Tag Archives: Amazon SageMaker Data Wrangler

AWS Week in Review – Generative AI with LLM Hands-on Course, Amazon SageMaker Data Wrangler Updates, and More – July 3, 2023

Post Syndicated from Donnie Prakoso original https://aws.amazon.com/blogs/aws/aws-week-in-review-generative-ai-with-llm-hands-on-course-amazon-sagemaker-data-wrangler-updates-and-more-july-3-2023/

In last week’s AWS Week in Review post, Danilo mentioned that it’s summer in London. Well, I’m based in Singapore, and it’s mostly summer here. But, June is a special month here as it marks the start of durian season.

Starting next week, I’ll be travelling to Thailand, Malaysia, and the Philippines. But before I go, I want to share some interesting updates from last week for you.

Let’s get started.

Last Week’s Launches
Here are some launches that caught my attention:

New Hands-on Course: Generative AI with Large Language Models – Generative AI has been a technology highlight for the past few months. If you are on your journey to learn large language models (LLM), then you can try the new hands-on course Generative AI with LLMs at Coursera. Antje wrote a post to announce this collaboration course between DeepLearning.AI and AWS. This course is designed to prepare data scientists and engineers to become experts in selecting, training, fine-tuning, and deploying LLMs for real-world applications.

Generative AI with large language models

Amazon SageMaker Data Wrangler direct connection to Snowflake – With this announcement, you can now browse databases, tables, schemas, and query data from Snowflake in SageMaker Data Wrangler. This unlocks the possibility for you to join your data with other popular data sources, such as S3, Amazon Athena, Amazon Redshift, Amazon EMR and over 50 SaaS applications to create the right data set for machine learning.

Amazon SageMaker Role Manager now provides CDK library to create fine-grained permissions — The CDK support for Amazon SageMaker Role Manager lets you define permissions with fine-grained access for SageMaker users, jobs, and SageMaker pipelines programmatically. This will reduce manual efforts and consistent permissions management. For example, the following code grants permissions with a set of related machine learning activities specific to a persona.


export class myCDKStack extends cdk.Stack {
  constructor(scope: cdk.App, id: string, props?: cdk.StackProps) {
    super(scope, id, props);

    const persona = new Persona(this, 'example-persona-id', {
        activities: [
            Activity.runStudioAppsV2(this, 'example-id1', {}), 
            Activity.accessS3Buckets(this, 'example-id2', {s3buckets: [s3.S3Bucket.fromBucketName('DOC-EXAMPLE-BUCKET')]}) 
            Activity.accessAwsServices(this, 'example-id3', {})
        ]
    });

    const role = persona.createRole(this, 'example-IAM-role-id', 'example-IAM-role-name');
    
    }
}

AWS SDK for SAP ABAP – Great news for SAP ABAP developers! We just recently announced the general availability of the AWS SDK for SAP ABAP. With this, ABAP developers can use simple, secure and configurable connections between ABAP environments and 200+ supported AWS services in all AWS Regions, including AWS GovCloud (US) Regions. This AWS SDK helps ABAP developers to modernize their business processes with AWS services.

Amazon OpenSearch Ingestion now supports ingesting events from Amazon Security Lake – Amazon OpenSearch Ingestion now lets you bring data in the Apache Parquet format. As Amazon Security Lake also uses Open Cybersecurity Schema Framework (OCSF) in Apache Parquet format, it means you can easily ingest data from Amazon Security Lake.

For a full list of AWS announcements, be sure to keep an eye on the What’s New at AWS page.

AWS Open-Source Updates
As always, my colleague Ricardo has curated the latest updates for open-source news at AWS. Here are some of the highlights.

lightsail-miab-installer – This handy command-line tool developed by my colleague Rio Astamal was designed to simplify the process of setting up Mail-in-a-Box on Amazon Lightsail. With lightsail-miab-installer, you can effortlessly streamline the installation and configuration of Mail-in-a-Box, making it even more accessible and user-friendly.

rdsconn – This amazing tool, created by AWS Hero Aidan Steele, simplifies the process of connecting to an AWS RDS instance within a VPC directly from your laptop. Using the recently launched EC2 Instance Connect, rdsconn eliminates the need for cumbersome SSH tunnels.

cdk-appflow – If you’re using AWS CDK to build your applications and Amazon AppFlow to create bidirectional data transfer integrations between various SaaS applications and AWS, then you’re going to love cdk-appflow, a new AWS CDK construct for Amazon AppFlow. It’s currently in technical preview, but you’re more than welcome to try it and provide us with your feedback.

Upcoming AWS Events
There are also upcoming events that you can join to learn. Let’s start with AWS Summit events:

And, let’s learn from our fellow builders and join AWS Community Days:

Open for Registration for AWS re:Invent
Before I end this post, AWS re:Invent registration is now open!

This learning conference hosted by AWS for the global cloud computing community will be held from Nov 27 to Dec 1, 2023 in Las Vegas.

Pro-tip: You can use information on the Justify Your Trip page to prove the value of your trip to AWS re:Invent.

That’s all for this week. Check back next Monday for another Week in Review.

Happy building.
— Donnie

This post is part of our Week in Review series. Check back each week for a quick roundup of interesting news and announcements from AWS!

New — Introducing Support for Real-Time and Batch Inference in Amazon SageMaker Data Wrangler

Post Syndicated from Donnie Prakoso original https://aws.amazon.com/blogs/aws/new-introducing-support-for-real-time-and-batch-inference-in-amazon-sagemaker-data-wrangler/

To build machine learning models, machine learning engineers need to develop a data transformation pipeline to prepare the data. The process of designing this pipeline is time-consuming and requires a cross-team collaboration between machine learning engineers, data engineers, and data scientists to implement the data preparation pipeline into a production environment.

The main objective of Amazon SageMaker Data Wrangler is to make it easy to do data preparation and data processing workloads. With SageMaker Data Wrangler, customers can simplify the process of data preparation and all of the necessary steps of data preparation workflow on a single visual interface. SageMaker Data Wrangler reduces the time to rapidly prototype and deploy data processing workloads to production, so customers can easily integrate with MLOps production environments.

However, the transformations applied to the customer data for model training need to be applied to new data during real-time inference. Without support for SageMaker Data Wrangler in a real-time inference endpoint, customers need to write code to replicate the transformations from their flow in a preprocessing script.

Introducing Support for Real-Time and Batch Inference in Amazon SageMaker Data Wrangler
I’m pleased to share that you can now deploy data preparation flows from SageMaker Data Wrangler for real-time and batch inference. This feature allows you to reuse the data transformation flow which you created in SageMaker Data Wrangler as a step in Amazon SageMaker inference pipelines.

SageMaker Data Wrangler support for real-time and batch inference speeds up your production deployment because there is no need to repeat the implementation of the data transformation flow. You can now integrate SageMaker Data Wrangler with SageMaker inference. The same data transformation flows created with the easy-to-use, point-and-click interface of SageMaker Data Wrangler, containing operations such as Principal Component Analysis and one-hot encoding, will be used to process your data during inference. This means that you don’t have to rebuild the data pipeline for a real-time and batch inference application, and you can get to production faster.

Get Started with Real-Time and Batch Inference
Let’s see how to use the deployment supports of SageMaker Data Wrangler. In this scenario, I have a flow inside SageMaker Data Wrangler. What I need to do is to integrate this flow into real-time and batch inference using the SageMaker inference pipeline.

First, I will apply some transformations to the dataset to prepare it for training.

I add one-hot encoding on the categorical columns to create new features.

Then, I drop any remaining string columns that cannot be used during training.

My resulting flow now has these two transform steps in it.

After I’m satisfied with the steps I have added, I can expand the Export to menu, and I have the option to export to SageMaker Inference Pipeline (via Jupyter Notebook).

I select Export to SageMaker Inference Pipeline, and SageMaker Data Wrangler will prepare a fully customized Jupyter notebook to integrate the SageMaker Data Wrangler flow with inference. This generated Jupyter notebook performs a few important actions. First, define data processing and model training steps in a SageMaker pipeline. The next step is to run the pipeline to process my data with Data Wrangler and use the processed data to train a model that will be used to generate real-time predictions. Then, deploy my Data Wrangler flow and trained model to a real-time endpoint as an inference pipeline. Last, invoke my endpoint to make a prediction.

This feature uses Amazon SageMaker Autopilot, which makes it easy for me to build ML models. I just need to provide the transformed dataset which is the output of the SageMaker Data Wrangler step and select the target column to predict. The rest will be handled by Amazon SageMaker Autopilot to explore various solutions to find the best model.

Using AutoML as a training step from SageMaker Autopilot is enabled by default in the notebook with the use_automl_step variable. When using the AutoML step, I need to define the value of target_attribute_name, which is the column of my data I want to predict during inference. Alternatively, I can set use_automl_step to False if I want to use the XGBoost algorithm to train a model instead.

On the other hand, if I would like to instead use a model I trained outside of this notebook, then I can skip directly to the Create SageMaker Inference Pipeline section of the notebook. Here, I would need to set the value of the byo_model variable to True. I also need to provide the value of algo_model_uri, which is the Amazon Simple Storage Service (Amazon S3) URI where my model is located. When training a model with the notebook, these values will be auto-populated.

In addition, this feature also saves a tarball inside the data_wrangler_inference_flows folder on my SageMaker Studio instance. This file is a modified version of the SageMaker Data Wrangler flow, containing the data transformation steps to be applied at the time of inference. It will be uploaded to S3 from the notebook so that it can be used to create a SageMaker Data Wrangler preprocessing step in the inference pipeline.

The next step is that this notebook will create two SageMaker model objects. The first object model is the SageMaker Data Wrangler model object with the variable data_wrangler_model, and the second is the model object for the algorithm, with the variable algo_model. Object data_wrangler_model will be used to provide input in the form of data that has been processed into algo_model for prediction.

The final step inside this notebook is to create a SageMaker inference pipeline model, and deploy it to an endpoint.

Once the deployment is complete, I will get an inference endpoint that I can use for prediction. With this feature, the inference pipeline uses the SageMaker Data Wrangler flow to transform the data from your inference request into a format that the trained model can use.

In the next section, I can run individual notebook cells in Make a Sample Inference Request. This is helpful if I need to do a quick check to see if the endpoint is working by invoking the endpoint with a single data point from my unprocessed data. Data Wrangler automatically places this data point into the notebook, so I don’t have to provide one manually.

Things to Know
Enhanced Apache Spark configuration — In this release of SageMaker Data Wrangler, you can now easily configure how Apache Spark partitions the output of your SageMaker Data Wrangler jobs when saving data to Amazon S3. When adding a destination node, you can set the number of partitions, corresponding to the number of files that will be written to Amazon S3, and you can specify column names to partition by, to write records with different values of those columns to different subdirectories in Amazon S3. Moreover, you can also define the configuration in the provided notebook.

You can also define memory configurations for SageMaker Data Wrangler processing jobs as part of the Create job workflow. You will find similar configuration as part of your notebook.

Availability — SageMaker Data Wrangler supports for real-time and batch inference as well as enhanced Apache Spark configuration for data processing workloads are generally available in all AWS Regions that Data Wrangler currently supports.

To get started with Amazon SageMaker Data Wrangler supports for real-time and batch inference deployment, visit AWS documentation.

Happy building
— Donnie

New — Amazon SageMaker Data Wrangler Supports SaaS Applications as Data Sources

Post Syndicated from Donnie Prakoso original https://aws.amazon.com/blogs/aws/new-amazon-sagemaker-data-wrangler-supports-saas-applications-as-data-sources/

Data fuels machine learning. In machine learning, data preparation is the process of transforming raw data into a format that is suitable for further processing and analysis. The common process for data preparation starts with collecting data, then cleaning it, labeling it, and finally validating and visualizing it. Getting the data right with high quality can often be a complex and time-consuming process.

This is why customers who build machine learning (ML) workloads on AWS appreciate the ability of Amazon SageMaker Data Wrangler. With SageMaker Data Wrangler, customers can simplify the process of data preparation and complete the required processes of the data preparation workflow on a single visual interface. Amazon SageMaker Data Wrangler helps to reduce the time it takes to aggregate and prepare data for ML.

However, due to the proliferation of data, customers generally have data spread out into multiple systems, including external software-as-a-service (SaaS) applications like SAP OData for manufacturing data, Salesforce for customer pipeline, and Google Analytics for web application data. To solve business problems using ML, customers have to bring all of these data sources together. They currently have to build their own solution or use third-party solutions to ingest data into Amazon S3 or Amazon Redshift. These solutions can be complex to set up and not cost-effective.

Introducing Amazon SageMaker Data Wrangler Supports SaaS Applications as Data Sources
I’m happy to share that starting today, you can aggregate external SaaS application data for ML in Amazon SageMaker Data Wrangler to prepare data for ML. With this feature, you can use more than 40 SaaS applications as data sources via Amazon AppFlow and have these data available on Amazon SageMaker Data Wrangler. Once the data sources are registered in AWS Glue Data Catalog by AppFlow, you can browse tables and schemas from these data sources using Data Wrangler SQL explorer. This feature provides seamless data integration between SaaS applications and SageMaker Data Wrangler using Amazon AppFlow.

Here is a quick preview of this new feature:

This new feature of Amazon SageMaker Data Wrangler works by using integration with Amazon AppFlow, a fully managed integration service that enables you to securely exchange data between SaaS applications and AWS services. With Amazon AppFlow, you can establish bidirectional data integration between SaaS applications, such as Salesforce, SAP, and Amplitude and all supported services, into your Amazon S3 or Amazon Redshift.

Then, with Amazon AppFlow, you can catalog the data in AWS Glue Data Catalog. This is a new feature where with Amazon AppFlow, you can create an integration with AWS Glue Data Catalog for Amazon S3 destination connector. With this new integration, customers can catalog SaaS data applications into AWS Glue Data Catalog with a few clicks, directly from the Amazon AppFlow Flow configuration, without the need to run any crawlers.

Once you’ve established a flow and inserted it into the AWS Glue Data Catalog, you can use this data inside the Amazon SageMaker Data Wrangler. Then, you can do the data preparation as you usually do. You can write Amazon Athena queries to preview data, join data from multiple sources, or import data to prepare for ML model training.

With this feature, you need to do a few simple steps to perform seamless data integration between SaaS applications into Amazon SageMaker Data Wrangler via Amazon AppFlow. This integration supports more than 40 SaaS applications, and for a complete list of supported applications, please check the Supported source and destination applications documentation.

Get Started with Amazon SageMaker Data Wrangler Support for Amazon AppFlow
Let’s see how this feature works in detail. In my scenario, I need to get data from Salesforce, and do the data preparation using Amazon SageMaker Data Wrangler.

To start using this feature, the first thing I need to do is to create a flow in Amazon AppFlow that registers the data source into the AWS Glue Data Catalog. I already have an existing connection with my Salesforce account, and all I need now is to create a flow.

One important thing to note is that to make SaaS application data available in Amazon SageMaker Data Wrangler, I need to create a flow with Amazon S3 as the destination. Then, I need to enable Create a Data Catalog table in the AWS Glue Data Catalog settings. This option will automatically catalog my Salesforce data into AWS Glue Data Catalog.

On this page, I need to select a user role with the required AWS Glue Data Catalog permissions and define the database name and the table name prefix. In addition, in this section, I can define the data format preference, be it in JSON, CSV, or Apache Parquet formats, and filename preference if I want to add a timestamp into the file name section.

To learn more about how to register SaaS data in Amazon AppFlow and AWS Glue Data Catalog, you can read Cataloging the data output from an Amazon AppFlow flow documentation page.

Once I’ve finished registering SaaS data, I need to make sure the IAM role can view the data sources in Data Wrangler from AppFlow. Here is an example of a policy in the IAM role:

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": "glue:SearchTables",
            "Resource": [
                "arn:aws:glue:*:*:table/*/*",
                "arn:aws:glue:*:*:database/*",
                "arn:aws:glue:*:*:catalog"
            ]
        }
    ]
}

By enabling data cataloging with AWS Glue Data Catalog, from this point on, Amazon SageMaker Data Wrangler will be able to automatically discover this new data source and I can browse tables and schema using the Data Wrangler SQL Explorer.

Now it’s time to switch to the Amazon SageMaker Data Wrangler dashboard then select Connect to data sources.

On the following page, I need to Create connection and select the data source I want to import. In this section, I can see all the available connections for me to use. Here I see the Salesforce connection is already available for me to use.

If I would like to add additional data sources, I can see a list of external SaaS applications that I can integrate into the Set up new data sources section. To learn how to recognize external SaaS applications as data sources, I can learn more with the select How to enable access.

Now I will import datasets and select the Salesforce connection.

On the next page, I can define connection settings and import data from Salesforce. When I’m done with this configuration, I select Connect.

On the following page, I see my Salesforce data that I already configured with Amazon AppFlow and AWS Glue Data Catalog called appflowdatasourcedb. I can also see a table preview and schema for me to review if this is the data I need.

Then, I start building my dataset using this data by performing SQL queries inside the SageMaker Data Wrangler SQL Explorer. Then, I select Import query.

Then, I define a name for my dataset.

At this point, I can start doing the data preparation process. I can navigate to the Analysis tab to run the data insight report. The analysis will provide me with a report on the data quality issues and what transform I need to use next to fix the issues based on the ML problem I want to predict. To learn more about how to use the data analysis feature, see Accelerate data preparation with data quality and insights in the Amazon SageMaker Data Wrangler blog post.

In my case, there are several columns I don’t need, and I need to drop these columns. I select Add step.

One feature I like is that Amazon SageMaker Data Wrangler provides numerous ML data transforms. It helps me to streamline the process of cleaning, transforming and feature engineering my data in one dashboard. For more about what SageMaker Data Wrangler provides for transformation data, please read this Transform Data documentation page.

In this list, I select Manage columns.

Then, in the Transform section, I select the Drop column option. Then, I select a few columns that I don’t need.

Once I’m done, the columns I don’t need are removed and the Drop column data preparation step I just created is listed in the Add step section.

I can also see the visual of my data flow inside the Amazon SageMaker Data Wrangler. In this example, my data flow is quite basic. But when my data preparation process becomes complex, this visual view makes it easy for me to see all the data preparation steps.

From this point on, I can do what I require with my Salesforce data. For example, I can export data directly to Amazon S3 by selecting Export to and choosing Amazon S3 from the Add destination menu. In my case, I specify Data Wrangler to store the data in Amazon S3 after it has processed it by selecting Add destination and then Amazon S3.

Amazon SageMaker Data Wrangler provides me flexibility to automate the same data preparation flow using scheduled jobs. I can also automate feature engineering with SageMaker Pipelines (via Jupyter Notebook) and SageMaker Feature Store (via Jupyter Notebook), and deploy to Inference end point with SageMaker Inference Pipeline (via Jupyter Notebook).

Things to Know
Related news – This feature will make it easy for you to do data aggregation and preparation with Amazon SageMaker Data Wrangler. As this feature is an integration with Amazon AppFlow and also AWS Glue Data Catalog, you might want to learn more on Amazon AppFlow now supports AWS Glue Data Catalog integration and provides enhanced data preparation page.

Availability – Amazon SageMaker Data Wrangler supports SaaS applications as data sources available in all the Regions currently supported by Amazon AppFlow.

Pricing – There is no additional cost to use SaaS applications supports in Amazon SageMaker Data Wrangler, but there is a cost to running Amazon AppFlow to get the data in Amazon SageMaker Data Wrangler.

Visit Import Data From Software as a Service (SaaS) Platforms documentation page to learn more about this feature, and follow the getting started guide to start data aggregating and preparing SaaS applications data with Amazon SageMaker Data Wrangler.

Happy building!
— Donnie

Detecting data drift using Amazon SageMaker

Post Syndicated from Shibu Nair original https://aws.amazon.com/blogs/architecture/detecting-data-drift-using-amazon-sagemaker/

As companies continue to embrace the cloud and digital transformation, they use historical data in order to identify trends and insights. This data is foundational to power tools, such as data analytics and machine learning (ML), in order to achieve high quality results.

This is a time where major disruptions are not only lasting longer, but also happening more frequently, as discussed in a McKinsey article on risk and resilience. Any disruption—a pandemic, hurricane, or even blocked sailing routes—has a major impact on the patterns of data and can create anomalous behavior.

ML models are dependent on data insights to help plan and support production-ready applications. With any disruptions, data drift can occur. Data drift is unexpected and undocumented changes to data structure, semantics, and/or infrastructure. If there is data drift, the model performance will degrade and no longer provide an accurate guidance. To mitigate the effects of the disruption, data drift needs to be detected and the ML models quickly trained and adjusted accordingly.

This blog post explains how to approach changing data patterns in the age of disruption and how to mitigate its effects on ML models. We also discuss the steps of building a feedback loop to capture the request data in the production environment and create a data pipeline to store the data for profiling and baselining. Then, we explain how Amazon SageMaker Clarify can help detect data drift.

How to detect data drift

There are three stages to detecting data drift: data quality monitoring, model quality monitoring, and drift evaluation (see Figure 1).

Stages in detecting data drift

Figure 1. Stages in detecting data drift

Data quality monitoring establishes a profile of the input data during model training, and then continuously compares incoming data with the profile. Deviations in the data profile signal a drift in the input data.

You can also detect drift through model quality monitoring, which requires capturing actual values that can be compared with the predictions. For example, using weekly demand forecasting, you can compare the forecast quantities one week later with the actual demand. Some use cases can require extra steps to collect actual values. For example, product recommendations may require you to ask a selected group of consumers for their feedback to the recommendation.

SageMaker Clarify provides insights into your trained models, including importance of model features and any biases towards certain segments of the input data. Changes of these attributes between re-trained models also signal drift. Drift evaluation constitutes the monitoring data and mechanisms to detect changes and triggering consequent actions. With Amazon CloudWatch, you can define rules and thresholds that prompt drift notifications.

Figure 2 illustrates a basic architecture with the data sources for training and production (on the left) and the observed data concerning drift (on the right). You can use Amazon SageMaker Data Wrangler, a visual data preparation tool, to clean and normalize your input data for your ML task. You can store the features that you defined for your models in the Amazon SageMaker Feature Store, a fully managed, purpose-built repository to store, update, retrieve, and share ML features.

The white, rectangular boxes in the architecture diagram represent the tasks for detecting data and model drift. You can integrate those tasks into your ML workflow with Amazon SageMaker Pipelines.

Basic architecture on how data drift is detected using Amazon SageMaker

Figure 2. Basic architecture on how data drift is detected using Amazon SageMaker

The drift observation data can be captured in tabular format, such as comma-separated values or Parquet, on Amazon Simple Storage Service (S3) and analyzed with Amazon Athena and Amazon QuickSight.

How to build a feedback loop

The baselining task establishes a data profile from training data. It uses Amazon SageMaker Model Monitor and runs before training or re-training the model. The baseline profile is stored on Amazon S3 to be referenced by the data drift monitoring job.

The data drift monitoring task continuously profiles the input data, compares it with baseline, and the results are captured in CloudWatch. This tasks runs on its own computation resources using Deequ, which checks that the monitoring job does not slow down your ML inference flow and scales with the data. The frequency of running this task can be adjusted to control cost, which can depend on how rapidly you anticipate that the data may change.

The model quality monitoring task computes model performance metrics from actuals and predicted values. The origin of these data points depends on the use case. Demand forecasting use cases naturally capture actuals that can be used to validate past predictions. Other use cases can require extra steps to acquire ground-truth data.

CloudWatch is a monitoring and observability service with which you can define rules to act on deviation in model performance or data drift. With CloudWatch, you can setup alerts to users via e-mail or SMS, and it can automatically start the ML model re-training process.

Run the baseline task on your updated data set before re-training your model. Use the SageMaker model registry to catalog your ML models for production, manage model versions, and control the associate training metrics.

Gaining insight into data and models

SageMaker Clarify provides greater visibility into your training data and models, helping identify and limit bias and explain predictions. For example, the trained models may consider some features more strongly than others when generating predictions. Compare the feature importance and bias between model-provided versions for a better understanding of the changes.

Conclusion

As companies continue to use data analytics and ML to inform daily activity, data drift may become a more common occurrence. Recognizing that drift can have a direct impact on models and production-ready applications, it is important to architect to identify potential data drift and avoid downgrading the models and negatively impacting results. Failure to capture changes in data can result in loss of process confidence, downgraded model accuracy, or a bottom-line impact to the business.