

Extractive summarization involves identifying and extracting key phrases, sentences, or segments from the original text without altering them. The system selects parts of the text deemed most informative or representative of the whole. Extractive summarization is useful if accuracy is critical and the summary needs to reflect the exact information from the original text. These could be use cases like highlighting specific legal terms, obligations, and rights outlined in the terms of use. The most common techniques used for extractive summarization are term frequency-inverse document frequency (TF-IDF), sentence scoring, text rank algorithm, and supervised machine learning (ML).

Abstractive summarization goes a step further by generating new phrases and sentences that were not in the original text, essentially paraphrasing and condensing the original content. This approach requires a deeper understanding of the text, because the AI needs to interpret the meaning and then express it in a new, concise form. Large language models (LLMs) are best suited for abstractive summarization because the transformer models use attention mechanisms to focus on relevant parts of the input text when generating summaries. The attention mechanism allows the model to assign different weights to different words or tokens in the input sequence, enabling it to capture long-range dependencies and contextually relevant information.

In addition to these two primary types, there are hybrid approaches that combine extractive and abstractive methods. These approaches might start with extractive summarization to identify the most important content and then use abstractive techniques to rewrite or condense that content into a fluent summary.

The challenge

Finding the optimal method to evaluate summary quality remains an open challenge. As organizations increasingly rely on automatic text summarization to distill key information from documents, the need grows for standardized techniques to measure summarization accuracy. Ideally, these evaluation metrics would quantify how well machine-generated summaries extract the most salient content from source texts and present coherent summaries reflecting the original meaning and context.

However, developing robust evaluation methodologies for text summarization presents difficulties:

• Human-authored reference summaries used for comparison often exhibit high variability based on subjective determinations of importance
• Nuanced aspects of summary quality like fluency, readability, and coherence prove difficult to quantify programmatically
• Wide variation exists across summarization methods from statistical algorithms to neural networks, complicating direct comparisons

Recall-Oriented Understudy for Gisting Evaluation (ROUGE)

ROUGE metrics, such as ROUGE-N and ROUGE-L, play a crucial role in evaluating the quality of machine-generated summaries compared to human-written reference summaries. These metrics focus on assessing the overlap between the content of machine-generated and human-crafted summaries by analyzing n-grams, which are groups of words or tokens. For instance, ROUGE-1 evaluates the match of individual words (unigrams), whereas ROUGE-2 considers pairs of words (bigrams). Additionally, ROUGE-N assesses the longest common subsequence of words between the two texts, allowing for flexibility in word order.

To illustrate this, consider the following examples:

• ROGUE-1 metric – ROUGE-1 evaluates the overlap of unigrams (single words) between a generated summary and a reference summary. For example, if a reference summary contains “The quick brown fox jumps,” and the generated summary is “The brown fox jumps quickly,” the ROUGE-1 metric would consider “brown,” “fox,” and “jumps” as overlapping unigrams. ROUGE-1 focuses on the presence of individual words in the summaries, measuring how well the generated summary captures the key words from the reference summary.
• ROGUE-2 metric – ROUGE-2 assesses the overlap of bigrams (pairs of adjacent words) between a generated summary and a reference summary. For instance, if the reference summary has “The cat is sleeping,” and the generated summary reads “A cat is sleeping,” ROUGE-2 would identify “cat is” and “is sleeping” as an overlapping bigram. ROUGE-2 provides insight into how well the generated summary maintains the sequence and context of word pairs compared to the reference summary.
• ROUGE-N metric – ROUGE-N is a generalized form where N represents any number, allowing evaluation based on n-grams (sequences of N words). Considering N=3, if the reference summary states “The sun is shining brightly,” and the generated summary is “Sun shining brightly,” ROUGE-3 would recognize “sun shining brightly” as a matching trigram. ROUGE-N offers flexibility to evaluate summaries based on different lengths of word sequences, providing a more comprehensive assessment of content overlap.

These examples illustrate how ROUGE-1, ROUGE-2, and ROUGE-N metrics function in evaluating automatic summarization or machine translation tasks by comparing generated summaries with reference summaries based on different levels of word sequences.

Calculate a ROUGE-N score

You can use the following steps to calculate a ROUGE-N score:

1. Tokenize the generated summary and the reference summary into individual words or tokens using basic tokenization methods like splitting by whitespace or natural language processing (NLP) libraries.
2. Generate n-grams (contiguous sequences of N words) from both the generated summary and the reference summary.
3. Count the number of overlapping n-grams between the generated summary and the reference summary.
4. Calculate precision, recall, and F1 score:
   • Precision – The number of overlapping n-grams divided by the total number of n-grams in the generated summary.
   • Recall – The number of overlapping n-grams divided by the total number of n-grams in the reference summary.
   • F1 score – The harmonic mean of precision and recall, calculated as (2 * precision * recall) / (precision + recall).
5. The aggregate F1 score obtained from calculating precision, recall, and F1 score for each row in the dataset is considered as the ROUGE-N score.

Limitations

ROGUE has the following limitations:

• Narrow focus on lexical overlap – The core idea behind ROUGE is to compare the system-generated summary to a set of reference or human-created summaries, and measure the lexical overlap between them. This means ROUGE has a very narrow focus on word-level similarity. It doesn’t actually evaluate semantic meaning, coherence, or readability of the summary. A system could achieve high ROUGE scores by simply extracting sentences word-for-word from the original text, without generating a coherent or concise summary.
• Insensitivity to paraphrasing – Because ROUGE relies on lexical matching, it can’t detect semantic equivalence between words and phrases. Therefore, paraphrasing and use of synonyms will often lead to lower ROUGE scores, even if the meaning is preserved. This disadvantages systems that paraphrase or summarize in an abstractive way.
• Lack of semantic understanding – ROUGE doesn’t evaluate whether the system truly understood the meanings and concepts in the original text. A summary could achieve high lexical overlap with references, while missing the main ideas or containing factual inconsistencies. ROUGE would not identify these issues.

When to use ROUGE

ROUGE is simple and fast to calculate. Use it as a baseline or benchmark for summary quality related to content selection. ROUGE metrics are most effectively employed in scenarios involving abstractive summarization tasks, automatic summarization evaluation, assessments of LLMs, and comparative analyses of different summarization approaches. By using ROUGE metrics in these contexts, stakeholders can quantitatively evaluate the quality and effectiveness of summary generation processes.

Metric for Evaluation of Translation with Explicit Ordering (METEOR)

One of the major challenges in evaluating summarization systems is assessing how well the generated summary flows logically, rather than just selecting relevant words and phrases from the source text. Simply extracting relevant keywords and sentences doesn’t necessarily produce a coherent and cohesive summary. The summary should flow smoothly and connect ideas logically, even if they aren’t presented in the same order as the original document.

The flexibility of matching by reducing words to their root or base form (For example, after stemming, words like “running,” “runs,” and “ran” all become “run”) and synonyms means METEOR correlates better with human judgements of summary quality. It can identify if important content is preserved, even if the wording differs. This is a key advantage over n-gram based metrics like ROUGE, which only look for exact token matches. METEOR also gives higher scores to summaries that focus on the most salient content from the reference. Lower scores are given to repetitive or irrelevant information. This aligns well with the goal of summarization to keep the most important content only. METEOR is a semantically meaningful metric that can overcome some of the limitations of n-gram matching for evaluating text summarization. The incorporation of stemming and synonyms allows for better assessment of information overlap and content accuracy.

To illustrate this, consider the following examples:

Reference Summary: Leaves fall during autumn.

Generated Summary 1: Leaves drop in fall.

Generated Summary 2: Leaves green in summer.

The words that match between the reference and generated summary 1 are highlighted:

Reference Summary: Leavesfall during autumn.

Generated Summary 1: Leaves drop in fall.

Even though “fall” and “autumn” are different tokens, METEOR recognizes them as synonyms through its synonym matching. “Drop” and “fall” are identified as a stemmed match. For generated summary 2, there are no matches with the reference summary besides “Leaves,” so this summary would receive a much lower METEOR score. The more semantically meaningful matches, the higher the METEOR score. This allows METEOR to better evaluate the content and accuracy of summaries compared to simple n-gram matching.

Calculate a METEOR score

Complete the following steps to calculate a METEOR score:

1. Tokenize the generated summary and the reference summary into individual words or tokens using basic tokenization methods like splitting by whitespace or NLP libraries.
2. Calculate the unigram precision, recall, and F-mean score, giving more weightage to recall than precision.
3. Apply a penalty for exact matches to avoid overemphasizing them. The penalty is chosen based on dataset characteristics, task requirements, and the balance between precision and recall. Subtract this penalty from the F-mean score calculated in Step 2.
4. Calculate the F-mean score for stemmed forms (reducing words to their base or root form) and synonyms for unigrams where applicable. Aggregate this with the earlier calculated F-mean score to obtain the final METEOR score. The METEOR score ranges from 0–1, where 0 indicates no similarity between the generated summary and reference summary, and 1 indicates perfect alignment. Typically, summarization scores fall between 0–0.6.

Limitations

When employing the METEOR metric for evaluating summarization tasks, several challenges may arise:

• Semantic complexity – METEOR’s emphasis on semantic similarity can struggle to capture the nuanced meanings and context in complex summarization tasks, potentially leading to inaccuracies in evaluation.
• Reference variability – Variability in human-generated reference summaries can impact METEOR scores, because differences in reference content may affect the evaluation of machine-generated summaries.
• Linguistic diversity – The effectiveness of METEOR may vary across languages due to linguistic variations, syntax differences, and semantic nuances, posing challenges in multilingual summarization evaluations.
• Length discrepancy – Evaluating summaries of varying lengths can be challenging for METEOR, because discrepancies in length compared to the reference summary may result in penalties or inaccuracies in assessment.
• Parameter tuning – Optimizing METEOR’s parameters for different datasets and summarization tasks can be time-consuming and require careful tuning to make sure the metric provides accurate evaluations.
• Evaluation bias – There is a risk of evaluation bias with METEOR if not properly adjusted or calibrated for specific summarization domains or tasks. This can potentially lead to skewed results and affect the reliability of the evaluation process.

By being aware of these challenges and considering them when using METEOR as a metric for summarization tasks, researchers and practitioners can navigate potential limitations and make more informed decisions in their evaluation processes.

When to use METEOR

METEOR is commonly used to automatically evaluate the quality of text summaries. It is preferable to use METEOR as an evaluation metric when the order of ideas, concepts, or entities in the summary matters. METEOR considers the order and matches n-grams between the generated summary and reference summaries. It rewards summaries that preserve sequential information. Unlike metrics like ROUGE, which rely on overlap of n-grams with reference summaries, METEOR matches stems, synonyms, and paraphrases. METEOR works better when there can be multiple correct ways of summarizing the original text. METEOR incorporates WordNet synonyms and stemmed tokens when matching n-grams. In short, summaries that are semantically similar but use different words or phrasing will still score well. METEOR has a built-in penalty for summaries with repetitive n-grams. Therefore, it discourages word-for-word extraction or lack of abstraction. METEOR is a good choice when semantic similarity, order of ideas, and fluent phrasing are important for judging summary quality. It is less appropriate for tasks where only lexical overlap with reference summaries matters.

BERTScore

Surface-level lexical measures like ROUGE and METEOR evaluate summarization systems by comparing the word overlap between a candidate summary and a reference summary. However, they rely heavily on exact string matching between words and phrases. This means they may miss semantic similarities between words and phrases that have different surface forms but similar underlying meanings. By relying only on surface matching, these metrics may underestimate the quality of system summaries that use synonymous words or paraphrase concepts differently from reference summaries. Two summaries could convey nearly identical information but receive low surface-level scores due to vocabulary differences.

BERTScore is a way to automatically evaluate how good a summary is by comparing it to a reference summary written by a human. It uses BERT, a popular NLP technique, to understand the meaning and context of words in the candidate summary and reference summary. Specifically, it looks at each word or token in the candidate summary and finds the most similar word in the reference summary based on the BERT embeddings, which are vector representations of the meaning and context of each word. It measures the similarity using cosine similarity, which tells how close the vectors are to each other. For each word in the candidate summary, it finds the most related word in the reference summary using BERT’s understanding of language. It compares all these word similarities across the whole summary to get an overall score of how semantically similar the candidate summary is to the reference summary. The more similar the words and meanings captured by BERT, the higher the BERTScore. This allows it to automatically evaluate the quality of a generated summary by comparing it to a human reference without needing human evaluation each time.

To illustrate this, imagine you have a machine-generated summary: “The quick brown fox jumps over the lazy dog.” Now, let’s consider a human-crafted reference summary: “A fast brown fox leaps over a sleeping canine.”

Calculate a BERTScore

Complete the following steps to calculate a BERTScore:

1. BERTScore uses contextual embeddings to represent each token in both the candidate (machine-generated) and reference (human-crafted) sentences. Contextual embeddings are a type of word representation in NLP that captures the meaning of a word based on its context within a sentence or text. Unlike traditional word embeddings that assign a fixed vector to each word regardless of its context, contextual embeddings consider the surrounding words to generate a unique representation for each word depending on how it is used in a specific sentence.
2. The metric then computes the similarity between each token in the candidate sentence with each token in the reference sentence using cosine similarity. Cosine similarity helps us quantify how closely related two sets of data are by focusing on the direction they point in a multi-dimensional space, making it a valuable tool for tasks like search algorithms, NLP, and recommendation systems.
3. By comparing the contextual embeddings and computing similarity scores for all tokens, BERTScore generates a comprehensive evaluation that captures the semantic relevance and context of the generated summary compared to the human-crafted reference.
4. The final BERTScore output provides a similarity score that reflects how well the machine-generated summary aligns with the reference summary in terms of meaning and context.

In essence, BERTScore goes beyond traditional metrics by considering the semantic nuances and context of sentences, offering a more sophisticated evaluation that closely mirrors human judgment. This advanced approach enhances the accuracy and reliability of evaluating summarization tasks, making BERTScore a valuable tool in assessing text generation systems.

Limitations:

Although BERTScore offers significant advantages in evaluating summarization tasks, it also comes with certain limitations that need to be considered:

• Computational intensity – BERTScore can be computationally intensive due to its reliance on pre-trained language models like BERT. This can lead to longer evaluation times, especially when processing large volumes of text data.
• Dependency on pre-trained models – The effectiveness of BERTScore is highly dependent on the quality and relevance of the pre-trained language model used. In scenarios where the pre-trained model may not adequately capture the nuances of the text, the evaluation results may be affected.
• Scalability – Scaling BERTScore for large datasets or real-time applications can be challenging due to its computational demands. Implementing BERTScore in production environments may require optimization strategies to provide efficient performance.
• Domain specificity – BERTScore’s performance may vary across different domains or specialized text types. Adapting the metric to specific domains or tasks may require fine-tuning or adjustments to produce accurate evaluations.
• Interpretability – Although BERTScore provides a comprehensive evaluation based on contextual embeddings, interpreting the specific reasons behind the similarity scores generated for each token can be complex and may require additional analysis.
• Reference-free evaluation – Although BERTScore reduces the reliance on reference summaries for evaluation, this reference-free approach may not fully capture all aspects of summarization quality, particularly in scenarios where human-crafted references are essential for assessing content relevance and coherence.

Acknowledging these limitations can help you make informed decisions when using BERTScore as a metric for evaluating summarization tasks, providing a balanced understanding of its strengths and constraints.

When to use BERTScore

BERTScore can evaluate the quality of text summarization by comparing a generated summary to a reference summary. It uses neural networks like BERT to measure semantic similarity beyond just exact word or phrase matching. This makes BERTScore very useful when semantic fidelity preserving the full meaning and content is critical for your summarization task. BERTScore will give higher scores to summaries that convey the same information as the reference summary, even if they use different words and sentence structures. The bottom line is that BERTScore is ideal for summarization tasks where retaining the full semantic meaning not just keywords or topics is vital. Its advanced neural scoring allows it to compare meaning beyond surface-level word matching. This makes it suitable for cases where subtle differences in wording can substantially alter overall meaning and implications. BERTScore, in particular, excels in capturing semantic similarity, which is crucial for assessing the quality of abstractive summaries like those produced by Retrieval Augmented Generation (RAG) models.

Model evaluation frameworks

Model evaluation frameworks are essential for accurately gauging the performance of various summarization models. These frameworks are instrumental in comparing models, providing coherence between generated summaries and source content, and pinpointing deficiencies in evaluation methods. By conducting thorough assessments and consistent benchmarking, these frameworks propel text summarization research by advocating standardized evaluation practices and enabling multifaceted model comparisons.

In AWS, the FMEval library within Amazon SageMaker Clarify streamlines the evaluation and selection of foundation models (FMs) for tasks like text summarization, question answering, and classification. It empowers you to evaluate FMs based on metrics such as accuracy, robustness, creativity, bias, and toxicity, supporting both automated and human-in-the-loop evaluations for LLMs. With UI-based or programmatic evaluations, FMEval generates detailed reports with visualizations to quantify model risks like inaccuracies, toxicity, or bias, helping organizations align with their responsible generative AI guidelines. In this section, we demonstrate how to use the FMEval library.

Evaluate Claude v2 on summarization accuracy using Amazon Bedrock

The following code snippet is an example of how to interact with the Anthropic Claude model using Python code:

import json
# We use Claude v2 in this example.
# See https://docs.anthropic.com/claude/reference/claude-on-amazon-bedrock#list-available-models
# for instructions on how to list the model IDs for all available Claude model variants.
model_id = 'anthropic.claude-v2'
accept = "application/json"
contentType = "application/json"
# `prompt_data` is structured in the format that the Claude model expects, as documented here:
# https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-claude.html#model-parameters-claude-request-body
prompt_data = """Human: Who is Barack Obama?
Assistant:
"""
# For more details on parameters that can be included in `body` (such as "max_tokens_to_sample"),
# see https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-claude.html#model-parameters-claude-request-body
body = json.dumps({"prompt": prompt_data, "max_tokens_to_sample": 500})
# Invoke the model
response = bedrock_runtime.invoke_model(
body=body, modelId=model_id, accept=accept, contentType=contentType
)
# Parse the invocation response
response_body = json.loads(response.get("body").read())
print(response_body.get("completion"))

In simple terms, this code performs the following actions:

1. Import the necessary libraries, including json, to work with JSON data.
2. Define the model ID as anthropic.claude-v2 and set the content type for the request.
3. Create a prompt_data variable that structures the input data for the Claude model. In this case, it asks the question “Who is Barack Obama?” and expects a response from the model.
4. Construct a JSON object named body that includes the prompt data, and specify additional parameters like the maximum number of tokens to generate.
5. Invoke the Claude model using bedrock_runtime.invoke_model with the defined parameters.
6. Parse the response from the model, extract the completion (generated text), and print it out.

Make sure the AWS Identity and Access Management (IAM) role associated with the Amazon SageMaker Studio user profile has access to the Amazon Bedrock models being invoked. Refer to Identity-based policy examples for Amazon Bedrock for guidance on best practices and examples of identity-based policies for Amazon Bedrock.

Using the FMEval library to evaluate the summarized output from Claude

We use the following code to evaluate the summarized output:

from fmeval.data_loaders.data_config import DataConfig
from fmeval.model_runners.bedrock_model_runner import BedrockModelRunner
from fmeval.constants import MIME_TYPE_JSONLINES
from fmeval.eval_algorithms.summarization_accuracy import SummarizationAccuracy
config = DataConfig(
    dataset_name="gigaword_sample",
    dataset_uri="gigaword_sample.jsonl",
    dataset_mime_type=MIME_TYPE_JSONLINES,
    model_input_location="document",
    target_output_location="summary"
)
bedrock_model_runner = BedrockModelRunner(
    model_id=model_id,
    output='completion',
    content_template='{"prompt": $prompt, "max_tokens_to_sample": 500}'
)
eval_algo = SummarizationAccuracy()
eval_output = eval_algo.evaluate(model=bedrock_model_runner, dataset_config=config,
prompt_template="Human: Summarise the following text in one sentence: $featurennAssistant:n", save=True)

In the preceding code snippet, to evaluate text summarization using the FMEval library, we complete the following steps:

1. Create a ModelRunner to perform invocation on your LLM. The FMEval library provides built-in support for Amazon SageMaker endpoints and Amazon SageMaker JumpStart LLMs. You can also extend the ModelRunner interface for any LLMs hosted anywhere.
2. Use supported eval_algorithms like toxicity, summarization, accuracy, semantic, and robustness, based on your evaluation needs.
3. Customize the evaluation configuration parameters for your specific use case.
4. Use the evaluation algorithm with either built-in or custom datasets to evaluate your LLM model. The dataset used in this case is sourced from the following GitHub repo.

Refer to the developer guide and examples for detailed usage of evaluation algorithms.

The following table summarizes the results of the evaluation.

Check out the sample notebook for more details about the summarization evaluation that we discussed in this post.

Conclusion

ROUGE, METEOR, and BERTScore all measure the quality of machine-generated summaries, but focus on different aspects like lexical overlap, fluency, or semantic similarity. Make sure to select the metric that aligns with what defines “good” for your specific summarization use case. You can also use a combination of metrics. This provides a more well-rounded evaluation and guards against potential weaknesses of any individual metric. With the right measurements, you can iteratively improve your summarizers to meet whichever notion of accuracy matters most.

Additionally, FM and LLM evaluation is necessary to be able to productionize these models at scale. With FMEval, you get a vast set of built-in algorithms across many NLP tasks, but also a scalable and flexible tool for large-scale evaluations of your own models, datasets, and algorithms. To scale up, you can use this package in your LLMOps pipelines to evaluate multiple models. To learn more about FMEval in AWS and how to use it effectively, refer to Use SageMaker Clarify to evaluate large language models. For further understanding and insights into the capabilities of SageMaker Clarify in evaluating FMs, see Amazon SageMaker Clarify Makes It Easier to Evaluate and Select Foundation Models.

About the Authors

Dinesh Kumar Subramani is a Senior Solutions Architect based in Edinburgh, Scotland. He specializes in artificial intelligence and machine learning, and is member of technical field community with in Amazon. Dinesh works closely with UK Central Government customers to solve their problems using AWS services. Outside of work, Dinesh enjoys spending quality time with his family, playing chess, and exploring a diverse range of music.

Pranav Sharma is an AWS leader driving technology and business transformation initiatives across Europe, the Middle East, and Africa. He has experience in designing and running artificial intelligence platforms in production that support millions of customers and deliver business outcomes. He has played technology and people leadership roles for Global Financial Services organizations. Outside of work, he likes to read, play tennis with his son, and watch movies.

Read More

Conversational artificial intelligence (AI) assistants are engineered to provide precise, real-time responses through intelligent routing of queries to the most suitable AI functions. With AWS generative AI services like Amazon Bedrock, developers can create systems that expertly manage and respond to user requests. Amazon Bedrock is a fully managed service that offers a choice of high-performing foundation models (FMs) from leading AI companies like AI21 Labs, Anthropic, Cohere, Meta, Stability AI, and Amazon using a single API, along with a broad set of capabilities you need to build generative AI applications with security, privacy, and responsible AI.

This post assesses two primary approaches for developing AI assistants: using managed services such as Agents for Amazon Bedrock, and employing open source technologies like LangChain. We explore the advantages and challenges of each, so you can choose the most suitable path for your needs.

What is an AI assistant?

An AI assistant is an intelligent system that understands natural language queries and interacts with various tools, data sources, and APIs to perform tasks or retrieve information on behalf of the user. Effective AI assistants possess the following key capabilities:

• Natural language processing (NLP) and conversational flow
• Knowledge base integration and semantic searches to understand and retrieve relevant information based on the nuances of conversation context
• Running tasks, such as database queries and custom AWS Lambda functions
• Handling specialized conversations and user requests

We demonstrate the benefits of AI assistants using Internet of Things (IoT) device management as an example. In this use case, AI can help technicians manage machinery efficiently with commands that fetch data or automate tasks, streamlining operations in manufacturing.

Agents for Amazon Bedrock approach

Agents for Amazon Bedrock allows you to build generative AI applications that can run multi-step tasks across a company’s systems and data sources. It offers the following key capabilities:

• Automatic prompt creation from instructions, API details, and data source information, saving weeks of prompt engineering effort
• Retrieval Augmented Generation (RAG) to securely connect agents to a company’s data sources and provide relevant responses
• Orchestration and running of multi-step tasks by breaking down requests into logical sequences and calling necessary APIs
• Visibility into the agent’s reasoning through a chain-of-thought (CoT) trace, allowing troubleshooting and steering of model behavior
• Prompt engineering abilities to modify the automatically generated prompt template for enhanced control over agents

You can use Agents for Amazon Bedrock and Knowledge Bases for Amazon Bedrock to build and deploy AI assistants for complex routing use cases. They provide a strategic advantage for developers and organizations by simplifying infrastructure management, enhancing scalability, improving security, and reducing undifferentiated heavy lifting. They also allow for simpler application layer code because the routing logic, vectorization, and memory is fully managed.

Solution overview

This solution introduces a conversational AI assistant tailored for IoT device management and operations when using Anthropic’s Claude v2.1 on Amazon Bedrock. The AI assistant’s core functionality is governed by a comprehensive set of instructions, known as a system prompt, which delineates its capabilities and areas of expertise. This guidance makes sure the AI assistant can handle a wide range of tasks, from managing device information to running operational commands.

"""The following is the system prompt that outlines the full scope of the AI assistant's capabilities:
You are an IoT Ops agent that handles the following activities:
- Looking up IoT device information
- Checking IoT operating metrics (historical data)
- Performing actions on a device-by-device ID
- Answering general questions
You can check device information (Device ID, Features, Technical Specifications, Installation Guide, Maintenance and Troubleshooting, Safety Guidelines, Warranty, and Support) from the "IotDeviceSpecs" knowledge base.
Additionally, you can access device historical data or device metrics. The device metrics are stored in an Athena DB named "iot_ops_glue_db" in a table named "iot_device_metrics". 
The table schema includes fields for oil level, temperature, pressure, received_at timestamp, and device_id.
The available actions you can perform on the devices include start, shutdown, and reboot."""

Equipped with these capabilities, as detailed in the system prompt, the AI assistant follows a structured workflow to address user questions. The following figure provides a visual representation of this workflow, illustrating each step from initial user interaction to the final response.

The workflow is composed of the following steps:

1. The process begins when a user requests the assistant to perform a task; for example, asking for the maximum data points for a specific IoT device device_xxx. This text input is captured and sent to the AI assistant.
2. The AI assistant interprets the user’s text input. It uses the provided conversation history, action groups, and knowledge bases to understand the context and determine the necessary tasks.
3. After the user’s intent is parsed and understood, the AI assistant defines tasks. This is based on the instructions that are interpreted by the assistant as per the system prompt and user’s input.
4. The tasks are then run through a series of API calls. This is done using ReAct prompting, which breaks down the task into a series of steps that are processed sequentially:
   1. For device metrics checks, we use the check-device-metrics action group, which involves an API call to Lambda functions that then query Amazon Athena for the requested data.
   2. For direct device actions like start, stop, or reboot, we use the action-on-device action group, which invokes a Lambda function. This function initiates a process that sends commands to the IoT device. For this post, the Lambda function sends notifications using Amazon Simple Email Service (Amazon SES).
   3. We use Knowledge Bases for Amazon Bedrock to fetch from historical data stored as embeddings in the Amazon OpenSearch Service vector database.
5. After the tasks are complete, the final response is generated by the Amazon Bedrock FM and conveyed back to the user.
6. Agents for Amazon Bedrock automatically stores information using a stateful session to maintain the same conversation. The state is deleted after a configurable idle timeout elapses.

Technical overview

The following diagram illustrates the architecture to deploy an AI assistant with Agents for Amazon Bedrock.

It consists of the following key components:

• Conversational interface – The conversational interface uses Streamlit, an open source Python library that simplifies the creation of custom, visually appealing web apps for machine learning (ML) and data science. It is hosted on Amazon Elastic Container Service (Amazon ECS) with AWS Fargate, and it is accessed using an Application Load Balancer. You can use Fargate with Amazon ECS to run containers without having to manage servers, clusters, or virtual machines.
• Agents for Amazon Bedrock – Agents for Amazon Bedrock completes the user queries through a series of reasoning steps and corresponding actions based on ReAct prompting:
  • Knowledge Bases for Amazon Bedrock – Knowledge Bases for Amazon Bedrock provides fully managed RAG to supply the AI assistant with access to your data. In our use case, we uploaded device specifications into an Amazon Simple Storage Service (Amazon S3) bucket. It serves as the data source to the knowledge base.
  • Action groups – These are defined API schemas that invoke specific Lambda functions to interact with IoT devices and other AWS services.
  • Anthropic Claude v2.1 on Amazon Bedrock – This model interprets user queries and orchestrates the flow of tasks.
  • Amazon Titan Embeddings – This model serves as a text embeddings model, transforming natural language text—from single words to complex documents—into numerical vectors. This enables vector search capabilities, allowing the system to semantically match user queries with the most relevant knowledge base entries for effective search.

The solution is integrated with AWS services such as Lambda for running code in response to API calls, Athena for querying datasets, OpenSearch Service for searching through knowledge bases, and Amazon S3 for storage. These services work together to provide a seamless experience for IoT device operations management through natural language commands.

Benefits

This solution offers the following benefits:

• Implementation complexity:
  • Fewer lines of code are required, because Agents for Amazon Bedrock abstracts away much of the underlying complexity, reducing development effort
  • Managing vector databases like OpenSearch Service is simplified, because Knowledge Bases for Amazon Bedrock handles vectorization and storage
  • Integration with various AWS services is more streamlined through pre-defined action groups
• Developer experience:
  • The Amazon Bedrock console provides a user-friendly interface for prompt development, testing, and root cause analysis (RCA), enhancing the overall developer experience
• Agility and flexibility:
  • Agents for Amazon Bedrock allows for seamless upgrades to newer FMs (such as Claude 3.0) when they become available, so your solution stays up to date with the latest advancements
  • Service quotas and limitations are managed by AWS, reducing the overhead of monitoring and scaling infrastructure
• Security:
  • Amazon Bedrock is a fully managed service, adhering to AWS’s stringent security and compliance standards, potentially simplifying organizational security reviews

Although Agents for Amazon Bedrock offers a streamlined and managed solution for building conversational AI applications, some organizations may prefer an open source approach. In such cases, you can use frameworks like LangChain, which we discuss in the next section.

LangChain dynamic routing approach

LangChain is an open source framework that simplifies building conversational AI by allowing the integration of large language models (LLMs) and dynamic routing capabilities. With LangChain Expression Language (LCEL), developers can define the routing, which allows you to create non-deterministic chains where the output of a previous step defines the next step. Routing helps provide structure and consistency in interactions with LLMs.

For this post, we use the same example as the AI assistant for IoT device management. However, the main difference is that we need to handle the system prompts separately and treat each chain as a separate entity. The routing chain decides the destination chain based on the user’s input. The decision is made with the support of an LLM by passing the system prompt, chat history, and user’s question.

Solution overview

The following diagram illustrates the dynamic routing solution workflow.

The workflow consists of the following steps:

1. The user presents a question to the AI assistant. For example, “What are the max metrics for device 1009?”
2. An LLM evaluates each question along with the chat history from the same session to determine its nature and which subject area it falls under (such as SQL, action, search, or SME). The LLM classifies the input and the LCEL routing chain takes that input.
3. The router chain selects the destination chain based on the input, and the LLM is provided with the following system prompt:

"""Given the user question below, classify it as one of the candidate prompts. You may want to modify the input considering the chat history and the context of the question. 
Sometimes the user may just assume that you have the context of the conversation and may not provide a clear input. Hence, you are being provided with the chat history for more context. 
Respond with only a Markdown code snippet containing a JSON object formatted EXACTLY as specified below. 
Do not provide an explanation to your classification beside the Markdown, I just need to know your decision on which destination and next_inputs
<candidate prompt>
physics: Good for answering questions about physics
sql: sql: Good for querying sql from AWS Athena. User input may look like: get me max or min for device x?
lambdachain: Good to execute actions with Amazon Lambda like shutting down a device or turning off an engine User input can be like, shutdown device x, or terminate process y, etc.
rag: Good to search knowledgebase and retrieve information about devices and other related information. User question can be like: what do you know about device x?
default: if the input is not well suited for any of the candidate prompts above. this could be used to carry on the conversation and respond to queries like provide a summary of the conversation
</candidate prompt>"""

The LLM evaluates the user’s question along with the chat history to determine the nature of the query and which subject area it falls under. The LLM then classifies the input and outputs a JSON response in the following format:

<Markdown>
```json
{{
"destination": string  name of the prompt to use
"next_inputs": string  a potentially modified version of the original input
}}
```

The router chain uses this JSON response to invoke the corresponding destination chain. There are four subject-specific destination chains, each with its own system prompt:

1. SQL-related queries are sent to the SQL destination chain for database interactions. You can use LCEL to build the SQL chain.
2. Action-oriented questions invoke the custom Lambda destination chain for running operations. With LCEL, you can define your own custom function; in our case, it’s a function to run a predefined Lambda function to send an email with a device ID parsed. Example user input might be “Shut down device 1009.”
3. Search-focused inquiries proceed to the RAG destination chain for information retrieval.
4. SME-related questions go to the SME/expert destination chain for specialized insights.
5. Each destination chain takes the input and runs the necessary models or functions:
   1. The SQL chain uses Athena for running queries.
   2. The RAG chain uses OpenSearch Service for semantic search.
   3. The custom Lambda chain runs Lambda functions for actions.
   4. The SME/expert chain provides insights using the Amazon Bedrock model.
6. Responses from each destination chain are formulated into coherent insights by the LLM. These insights are then delivered to the user, completing the query cycle.
7. User input and responses are stored in Amazon DynamoDB to provide context to the LLM for the current session and from past interactions. The duration of persisted information in DynamoDB is controlled by the application.

Technical overview

The following diagram illustrates the architecture of the LangChain dynamic routing solution.

The web application is built on Streamlit hosted on Amazon ECS with Fargate, and it is accessed using an Application Load Balancer. We use Anthropic’s Claude v2.1 on Amazon Bedrock as our LLM. The web application interacts with the model using LangChain libraries. It also interacts with variety of other AWS services, such as OpenSearch Service, Athena, and DynamoDB to fulfill end-users’ needs.

Benefits

This solution offers the following benefits:

• Implementation complexity:
  • Although it requires more code and custom development, LangChain provides greater flexibility and control over the routing logic and integration with various components.
  • Managing vector databases like OpenSearch Service requires additional setup and configuration efforts. The vectorization process is implemented in code.
  • Integrating with AWS services may involve more custom code and configuration.
• Developer experience:
  • LangChain’s Python-based approach and extensive documentation can be appealing to developers already familiar with Python and open source tools.
  • Prompt development and debugging may require more manual effort compared to using the Amazon Bedrock console.
• Agility and flexibility:
  • LangChain supports a wide range of LLMs, allowing you to switch between different models or providers, fostering flexibility.
  • The open source nature of LangChain enables community-driven improvements and customizations.
• Security:
  • As an open source framework, LangChain may require more rigorous security reviews and vetting within organizations, potentially adding overhead.

Conclusion

Conversational AI assistants are transformative tools for streamlining operations and enhancing user experiences. This post explored two powerful approaches using AWS services: the managed Agents for Amazon Bedrock and the flexible, open source LangChain dynamic routing. The choice between these approaches hinges on your organization’s requirements, development preferences, and desired level of customization. Regardless of the path taken, AWS empowers you to create intelligent AI assistants that revolutionize business and customer interactions

Find the solution code and deployment assets in our GitHub repository, where you can follow the detailed steps for each conversational AI approach.

About the Authors

Ameer Hakme is an AWS Solutions Architect based in Pennsylvania. He collaborates with Independent Software Vendors (ISVs) in the Northeast region, assisting them in designing and building scalable and modern platforms on the AWS Cloud. An expert in AI/ML and generative AI, Ameer helps customers unlock the potential of these cutting-edge technologies. In his leisure time, he enjoys riding his motorcycle and spending quality time with his family.

Sharon Li is an AI/ML Solutions Architect at Amazon Web Services based in Boston, with a passion for designing and building Generative AI applications on AWS. She collaborates with customers to leverage AWS AI/ML services for innovative solutions.

Kawsar Kamal is a senior solutions architect at Amazon Web Services with over 15 years of experience in the infrastructure automation and security space. He helps clients design and build scalable DevSecOps and AI/ML solutions in the Cloud.

Read More

The Amazon EU Design and Construction (Amazon D&C) team is the engineering team designing and constructing Amazon warehouses. The team navigates a large volume of documents and locates the right information to make sure the warehouse design meets the highest standards. In the post A generative AI-powered solution on Amazon SageMaker to help Amazon EU Design and Construction, we presented a question answering bot solution using a Retrieval Augmented Generation (RAG) pipeline with a fine-tuned large language model (LLM) for Amazon D&C to efficiently retrieve accurate information from a large volume of unorganized documents, and provide timely and high-quality services in their construction projects. The Amazon D&C team implemented the solution in a pilot for Amazon engineers and collected user feedback.

In this post, we share how we analyzed the feedback data and identified limitations of accuracy and hallucinations RAG provided, and used the human evaluation score to train the model through reinforcement learning. To increase training samples for better learning, we also used another LLM to generate feedback scores. This method addressed the RAG limitation and further improved the bot response quality. We present the reinforcement learning process and the benchmarking results to demonstrate the LLM performance improvement. The solution uses Amazon SageMaker JumpStart as the core service for model deployment, fine-tuning, and reinforcement learning.

Collect feedback from Amazon engineers in a pilot project

After developing the solution described in A generative AI-powered solution on Amazon SageMaker to help Amazon EU Design and Construction, the Amazon D&C team deployed the solution and ran a pilot project with Amazon engineers. The engineers accessed the pilot system through a web application developed by Streamlit, connected with the RAG pipeline. In the pipeline, we used Amazon OpenSearch Service for the vector database, and deployed a fine-tuned Mistral-7B-Instruct model on Amazon SageMaker.

One of the key objectives of the pilot is to collect feedback from Amazon engineers and use the feedback to further reduce LLM hallucinations. To achieve this, we developed a feedback collection module in the UI, as shown in the following figure, and stored the web session information and user feedback in Amazon DynamoDB. Through the feedback collection UI, Amazon engineers can select from five satisfaction levels: strongly disagree, disagree, neutral, agree, and strongly agree, corresponding to the feedback scores from 1–5. They can also provide a better answer to the question or comment on why the LLM response is not satisfactory.

During the pilot, users provided 118 feedback responses. The following chart shows the distribution from negative to positive. We observed the incorrect answers marked by the negative feedback (disagree and strongly disagree) are 45% among the total feedback (53 out of 118). Some of the errors (about 10% of negative feedback and 7.5% of overall responses) can be addressed by user education and prompt engineering. Additionally, we can address the issue with the solution of LLM fine-tuning and reinforcement learning, described in the next section.

Some users simply left a note, such as “Great!” for the strongly agree answers, or “Doesn’t answer the question,” for the strongly disagree answers. Other users provided scores and explained how they justify the LLM answers in their notes. In the following example, the user specifically provided the adequate document and content to correct the LLM hallucination.

Improve bot response with supervised fine-tuning and reinforcement learning

The solution consists of three steps of fine-tuning:

1. Conduct supervised fine-tuning using labeled data. This method was described in A generative AI-powered solution on Amazon SageMaker to help Amazon EU Design and Construction.
2. Collect user feedback to label the question-answer pairs for further LLM tuning.
3. When the training data is ready, further tune the model using reinforcement learning from human feedback (RLHF).

RLHF is widely used throughout generative artificial intelligence (AI) and LLM applications. It incorporates human feedback in the rewards function and trains the model with a reinforcement learning algorithm to maximize rewards, which makes the model perform tasks more aligned with human goals. The following diagram shows the pipeline of the steps.

We tested the methodology using the Amazon D&C documents with a Mistral-7B model on SageMaker JumpStart.

Supervised fine-tuning

In the previous post, we demonstrated how the fine-tuned Falcon-7B model outperforms the RAG pipeline and improves the quality and accuracy of QA bot response. For this post, we performed supervised fine-tuning on the Mistral-7B model. The supervised fine-tuning used the PEFT/LoRA technique (LoRA_r = 512, LoRA_alpha = 1024) on 436,207,616 parameters (5.68% of the total 7,677,964,288 parameters). The training was conducted on a p3.8x node with 137 samples synthetically generated by LLM and validated by humans; the process is well converged after 20 epochs, as shown in the following figure.

The fine-tuned model was validated by 274 samples, and the inference results were compared with the reference answers by the semantic similarity score. The score is 0.8100, which is higher than the score of 0.6419 from the traditional RAG.

Collect human and AI feedback for reinforcement learning

For RLHF, a sufficient amount of high-quality training samples labeled by subject matter experts (SMEs) are essential. However, poor-quality human labels will likely cause worse model performance than the original model after RLHF training. SMEs’ time is a scarce resource in any organization; reviewing hundreds or thousands of LLM responses and providing feedback requires a significant time investment from SMEs that may not have a clear return on investment.

To address this challenge, we adopted the reinforcement learning from AI feedback (RLAIF) method. RLAIF employs an AI assistant (another LLM) to provide evaluation scores, rather than from humans. In this hybrid learning approach, the learning agent refines the actions not only based on the interaction with a human but also from feedback provided by another AI model. It is much more scalable to provide sufficient training data for reinforcement learning, and much less subjective because the training is not dependent on particular preferences from a small group of SMEs. The SMEs may still be involved, but only to supervise the process and examine the AI feedback quality. This tremendously reduces the SMEs’ workload, because the AI-generated scores and labels are much less complicated for SMEs to analyze, filter, and sort.

In this use case, we provide the following prompt to an Anthropic Claude 2 model to generate the evaluation scores as the training samples for the reinforcement learning step:

"""

Human:

I am going to provide you with a question, a reference answer, and an machine generated answer to the question.

question: {question}

reference answer: {reference_answer}

machine_answer: {machine_answer}

Please analyze the question, the reference answer, and the machine generated answer to determine if the machine generated answer semantically matches the reference answer.

First think through your answer inside of <thinking> tags, then assign a score between 0.0 and 5.0 based on how well the machine answer matches the reference answer. Score 5.0 means a completely match, score 0.0 means a completely mismatch. Answer the score inside of <scoring> tags.

Also provide the reason to give this score inside of <reason> tags.

Assistant: <thinking>

"""

The Claude 2 model generates an evaluation score and a comment on why the score is given, as shown in the following example.

Out of the 274 validation questions, the supervised fine-tuned model generated 159 responses that have AI scores greater than 4. We observed 60 answers with scores lower than 3; there is space to improve the overall response quality.

The Amazon Engineering SMEs validated this AI feedback and acknowledged the benefits of using AI scores. Without AI feedback, the SMEs would need some time to review and analyze each LLM response to identify the cut-off answers and hallucinations, and to judge whether the LLM is returning correct contents and key concepts. AI feedback provides AI scores automatically and enables the SMEs to use filtering, sorting, and grouping to validate the scores and identify trends in the responses. This reduces the average SME’s review time by 80%.

Reinforcement learning from human and AI feedback

When the training samples are ready, we use the proximal policy optimization (PPO) algorithm to perform reinforcement learning. PPO uses a policy gradient method, which takes small steps to update the policy in the learning process, so that the learning agents can reliably reach the optimal policy network. This makes the training process more stable and reduces the possibility of divergence.

During the training, first we use the human- and AI-labeled data to build a reward model, which will be used guide the weights update in the learning process. For this use case, we select a distilroberta-base reward model and train it by samples in the following format:

[Instruction, Chosen_response, Rejected_response]

The following is an example of a training record.

The reward model is trained with the learning rate 1e-5. As shown in the following chart, the training well converges after 10 epochs.

Then the reward model is used for reinforcement learning. For this use case, we use PEFT/LoRA to accelerate the learning. The LoRA_r is 32 and LoRA_alpha is 64. The PPO-based learning is on a rate of 1.41e-5 in three epochs. Reinforcement training is highly volatile and very sensitive to the label data and feedback data. We use the reward_mean metric, the average reward obtained from the environment by specifically monitoring the reward model, to observe the learning process.

After RLAIF, we can see the number of bot responses with positive scores (above 3) increase from 78.1% to 85.5%, and those with negative scores (below 3) decrease from 21.9% to 14.5%. In total, the average AI feedback score increases up to 8%, from 3.9 to 4.2.

You can see the improvement in the following example.

Conclusion

This post is a continuation of our work with the Amazon EU Design and Construction team as discussed in A generative AI-powered solution on Amazon SageMaker to help Amazon EU Design and Construction. In this post, we showed how we generated human and AI feedback data to fine-tune the Mistral-7B model with reinforcement learning. The model after RLAIF provided better performance for Amazon Engineering’s question answering bot, improved the AI feedback score by 8%. In the Amazon D&C team’s pilot project, using RLAIF reduced the validation workload for SMEs by an estimated 80%. As the next step, we will scale up this solution by connecting with Amazon Engineering’s data infrastructure, and design a framework to automate the continuous learning process with a human in the loop. We will also further improve the AI feedback quality by tuning the prompt template.

Through this process, we learned how to further improve the quality and performance of question answering tasks through RLHF and RLAIF.

• Human validation and augmentation are essential to provide accurate and responsible outputs from LLM. The human feedback can be used in RLHF to further improve the model response.
• RLAIF automates the evaluation and learning cycle. The AI-generated feedback is less subjective because it doesn’t depend on a particular preference from a small pool of SMEs.
• RLAIF is more scalable to improve the bot quality through continued reinforcement learning while minimizing the efforts required from SMEs. It is especially useful for developing domain-specific generative AI solutions within large organizations.
• This process should be done on a regular basis, especially when new domain data is available to be covered by the solution.

In this use case, we used SageMaker JumpStart to test multiple LLMs and experiment with multiple LLM training approaches. It significantly accelerates the AI feedback and learning cycle with maximized efficiency and quality. For your own project, you can introduce the human-in-the-loop approach to collect your users’ feedback, or generate AI feedback using another LLM. Then you can follow the three-step process defined in this post to fine-tune your models using RLHF and RLAIF. We recommend experimenting with the methods using SageMaker JumpStart to speed up the process.

About the Author

Yunfei Bai is a Senior Solutions Architect at AWS. With a background in AI/ML, data science, and analytics, Yunfei helps customers adopt AWS services to deliver business results. He designs AI/ML and data analytics solutions that overcome complex technical challenges and drive strategic objectives. Yunfei has a PhD in Electronic and Electrical Engineering. Outside of work, Yunfei enjoys reading and music.

Elad Dwek is a Construction Technology Manager at Amazon. With a background in construction and project management, Elad helps teams adopt new technologies and data-based processes to deliver construction projects. He identifies needs and solutions, and facilitates the development of the bespoke attributes. Elad has an MBA and a BSc in Structural Engineering. Outside of work, Elad enjoys yoga, woodworking, and traveling with his family.

Luca Cerabone is a Business Intelligence Engineer at Amazon. Drawing from his background in data science and analytics, Luca crafts tailored technical solutions to meet the unique needs of his customers, driving them towards more sustainable and scalable processes. Armed with an MSc in Data Science, Luca enjoys engaging in DIY projects, gardening and experimenting with culinary delights in his leisure moments.

Read More

In various industries, such as financial services, telecommunications, and healthcare, customers use a digital identity process, which usually involves several steps to verify end-users during online onboarding or step-up authentication. An example of one step that can be used is face search, which can help determine whether a new end-user’s face matches those associated with an existing account.

Building an accurate face search system involves several steps. The system must be able to detect human faces in images, extract the faces into vector representations, store face vectors in a database, and compare new faces against existing entries. Amazon Rekognition makes this effortless by giving you pre-trained models that are invoked via simple API calls.

Amazon Rekognition enables you to achieve very high face search accuracy with a single face image. In some cases, you can use multiple images of the same person’s face to create user vectors and improve accuracy even further. This is especially helpful when images have variations in lighting, poses, and appearances.

In this post, we demonstrate how to use the Amazon Rekognition Face Search APIs with user vectors to increase the similarity score for true matches and decrease the similarity score for true non-matches.

We compare the results of performing face matching with and without user vectors.

Amazon Rekognition face matching

Amazon Rekognition face matching enables measuring the similarity of a face vector extracted from one image to a face vector extracted from another image. A pair of face images is said to be a true match if both images contain the face of the same person, and a true non-match otherwise. Amazon Rekognition returns a score for the similarity of the source and target faces. The minimum similarity score is 0, implying very little similarity, and the maximum is 100.

For comparing a source face with a collection of target faces (1:N matching), Amazon Rekognition allows you to create a Collection object and populate it with faces from images using API calls.

When adding a face to a collection, Amazon Rekognition doesn’t store the actual image of the face but rather the face vector, a mathematical representation of the face. With the SearchFaces API, you can compare a source face with one or several collections of target faces.

In June 2023, AWS launched user vectors, a new capability that significantly improves face search accuracy by using multiple face images of a user. Now, you can create user vectors, which aggregate multiple face vectors of the same user. User vectors offer higher face search accuracy with more robust depictions, because they contain varying degrees of lighting, sharpness, pose, appearance, and more. This improves the accuracy compared to searching against individual face vectors.

In the following sections, we outline the process of using Amazon Rekognition user vectors. We guide you through creating a collection, storing face vectors in that collection, aggregating those face vectors into user vectors, and then comparing the results of searching against those individual face vectors and user vectors.

Solution overview

For this solution, we use an Amazon Rekognition collection of users, each with its associated indexed face vectors from a number of different images of faces for each user.

Let’s look at the workflow to build a collection with users and faces:

1. Create an Amazon Rekognition collection.
2. For each user, create a user in the collection.
3. For each image of the user, add the face to the collection (IndexFaces, which returns face ID corresponding to each face vector).
4. Associate all indexed face IDs with the user (this is necessary for user vectors).

Then, we will compare the following workflows:

Searching with a new given input image against individual face vectors in our collection:

1. Get all faces from an image (DetectFaces).
2. For each face, compare against individual faces in our collection (SearchFacesByImage).

Searching with a new given input image against user vectors in our collection:

1. Get all faces from an image (DetectFaces).
2. For each face, compare to the user vector (SearchUsersByImage).

Now let’s describe the solution in details.

Prerequisites

Add the following policy to your AWS Identity and Access Management (IAM) user or role. The policy grants you permission to the relevant Amazon Rekognition APIs and allows access to an Amazon Simple Storage Service (Amazon S3) bucket to store the images:

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Sid": "RekognitionPermissions",
            "Effect": "Allow",
            "Action": [
                "rekognition:CreateCollection",
                "rekognition:DeleteCollection",
                "rekognition:CreateUser",
                "rekognition:IndexFaces",
                "rekognition:DetectFaces",
                "rekognition:AssociateFaces",
                "rekognition:SearchUsersByImage",
                "rekognition:SearchFacesByImage"
            ],
            "Resource": "*"
        },
        {
            "Sid": "S3BucketPermissions",
            "Effect": "Allow",
            "Action": [
                "s3:GetObject",
                "s3:PutObject",
                "s3:ListBucket"
            ],
            "Resource": [
                "arn:aws:s3:::<replace_with_your_bucket>/*",
                "arn:aws:s3:::<replace_with_your_bucket>"
            ]
        }
    ]
}

Create an Amazon Rekognition collection and add users and faces

First, we create an S3 bucket to store users’ images. We organize the bucket by creating a folder for each user that contains their personal images. Our images folder looks like the following structure:

── images
│   ├── photo.jpeg
│   ├── Swami
│   │   ├── Swami1.jpeg
│   │   └── Swami2.jpeg
│   └── Werner
│       ├── Werner1.jpeg
│       ├── Werner2.jpeg
│       └── Werner3.jpeg

Our S3 bucket has a directory for each user that stores their images. There are currently two folders, and each contains several images. You can add more folders for your users, each containing one or more images to be indexed.

Next, we create our Amazon Rekognition collection. We have supplied helpers.py, which contains different methods that we use:

• create_collection – Create a new collection
• delete_collection – Delete a collection
• create_user – Create a new user in a collection
• add_faces_to_collection – Add faces to collection
• associate_faces – Associate face_ids to a user in a collection
• get_subdirs – Get all subdirectories under the S3 prefix
• get_files – Get all files under the S3 prefix

The following is an example method for creating an Amazon Rekognition collection:

import boto3
session = boto3.Session()
client = session.client('rekognition')

def create_collection(collection_id):
    try:
        # Create a collection
        print('Creating collection:' + collection_id)
        response = client.create_collection(CollectionId=collection_id)
        print('Collection ARN: ' + response['CollectionArn'])
        print('Status code: ' + str(response['StatusCode']))
        print('Done...')
    except client.exceptions.ResourceAlreadyExistsException:
        print('Resource already exits...')

Create the collection with the following code:

import helpers
collection_id = "faces-collection"
helpers.create_collection(collection_id)

Next, let’s add the face vectors into our collection and aggregate them into user vectors.

For each user in the S3 directory, we create a user vector in the collection. Then we index the face images for each user into the collection as individual face vectors, which generates face IDs. Lastly, we associate the face IDs to the appropriate user vector.

This creates two types of vectors in our collection:

• Individual face vectors
• User vectors, which are built based on the face vector IDs supplied using the method associate_faces

See the following code:

bucket = '<replace_with_your_bucket>'
prefix = 'images/'

# Get all the users directories from s3 containing the images
folder_list = helpers.get_subdirs(bucket, prefix)
print(f"Found users folders: {folder_list}")
print()

for user_id in folder_list:
    face_ids = []
    helpers.create_user(collection_id, user_id)
    # Get all files per user under the s3 user directory
    images = helpers.get_files(bucket, prefix + user_id + "/")
    print (f"Found images={images} for {user_id}")
    for image in images:
        face_id = helpers.add_faces_to_collection(bucket, image, collection_id)
        face_ids.append(face_id)
    helpers.associate_faces(collection_id, user_id, face_ids)
    print()

We use the following methods:

• get_subdirs – Returns a list of all the users’ directories. In our example, the value is [Swami,Werner].
• get_files – Returns all the images files under the S3 prefix for the user.
• face_ids – This is a list containing all the face IDs belonging to a user. We use this list when calling the AssociateFaces API.

As explained earlier, you can add more users by adding folders for them (the folder dictates the user ID) and add your images in that folder (no ordering is required for the files).

Now that our environment is set up and we have both individual face vectors and user vectors, let’s compare our search quality against each of them. To do that, we use a new photo with multiple people and attempt to match their faces against our collection, first against the individual face vectors and then against the user vectors.

Face search of image against a collection of individual face vectors

To search against our individual face vectors, we use the Amazon Rekognition SearchFacesByImage API. This function uses a source face image to search against individual face vectors in our collection and returns faces that match our defined similarity score threshold.

An important consideration is that the SearchFacesByImage API will only operate on the largest face detected in the image. If multiple faces are present, you need to crop each individual face and pass it separately to the method for identification.

For extracting faces details from an image (such as their location on the image), we use the Amazon Rekognition DetectFaces API.

The following detect_faces_in_image method detects faces in an image. For each face, it performs the following actions:

• Print its bounding box location
• Crop the face from the image and check if such face exists in the collection and print the user or ‘Unknown’
• Print the similarity score

The example Python code uses the Pillow library for doing the image manipulations (such as printing, drawing, and cropping).

We use a similarity score threshold of 99%, which is a common setting for identity verification use cases.

Run the following code:

import detect_users
from PIL import Image

# The image we would like to match faces against our collection.
file_key= "images/photo.jpeg"

img = detect_users.detect_faces_in_image(
    bucket, 
    file_key, 
    collection_id, 
    threshold=99
)
img.show() # or in Jupyter use display(img)

file_key is the S3 object key we want to match against our collection. We have supplied an example image (photo.jpeg) under the images folder.

The following image shows our results.

Using a threshold of 99%, only one person was identified. Dr. Werner Vogels was flagged as Unknown. If we run the same code using a lower threshold of 90 (set threshold=90), we get the following results.

Now we see Dr. Werner Vogel’s face has a similarity score of 96.86%. Next, let’s check if we can get the similarity score above our defined threshold by using user vectors.

Face search of image against a collection of user vectors

To search against our user vectors, we use the Amazon Rekognition SearchUsersByImage API. This function uses a source face image to search against user vectors in our collection and returns users that match our defined similarity score threshold.

The same consideration is relevant here – the SearchUsersByImage API will only operate on the largest face detected in the image. If there are multiple faces present, you need to crop each individual face and pass it separately to the method for identification.

For extracting faces details from an image (such as their location on the image), we use the Amazon Rekognition DetectFaces API.

The following detect_users_in_image method detects faces in an image. For each face, it performs the following actions:

• Print its bounding box location
• Crop the face from the image and check if such user face exists in our collection and print the user or ‘Unknown’
• Print the similarity score

See the following code:

import boto3
import io
import math
from PIL import Image, ImageDraw, ImageFont

def detect_users_in_image(bucket, key, collection_id, threshold=80):

    session = boto3.Session()
    client = session.client('rekognition')

    # Load image from S3 bucket
    s3_connection = boto3.resource('s3')
    s3_object = s3_connection.Object(bucket, key)
    s3_response = s3_object.get()

    stream = io.BytesIO(s3_response['Body'].read())
    image = Image.open(stream)

    # Call DetectFaces to find faces in image
    response = client.detect_faces(
        Image={'S3Object': {'Bucket': bucket, 'Name': key}},
        Attributes=['ALL']
    )

    imgWidth, imgHeight = image.size
    draw = ImageDraw.Draw(image)

    # Calculate and display bounding boxes for each detected face
    for faceDetail in response['FaceDetails']:
        print('The detected face is between ' + str(faceDetail['AgeRange']['Low'])
              + ' and ' + str(faceDetail['AgeRange']['High']) + ' years old')

        box = faceDetail['BoundingBox']
        left = imgWidth * box['Left']
        top = imgHeight * box['Top']
        width = imgWidth * box['Width']
        height = imgHeight * box['Height']

        print('Left: ' + '{0:.0f}'.format(left))
        print('Top: ' + '{0:.0f}'.format(top))
        print('Face Width: ' + "{0:.0f}".format(width))
        print('Face Height: ' + "{0:.0f}".format(height))

        points = (
            (left, top),
            (left + width, top),
            (left + width, top + height),
            (left, top + height),
            (left, top)
        )

        # Crop the face box and convert it to byte array
        face = image.crop((left, top, left + width, top + height))
        imgByteArr = image_to_byte_array(face, image.format)

        # Search for a user in our collection using the cropped image
        user_response = client.search_users_by_image(
            CollectionId=collection_id,
            Image={'Bytes': imgByteArr},
            UserMatchThreshold=threshold
        )
        # print (user_response)

        # Extract user id and the similarity from the response
        if (user_response['UserMatches']):
            similarity = user_response['UserMatches'][0]['Similarity']
            similarity = (math.trunc(similarity * 100) / 100) if isinstance(similarity, float) else similarity
            user_id = user_response['UserMatches'][0]['User']['UserId']
            print(f"User {user_id} was found, similarity of {similarity}%")
            print("")
        else:
            user_id = "Unknown"
            similarity = 0

        draw.line(points, fill='#00d400', width=4)
        font = ImageFont.load_default(size=25)
        draw.text((left, top - 30), user_id, fill='#00d400', font=font)
        if similarity > 0:
            draw.text((left, top + 1), str(similarity), fill='#00d400', font=font)

    return image

The function returns a modified image with the results that can be saved to Amazon S3 or printed. The function also outputs statistics about the estimated ages of the faces to the terminal.

Run the following code:

import detect_users
from PIL import Image

# The image we would like to match faces against our collection.
file_key= "images/photo.jpeg"

img = detect_users.detect_users_in_image(
    bucket, 
    file_key, 
    collection_id, 
    threshold=99
)
img.show() # or in Jupyter use display(img)

The following image shows our results.

The users that exist in our collection were identified correctly with high similarity (over 99%).

We were able to increase the similarity score by using three face vectors per user vector. As we increase the number of face vectors used, we expect the similarity score for true matches to also increase. You can use up to 100 face vectors per user vector.

An end-to-end example code can be found in the GitHub repository. It includes a detailed Jupyter notebook that you can run on Amazon SageMaker Studio (or other alternatives).

Clean up

To delete the collection, use the following code:

helpers.delete_collection(collection_id)

Conclusion

In this post, we presented how to use Amazon Rekognition user vectors to implement face search against a collection of users’ faces. We demonstrated how to improve face search accuracy by using multiple face images per user and compared it against individual face vectors. Additionally, we described how you can use the different Amazon Rekognition APIs to detect faces. The provided example code serves as a solid foundation for constructing a functional face search system.

For more information about Amazon Rekognition user vectors, refer to Searching faces in a collection. If you’re new to Amazon Rekognition, you can use our Free Tier, which lasts 12 months and includes processing 5,000 images per month and storing 1,000 user vector objects per month.

About the Authors

Arik Porat is a Senior Startups Solutions Architect at Amazon Web Services. He works with startups to help them build and design their solutions in the cloud, and is passionate about machine learning and container-based solutions. In his spare time, Arik likes to play chess and video games.

Eliran Efron is a Startups Solutions Architect at Amazon Web Services. Eliran is a data and compute enthusiast, assisting startups designing their system architectures. In his spare time, Eliran likes to build and race cars in Touring races and build IoT devices.

Read More

We are excited to announce two new capabilities in Amazon SageMaker Studio that will accelerate iterative development for machine learning (ML) practitioners: Local Mode and Docker support. ML model development often involves slow iteration cycles as developers switch between coding, training, and deployment. Each step requires waiting for remote compute resources to start up, which delays validating implementations and getting feedback on changes.

With Local Mode, developers can now train and test models, debug code, and validate end-to-end pipelines directly on their SageMaker Studio notebook instance without the need for spinning up remote compute resources. This reduces the iteration cycle from minutes down to seconds, boosting developer productivity. Docker support in SageMaker Studio notebooks enables developers to effortlessly build Docker containers and access pre-built containers, providing a consistent development environment across the team and avoiding time-consuming setup and dependency management.

Local Mode and Docker support offer a streamlined workflow for validating code changes and prototyping models using local containers running on a SageMaker Studio notebook

instance. In this post, we guide you through setting up Local Mode in SageMaker Studio, running a sample training job, and deploying the model on an Amazon SageMaker endpoint from a SageMaker Studio notebook.

SageMaker Studio Local Mode

SageMaker Studio introduces Local Mode, enabling you to run SageMaker training, inference, batch transform, and processing jobs directly on your JupyterLab, Code Editor, or SageMaker Studio Classic notebook instances without requiring remote compute resources. Benefits of using Local Mode include:

• Instant validation and testing of workflows right within integrated development environments (IDEs)
• Faster iteration through local runs for smaller-scale jobs to inspect outputs and identify issues early
• Improved development and debugging efficiency by eliminating the wait for remote training jobs
• Immediate feedback on code changes before running full jobs in the cloud

The following figure illustrates the workflow using Local Mode on SageMaker.

To use Local Mode, set instance_type='local' when running SageMaker Python SDK jobs such as training and inference. This will run them on the instances used by your SageMaker Studio IDEs instead of provisioning cloud resources.

Although certain capabilities such as distributed training are only available in the cloud, Local Mode removes the need to switch contexts for quick iterations. When you’re ready to take advantage of the full power and scale of SageMaker, you can seamlessly run your workflow in the cloud.

Docker support in SageMaker Studio

SageMaker Studio now also enables building and running Docker containers locally on your SageMaker Studio notebook instance. This new feature allows you to build and validate Docker images in SageMaker Studio before using them for SageMaker training and inference.

The following diagram illustrates the high-level Docker orchestration architecture within SageMaker Studio.

With Docker support in SageMaker Studio, you can:

• Build Docker containers with integrated models and dependencies directly within SageMaker Studio
• Eliminate the need for external Docker build processes to simplify image creation
• Run containers locally to validate functionality before deploying models to production
• Reuse local containers when deploying to SageMaker for training and hosting

Although some advanced Docker capabilities like multi-container and custom networks are not supported as of this writing, the core build and run functionality is available to accelerate developing containers for bring your own container (BYOC) workflows.

Prerequisites

To use Local Mode in SageMaker Studio applications, you must complete the following prerequisites:

• For pulling images from Amazon Elastic Container Registry (Amazon ECR), the account hosting the ECR image must provide access permission to the user’s Identity and Access Management (IAM) role. The domain’s role must also allow Amazon ECR access.
• To enable Local Mode and Docker capabilities, you must set the EnableDockerAccess parameter to true for the domain’s DockerSettings using the AWS Command Line Interface (AWS CLI). This allows users in the domain to use Local Mode and Docker features. By default, Local Mode and Docker are disabled in SageMaker Studio. Any existing SageMaker Studio apps will need to be restarted for the Docker service update to take effect. The following is an example AWS CLI command for updating a SageMaker Studio domain:

aws sagemaker --region <REGION> 
update-domain --domain-id <DOMAIN-ID> 
--domain-settings-for-update '{"DockerSettings": {"EnableDockerAccess": "ENABLED"}}'

• You need to update the SageMaker IAM role in order to be able to push Docker images to Amazon ECR:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "ecr:CompleteLayerUpload",
        "ecr:UploadLayerPart",
        "ecr:InitiateLayerUpload",
        "ecr:BatchCheckLayerAvailability",
        "ecr:PutImage"
      ],
      "Resource": "arn:aws:ecr:us-east-2:123456789012:repository/<repositoryname>"
    },
    {
      "Effect": "Allow",
      "Action": "ecr:GetAuthorizationToken",
      "Resource": "*"
    }
  ]
}

Run Python files in SageMaker Studio spaces using Local Mode

SageMaker Studio JupyterLab and Code Editor (based on Code-OSS, Visual Studio Code – Open Source), extends SageMaker Studio so you can write, test, debug, and run your analytics and ML code using the popular lightweight IDE. For more details on how to get started with SageMaker Studio IDEs, refer to Boost productivity on Amazon SageMaker Studio: Introducing JupyterLab Spaces and generative AI tools and New – Code Editor, based on Code-OSS VS Code Open Source now available in Amazon SageMaker Studio. Complete the following steps:

• Create a new Code Editor or JupyterLab space called my-sm-code-editor-space or my-sm-jupyterlab-space, respectively.
• Choose Create space. 
• Choose the ml.m5.large instance and set storage to 32 GB.
• Choose Run space. 
• Open the JupyterLab or Code Editor space and clone the GitHub repo. 
• Clone the GitHub repo, with /home/sagemaker-user/ as the target folder.

• Create a new terminal. 
• Install the Docker CLI and Docker Compose plugin following the instructions in the following GitHub repo. If chained commands fail, run the commands one at a time.

You must update the SageMaker SDK to the latest version.

• Run pip install sagemaker -Uq in the terminal.

For Code Editor only, you need to set the Python environment to run in the current terminal.

• In Code Editor, on the File menu¸ choose Preferences and Settings.

• Search for and select Terminal: Execute in File Dir.

• In Code Editor or JupyterLab, open the scikit_learn_script_mode_local_training_and_serving folder and run the scikit_learn_script_mode_local_training_and_serving.py file.

You can run the script by choosing Run in Code Editor or using the CLI in a JupyterLab terminal. You will be able to see how the model is trained locally. Then you deploy the model to a SageMaker endpoint locally, and calculate the root mean square error (RMSE).

Simulate training and inference in SageMaker Studio Classic using Local Mode

You can also use a notebook in SageMaker Studio Classic to run a small-scale training job on CIFAR10 using Local Mode, deploy the model locally, and perform inference.

Set up your notebook

To set up the notebook, complete the following steps:

• Open SageMaker Studio Classic and clone the following GitHub repo.

• Open the pytorch_local_mode_cifar10.ipynb notebook in blog/pytorch_cnn_cifar10.

• For Image, choose PyTorch 2.1.0 Python 3.10 CPU Optimized.

Confirm that your notebook shows the correct instance and kernel selection.

• Open a terminal by choosing Launch Terminal in the current SageMaker image.

• Install the Docker CLI and Docker Compose plugin following the instructions in the following GitHub repo.

Because you’re using Docker from SageMaker Studio Classic, remove sudo when running commands because the terminal already runs under superuser. For SageMaker Studio Classic, the installation commands depend on the SageMaker Studio app image OS. For example, DLC-based framework images are Ubuntu based, in which the following instructions would work. However, for a Debian-based image like DataScience Images, you must follow the instructions in the following GitHub repo. If chained commands fail, run the commands one at a time. You should see the Docker version displayed.

• Leave the terminal window open, go back to the notebook, and start running it cell by cell.

Make sure to run the cell with pip install -U sagemaker so you’re using the latest version of the SageMaker Python SDK.

Local training

When you start running the local SageMaker training job, you will see the following log lines:

INFO:sagemaker.local.image:'Docker Compose' found using Docker CLI.
INFO:sagemaker.local.local_session:Starting training job

This indicates that the training was running locally using Docker.

Be patient while the pytorch-training:2.1-cpu-py310 Docker image is pulled. Due to its large size (5.2 GB), it could take a few minutes.

Docker images will be stored in the SageMaker Studio app instance’s root volume, which is not accessible to end-users. The only way to access and interact with Docker images is via the exposed Docker API operations.

From a user confidentiality standpoint, the SageMaker Studio platform never accesses or stores user-specific images.

When the training is complete, you’ll be able to see the following success log lines:

8zlz1zbfta-sagemaker-local exited with code 0
Aborting on container exit...
Container 8zlz1zbfta-sagemaker-local  Stopping
Container 8zlz1zbfta-sagemaker-local  Stopped
INFO:sagemaker.local.image:===== Job Complete =====

Local inference

Complete the following steps:

• Deploy the SageMaker endpoint using SageMaker Local Mode.

Be patient while the pytorch-inference:2.1-cpu-py310 Docker image is pulled. Due to its large size (4.32 GB), it could take a few minutes.

• Invoke the SageMaker endpoint deployed locally using the test images.

You will be able to see the predicted classes: frog, ship, car, and plane:

Predicted:  frog ship  car plane

• Because the SageMaker Local endpoint is still up, navigate back to the open terminal window and list the running containers:

docker ps

You’ll be able to see the running pytorch-inference:2.1-cpu-py310 container backing the SageMaker endpoint.

• To shut down the SageMaker local endpoint and stop the running container, because you can only run one local endpoint at a time, run the cleanup code.

• To make sure the Docker container is down, you can navigate to the opened terminal window, run docker ps, and make sure there are no running containers.
• If you see a container running, run docker stop <CONTAINER_ID> to stop it.

Tips for using SageMaker Local Mode

If you’re using SageMaker for the first time, refer to Train machine learning models. To learn more about deploying models for inference with SageMaker, refer to Deploy models for inference.

Keep in mind the following recommendations:

• Print input and output files and folders to understand dataset and model loading
• Use 1–2 epochs and small datasets for quick testing
• Pre-install dependencies in a Dockerfile to optimize environment setup
• Isolate serialization code in endpoints for debugging

Configure Docker installation as a Lifecycle Configuration

You can define the Docker install process as a Lifecycle Configuration (LCC) script to simplify setup each time a new SageMaker Studio space starts. LCCs are scripts that SageMaker runs during events like space creation. Refer to the JupyterLab, Code Editor, or SageMaker Studio Classic LCC setup (using docker install cli as reference) to learn more.

Build and test custom Docker images in SageMaker Studio spaces

In this step, you install Docker inside the JupyterLab (or Code Editor) app space and use Docker to build, test, and publish custom Docker images with SageMaker Studio spaces. Spaces are used to manage the storage and resource needs of some SageMaker Studio applications. Each space has a 1:1 relationship with an instance of an application. Every supported application that is created gets its own space. To learn more about SageMaker spaces, refer to Boost productivity on Amazon SageMaker Studio: Introducing JupyterLab Spaces and generative AI tools. Make sure you provision a new space with at least 30 GB of storage to allow sufficient storage for Docker images and artifacts.

Install Docker inside a space

To install the Docker CLI and Docker Compose plugin inside a JupyterLab space, run the commands in the following GitHub repo. SageMaker Studio only supports Docker version 20.10.X.

Build Docker images

To confirm that Docker is installed and working inside your JupyterLab space, run the following code:

# to verify docker service
sagemaker-user@default:~$ docker version
Client: Docker Engine - Community
Version:           24.0.7
API version:       1.41 (downgraded from 1.43)
Go version:        go1.20.10
Git commit:        afdd53b
Built:             Thu Oct 26 09:07:41 2023
OS/Arch:           linux/amd64
Context:           default

Server:
Engine:
Version:          20.10.25
API version:      1.41 (minimum version 1.12)
Go version:       go1.20.10
Git commit:       5df983c
Built:            Fri Oct 13 22:46:59 2023
OS/Arch:          linux/amd64
Experimental:     false
containerd:
Version:          1.7.2
GitCommit:        0cae528dd6cb557f7201036e9f43420650207b58
runc:
Version:          1.1.7
GitCommit:        f19387a6bec4944c770f7668ab51c4348d9c2f38
docker-init:
Version:          0.19.0
GitCommit:        de40ad0

To build a custom Docker image inside a JupyterLab (or Code Editor) space, complete the following steps:

• Create an empty Dockerfile:

touch Dockerfile

• Edit the Dockerfile with the following commands, which create a simple flask web server image from the base python:3.10.13-bullseye image hosted on Docker Hub:

# Use the specified Python base image
FROM python:3.10.13-bullseye

# Create a code dir
RUN mkdir /code/

# Set the working directory in the container
WORKDIR /code

# Upgrade pip and install required packages
RUN python3 -m pip install --upgrade pip && 
python3 -m pip install flask

# Copy the app.py file to the container
COPY app.py /code/

# Set the command to run the app
ENTRYPOINT ["python", "app.py"]

The following code shows the contents of an example flask application file app.py:

from flask import Flask, jsonify

app = Flask(__name__)

@app.route('/')
def hello():
return jsonify({"response": "Hello"})

if __name__ == '__main__':
app.run(host='0.0.0.0', port=6006)

Additionally, you can update the reference Dockerfile commands to include packages and artifacts of your choice.

• Build a Docker image using the reference Dockerfile:

docker build --network sagemaker --tag myflaskapp:v1 --file ./Dockerfile .

Include --network sagemaker in your docker build command, otherwise the build will fail. Containers can’t be run in Docker default bridge or custom Docker networks. Containers are run in same network as the SageMaker Studio application container. Users can only use sagemaker for the network name.

• When your build is complete, validate if the image exists. Re-tag the build as an ECR image and push. If you run into permission issues, run the aws ecr get-login-password… command and try to rerun the Docker push/pull:

sagemaker-user@default:~$ docker image list
REPOSITORY      TAG       IMAGE ID       CREATED          SIZE
myflaskapp      v1        d623f1538f20   27 minutes ago   489MB

sagemaker-user@default:~$ docker tag myflaskapp:v1 123456789012.dkr.ecr.us-east-2.amazonaws.com/myflaskapp:v1

sagemaker-user@default:~$ docker image list
REPOSITORY                                                  TAG       IMAGE ID       CREATED          SIZE
123456789012.dkr.ecr.us-east-2.amazonaws.com/myflaskapp     latest    d623f1538f20   27 minutes ago   489MB
myflaskapp                                                  v1        d623f1538f20   27 minutes ago   489MB

sagemaker-user@default:~$ aws ecr get-login-password --region region | docker login --username AWS --password-stdin aws_account_id.dkr.ecr.region.amazonaws.com

sagemaker-user@default:~$ docker push 123456789012.dkr.ecr.us-east-2.amazonaws.com/myflaskapp:latest

Test Docker images

Having Docker installed inside a JupyterLab (or Code Editor) SageMaker Studio space allows you to test pre-built or custom Docker images as containers (or containerized applications). In this section, we use the docker run command to provision Docker containers inside a SageMaker Studio space to test containerized workloads like REST web services and Python scripts. Complete the following steps:

• Check if the image you’re testing exists on the space’s Amazon Elastic Block Store (Amazon EBS) volume:

sagemaker-user@default:~$ docker image list
REPOSITORY                                                  TAG       IMAGE ID       CREATED       SIZE

• If the test image doesn’t exist, run docker pull to pull the image into your local machine:

sagemaker-user@default:~$ docker pull 123456789012.dkr.ecr.us-east-2.amazonaws.com/myflaskapp:v1

• If you encounter authentication issues, run the following commands:

aws ecr get-login-password --region region | docker login --username AWS --password-stdin aws_account_id.dkr.ecr.region.amazonaws.com

• Create a container to test your workload:

docker run --network sagemaker 123456789012.dkr.ecr.us-east-2.amazonaws.com/myflaskapp:v1

This spins up a new container instance and runs the application defined using Docker’s ENTRYPOINT:

sagemaker-user@default:~$ docker run --network sagemaker 905418447590.dkr.ecr.us-east-2.amazonaws.com/myflaskapp:v1
* Serving Flask app 'app'
* Debug mode: off
WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead.
* Running on all addresses (0.0.0.0)
* Running on http://127.0.0.1:6006
* Running on http://169.255.255.2:6006

• To test if your web endpoint is active, navigate to the URL https://<sagemaker-space-id>.studio.us-east-2.sagemaker.aws/jupyterlab/default/proxy/6006/.

You should see a JSON response similar to following screenshot.

Clean up

To avoid incurring unnecessary charges, delete the resources that you created while running the examples in this post:

1. In your SageMaker Studio domain, choose Studio Classic in the navigation pane, then choose Stop.
2. In your SageMaker Studio domain, choose JupyterLab or Code Editor in the navigation pane, choose your app, and then choose Stop.

Conclusion

SageMaker Studio Local Mode and Docker support empower developers to build, test, and iterate on ML implementations faster without leaving their workspace. By providing instant access to test environments and outputs, these capabilities optimize workflows and improve productivity. Try out SageMaker Studio Local Model and Docker support using our quick onboard feature, which allows you to spin up a new domain for single users within minutes. Share your thoughts in the comments section!

About the Authors

Shweta Singh is a Senior Product Manager in the Amazon SageMaker Machine Learning (ML) platform team at AWS, leading SageMaker Python SDK. She has worked in several product roles in Amazon for over 5 years. She has a Bachelor of Science degree in Computer Engineering and Masters of Science in Financial Engineering, both from New York University

Eitan Sela is a Generative AI and Machine Learning Specialist Solutions Architect ta AWS. He works with AWS customers to provide guidance and technical assistance, helping them build and operate Generative AI and Machine Learning solutions on AWS. In his spare time, Eitan enjoys jogging and reading the latest machine learning articles.

Pranav Murthy is an AI/ML Specialist Solutions Architect at AWS. He focuses on helping customers build, train, deploy and migrate machine learning (ML) workloads to SageMaker. He previously worked in the semiconductor industry developing large computer vision (CV) and natural language processing (NLP) models to improve semiconductor processes using state of the art ML techniques. In his free time, he enjoys playing chess and traveling. You can find Pranav on LinkedIn.

Mufaddal Rohawala is a Software Engineer at AWS. He works on the SageMaker Python SDK library for Amazon SageMaker. In his spare time, he enjoys travel, outdoor activities and is a soccer fan.

Read More

We introduced Amazon Bedrock to the world a little over a year ago, delivering an entirely new way to build generative artificial intelligence (AI) applications. With the broadest selection of first- and third-party foundation models (FMs) as well as user-friendly capabilities, Amazon Bedrock is the fastest and easiest way to build and scale secure generative AI applications. Now tens of thousands of customers are using Amazon Bedrock to build and scale impressive applications. They are innovating quickly, easily, and securely to advance their AI strategies. And we’re supporting their efforts by enhancing Amazon Bedrock with exciting new capabilities including even more model choice and features that make it easier to select the right model, customize the model for a specific use case, and safeguard and scale generative AI applications.

Customers across diverse industries from finance to travel and hospitality to healthcare to consumer technology are making remarkable progress. They are realizing real business value by quickly moving generative AI applications into production to improve customer experiences and increase operational efficiency. Consider the New York Stock Exchange (NYSE), the world’s largest capital market processing billions of transactions each day. NYSE is leveraging Amazon Bedrock’s choice of FMs and cutting-edge AI generative capabilities across several use cases, including the processing of thousands of pages of regulations to provide answers in easy-to-understand language

Global airline United Airlines modernized their Passenger Service System to translate legacy passenger reservation codes into plain English so that agents can provide swift and efficient customer support. LexisNexis Legal & Professional, a leading global provider of information and analytics, developed a personalized legal generative AI assistant on Lexis+ AI. LexisNexis customers receive trusted results two times faster than the nearest competing product and can save up to five hours per week for legal research and summarization. And HappyFox, an online help desk software, selected Amazon Bedrock for its security and performance, boosting the efficiency of its AI-powered automated ticket system in its customer support solution by 40% and agent productivity by 30%.

And across Amazon, we are continuing to innovate with generative AI to deliver more immersive, engaging experiences for our customers. Just last week Amazon Music announced Maestro. Maestro is an AI playlist generator powered by Amazon Bedrock that gives Amazon Music subscribers an easier, more fun way to create playlists based on prompts. Maestro is now rolling out in beta to a small number of U.S. customers on all tiers of Amazon Music.

With Amazon Bedrock, we’re focused on the key areas that customers need to build production-ready, enterprise-grade generative AI applications at the right cost and speed. Today I’m excited to share new features that we’re announcing across the areas of model choice, tools for building generative AI applications, and privacy and security.

1. Amazon Bedrock expands model choice with Llama 3 models and helps you find the best model for your needs

In these early days, customers are still learning and experimenting with different models to determine which ones to use for various purposes. They want to be able to easily try the latest models, and test which capabilities and features will give them the best results and cost characteristics for their use cases. The majority of Amazon Bedrock customers use more than one model, and Amazon Bedrock provides the broadest selection of first- and third-party large language models (LLMs) and other FMs.  This includes models from AI21 labs, Anthropic, Cohere, Meta, Mistral AI, and Stability AI, as well as our own Amazon Titan models. In fact, Joel Hron, head of AI and Thomson Reuters Labs at Thomson Reuters recently said this about their adoption of Amazon Bedrock, “Having the ability to use a diverse range of models as they come out was a key driver for us, especially given how quickly this space is evolving.” The cutting-edge models of the Mistral AI model family including Mistral 7B, Mixtral 8x7B, and Mistral Large have customers excited about their high performance in text generation, summarization, Q&A, and code generation. Since we introduced the Anthropic Claude 3 model family, thousands of customers have experienced how Claude 3 Haiku, Sonnet, and Opus have established new benchmarks across cognitive tasks with unrivaled intelligence, speed, and cost-efficiency. After the initial evaluation using Claude 3 Haiku and Opus in Amazon Bedrock, BlueOcean.ai, a brand intelligence platform, saw a cost reduction of over 50% when they were able to consolidate four separate API calls into a single, more efficient call.

Masahiro Oba, General Manager, Group Federated Governance of DX Platform at Sony Group corporation shared,

“While there are many challenges with applying generative AI to the business, Amazon Bedrock’s diverse capabilities help us to tailor generative AI applications to Sony’s business. We are able to take advantage of not only the powerful LLM capabilities of Claude 3, but also capabilities that help us safeguard applications at the enterprise-level. I’m really proud to be working with the Bedrock team to further democratize generative AI within the Sony Group.”

I recently sat down with Aaron Linsky, CTO of Artificial Investment Associate Labs at Bridgewater Associates, a premier asset management firm, where they are using generative AI to enhance their “Artificial Investment Associate,” a major leap forward for their customers. It builds on their experience of giving rules-based expert advice for investment decision-making. With Amazon Bedrock, they can use the best available FMs, such as Claude 3, for different tasks-combining fundamental market understanding with the flexible reasoning capabilities of AI. Amazon Bedrock allows for seamless model experimentation, enabling Bridgewater to build a powerful, self-improving investment system that marries systematic advice with cutting-edge capabilities–creating an evolving, AI-first process.

To bring even more model choice to customers, today, we are making Meta Llama 3 models available in Amazon Bedrock. Llama 3’s Llama 3 8B and Llama 3 70B models are designed for building, experimenting, and responsibly scaling generative AI applications. These models were significantly improved from the previous model architecture, including scaling up pretraining, as well as instruction fine-tuning approaches. Llama 3 8B excels in text summarization, classification, sentiment analysis, and translation, ideal for limited resources and edge devices. Llama 3 70B shines in content creation, conversational AI, language understanding, R&D, enterprises, accurate summarization, nuanced classification/sentiment analysis, language modeling, dialogue systems, code generation, and instruction following. Read more about Meta Llama 3 now available in Amazon Bedrock.

We are also announcing support coming soon for Cohere’s Command R and Command R+ enterprise FMs. These models are highly scalable and optimized for long-context tasks like retrieval-augmented generation (RAG) with citations to mitigate hallucinations, multi-step tool use for automating complex business tasks, and support for 10 languages for global operations. Command R+ is Cohere’s most powerful model optimized for long-context tasks, while Command R is optimized for large-scale production workloads. With the Cohere models coming soon in Amazon Bedrock, businesses can build enterprise-grade generative AI applications that balance strong accuracy and efficiency for day-to-day AI operations beyond proof-of-concept.

Amazon Titan Image Generator now generally available and Amazon Titan Text Embeddings V2 coming soon

In addition to adding the most capable 3P models, Amazon Titan Image Generator is generally available today. With Amazon Titan Image Generator, customers in industries like advertising, e-commerce, media, and entertainment can efficiently generate realistic, studio-quality images in large volumes and at low cost, utilizing natural language prompts. They can edit generated or existing images using text prompts, configure image dimensions, or specify the number of image variations to guide the model. By default, every image produced by Amazon Titan Image Generator contains an invisible watermark, which aligns with AWS’s commitment to promoting responsible and ethical AI by reducing the spread of misinformation. The Watermark Detection feature identifies images created by Image Generator, and is designed to be tamper-resistant, helping increase transparency around AI-generated content. Watermark Detection helps mitigate intellectual property risks and enables content creators, news organizations, risk analysts, fraud-detection teams, and others, to better identify and mitigate dissemination of misleading AI-generated content. Read more about Watermark Detection for Titan Image Generator.

Coming soon, Amazon Titan Text Embeddings V2 efficiently delivers more relevant responses for critical enterprise use cases like search. Efficient embeddings models are crucial to performance when leveraging RAG to enrich responses with additional information. Embeddings V2 is optimized for RAG workflows and provides seamless integration with Knowledge Bases for Amazon Bedrock to deliver more informative and relevant responses efficiently. Embeddings V2 enables a deeper understanding of data relationships for complex tasks like retrieval, classification, semantic similarity search, and enhancing search relevance. Offering flexible embedding sizes of 256, 512, and 1024 dimensions, Embeddings V2 prioritizes cost reduction while retaining 97% of the accuracy for RAG use cases, out-performing other leading models. Additionally, the flexible embedding sizes cater to diverse application needs, from low-latency mobile deployments to high-accuracy asynchronous workflows.

New Model Evaluation simplifies the process of accessing, comparing, and selecting LLMs and FMs

Choosing the appropriate model is a critical first step toward building any generative AI application. LLMs can vary drastically in performance based on the task, domain, data modalities, and other factors. For example, a biomedical model is likely to outperform general healthcare models in specific medical contexts, whereas a coding model may face challenges with natural language processing tasks. Using an excessively powerful model could lead to inefficient resource usage, while an underpowered model might fail to meet minimum performance standards – potentially providing incorrect results. And selecting an unsuitable FM at a project’s onset could undermine stakeholder confidence and trust.

With so many models to choose from, we want to make it easier for customers to pick the right one for their use case.

Amazon Bedrock’s Model Evaluation tool, now generally available, simplifies the selection process by enabling benchmarking and comparison against specific datasets and evaluation metrics, ensuring developers select the model that best aligns with their project goals. This guided experience allows developers to evaluate models across criteria tailored to each use case. Through Model Evaluation, developers select candidate models to assess – public options, imported custom models, or fine-tuned versions. They define relevant test tasks, datasets, and evaluation metrics, such as accuracy, latency, cost projections, and qualitative factors. Read more about Model Evaluation in Amazon Bedrock.

The ability to select from the top-performing FMs in Amazon Bedrock has been extremely beneficial for Elastic Security. James Spiteri, Director of Product Management at Elastic shared,

“With just a few clicks, we can assess a single prompt across multiple models simultaneously. This model evaluation functionality enables us to compare the outputs, metrics, and associated costs across different models, allowing us to make an informed decision on which model would be most suitable for what we are trying to accomplish. This has significantly streamlined our process, saving us a considerable amount of time in deploying our applications to production.”

2. Amazon Bedrock offers capabilities to tailor generative AI to your business needs

While models are incredibly important, it takes more than a model to build an application that is useful for an organization. That’s why Amazon Bedrock has capabilities to help you easily tailor generative AI solutions to specific use cases. Customers can use their own data to privately customize applications through fine-tuning or by using Knowledge Bases for a fully managed RAG experience to deliver more relevant, accurate, and customized responses. Agents for Amazon Bedrock allows developers to define specific tasks, workflows, or decision-making processes, enhancing control and automation while ensuring consistent alignment with an intended use case. Starting today, you can now use Agents with Anthropic Claude 3 Haiku and Sonnet models. We are also introducing an updated AWS console experience, supporting a simplified schema and return of control to make it easy for developers to get started. Read more about Agents for Amazon Bedrock, now faster and easier to use.

With new Custom Model Import, customers can leverage the full capabilities of Amazon Bedrock with their own models

All these features are essential to building generative AI applications, which is why we wanted to make them available to even more customers including those who have already invested significant resources in fine-tuning LLMs with their own data on different services or in training custom models from scratch. Many customers have customized models available on Amazon SageMaker, which provides the broadest array of over 250 pre-trained FMs. These FMs include cutting-edge models such as Mistral, Llama2, CodeLlama, Jurassic-2, Jamba, pplx-7B, 70B, and the impressive Falcon 180B. Amazon SageMaker helps with getting data organized and fine-tuned, building scalable and efficient training infrastructure, and then deploying models at scale in a low latency, cost-efficient manner. It has been a game changer for developers in preparing their data for AI, managing experiments, training models faster (e.g. Perplexity AI trains models 40% faster in Amazon SageMaker), lowering inference latency (e.g. Workday has reduced inference latency by 80% with Amazon SageMaker), and improving developer productivity (e.g. NatWest reduced its time-to-value for AI from 12-18 months to under seven months using Amazon SageMaker). However, operationalizing these customized models securely and integrating them into applications for specific business use cases still has challenges.

That is why today we’re introducing Amazon Bedrock Custom Model Import, which enables organizations to leverage their existing AI investments along with Amazon Bedrock’s capabilities. With Custom Model Import, customers can now import and access their own custom models built on popular open model architectures including Flan-T5, Llama, and Mistral, as a fully managed application programming interface (API) in Amazon Bedrock. Customers can take models that they customized on Amazon SageMaker, or other tools, and easily add them to Amazon Bedrock. After an automated validation, they can seamlessly access their custom model, as with any other model in Amazon Bedrock. They get all the same benefits, including seamless scalability and powerful capabilities to safeguard their applications, adherence to responsible AI principles – as well as the ability to expand a model’s knowledge base with RAG, easily create agents to complete multi-step tasks, and carry out fine tuning to keep teaching and refining models. All without needing to manage the underlying infrastructure.

With this new capability, we’re making it easy for organizations to choose a combination of Amazon Bedrock models and their own custom models while maintaining the same streamlined development experience. Today, Amazon Bedrock Custom Model Import is available in preview and supports three of the most popular open model architectures and with plans for more in the future. Read more about Custom Model Import for Amazon Bedrock.

ASAPP is a generative AI company with a 10-year history of building ML models.

“Our conversational generative AI voice and chat agent leverages these models to redefine the customer service experience. To give our customers end to end automation, we need LLM agents, knowledge base, and model selection flexibility. With Custom Model Import, we will be able to use our existing custom models in Amazon Bedrock. Bedrock will allow us to onboard our customers faster, increase our pace of innovation, and accelerate time to market for new product capabilities.”

– Priya Vijayarajendran, President, Technology.

3. Amazon Bedrock provides a secure and responsible foundation to implement safeguards easily

As generative AI capabilities progress and expand, building trust and addressing ethical concerns becomes even more important. Amazon Bedrock addresses these concerns by leveraging AWS’s secure and trustworthy infrastructure with industry-leading security measures, robust data encryption, and strict access controls.

Guardrails for Amazon Bedrock, now generally available, helps customers prevent harmful content and manage sensitive information within an application.

We also offer Guardrails for Amazon Bedrock, which is now generally available. Guardrails offers industry-leading safety protection, giving customers the ability to define content policies, set application behavior boundaries, and implement safeguards against potential risks. Guardrails for Amazon Bedrock is the only solution offered by a major cloud provider that enables customers to build and customize safety and privacy protections for their generative AI applications in a single solution. It helps customers block as much as 85% more harmful content than protection natively provided by FMs on Amazon Bedrock. Guardrails provides comprehensive support for harmful content filtering and robust personal identifiable information (PII) detection capabilities. Guardrails works with all LLMs in Amazon Bedrock as well as fine-tuned models, driving consistency in how models respond to undesirable and harmful content. You can configure thresholds to filter content across six categories – hate, insults, sexual, violence, misconduct (including criminal activity), and prompt attack (jailbreak and prompt injection). You can also define a set of topics or words that need to be blocked in your generative AI application, including harmful words, profanity, competitor names, and products. For example, a banking application can configure a guardrail to detect and block topics related to investment advice. A contact center application summarizing call center transcripts can use PII redaction to remove PIIs in call summaries, or a conversational chatbot can use content filters to block harmful content. Read more about Guardrails for Amazon Bedrock.

Companies like Aha!, a software company that helps more than 1 million people bring their product strategy to life, uses Amazon Bedrock to power many of their generative AI capabilities.

“We have full control over our information through Amazon Bedrock’s data protection and privacy policies, and can block harmful content through Guardrails for Amazon Bedrock. We just built on it to help product managers discover insights by analyzing feedback submitted by their customers. This is just the beginning. We will continue to build on advanced AWS technology to help product development teams everywhere prioritize what to build next with confidence.”

With even more choice of leading FMs and features that help you evaluate models and safeguard applications as well as leverage your prior investments in AI along with the capabilities of Amazon Bedrock, today’s launches make it even easier and faster for customers to build and scale generative AI applications. This blog post highlights only a subset of the new features. You can learn more about everything we’ve launched in the resources of this post, including asking questions and summarizing data from a single document without setting up a vector database in Knowledge Bases and the general availability of support for multiple data sources with Knowledge Bases.

Early adopters leveraging Amazon Bedrock’s capabilities are gaining a crucial head start – driving productivity gains, fueling ground-breaking discoveries across domains, and delivering enhanced customer experiences that foster loyalty and engagement. I’m excited to see what our customers will do next with these new capabilities.

As my mentor Werner Vogels always says “Now Go Build” and I’ll add “…with Amazon Bedrock!”

Resources

Check out the following resources to learn more about this announcement:

• Visit our community.aws site to find deep-dive technical content and to discover how our builder communities are using Amazon Bedrock in their solutions
• Learn more about Generative AI on AWS
• Learn more about Amazon Bedrock
• Learn more about customers achieving success with Amazon Bedrock

About the author

Swami Sivasubramanian is Vice President of Data and Machine Learning at AWS. In this role, Swami oversees all AWS Database, Analytics, and AI & Machine Learning services. His team’s mission is to help organizations put their data to work with a complete, end-to-end data solution to store, access, analyze, and visualize, and predict.

Read More