Migrating Amazon SageMaker Knowledge Wrangler flows to Amazon SageMaker Canvas for quicker information preparation

Amazon SageMaker Knowledge Wrangler offers a visible interface to streamline and speed up information preparation for machine studying (ML), which is usually essentially the most time-consuming and tedious job in an ML mission. Amazon SageMaker Canvas is a low-code, no-code visible interface to construct and deploy ML fashions with out writing any code. Based mostly on buyer suggestions, we’ve built-in the superior ML-specific information preparation capabilities of SageMaker Knowledge Wrangler inside SageMaker Canvas, offering customers with an end-to-end, no-code workspace for information preparation, constructing and deploying ML fashions.

SageMaker Canvas abstracts a lot of the complexity of the ML workflow so you may put together your information and construct or use fashions to generate extremely correct enterprise insights with out writing any code. Moreover, getting ready your information in SageMaker Canvas affords many enhancements, together with as much as 10x quicker web page hundreds, a pure language interface for information preparation, the power to see the scale and form of your information at every step, and enhancements to exchange and reorder transformations for iterating by way of your dataflow. Lastly, you may create fashions with one click on in the identical interface, or create SageMaker Canvas datasets to fine-tune your basis mannequin (FM).

On this submit, we present learn how to convey your current SageMaker Knowledge Wrangler flows (directions created when constructing information transformations) from SageMaker Studio Basic to SageMaker Canvas. We offer an instance of transferring information from SageMaker Studio Basic to Amazon Easy Storage Service (Amazon S3) as an intermediate step earlier than importing the information into SageMaker Canvas.

Resolution overview

The high-level steps are as follows:

1. Open a terminal in SageMaker Studio and replica the move file to Amazon S3.
2. Import the move file from Amazon S3 into SageMaker Canvas.

Stipulations

On this instance, data-wrangler-classic-flows As a staging folder for migrating move information to Amazon S3. You do not have to create a migration folder, however on this instance, a folder was created utilizing the file system browser portion of SageMaker Studio Basic. After you create the folder, watch out to maneuver and consolidate associated SageMaker Knowledge Wrangler move information collectively. Within the following screenshot, the three move information required for migration have been moved into the folder. data-wrangler-classic-flows, One in all these information, as proven within the left pane, is titanic.movewill open and seem in the appropriate pane.

Copy the move file to Amazon S3

To repeat your move information to Amazon S3, full the next steps:

1. To open a brand new terminal in SageMaker Studio Basic, file Menu, Choose Terminal.
   
2. Upon getting a brand new terminal open, you may copy the flowfile to your most well-liked Amazon S3 location by coming into the next command (exchange NNNNNNNNNNNN together with your AWS account quantity):

   cd data-wrangler-classic-flows
   goal="s3://sagemaker-us-west-2-NNNNNNNNNNNN/data-wrangler-classic-flows/"
   aws s3 sync . $goal --exclude "*.*" --include "*.move"

The next screenshot reveals an instance of the Amazon S3 sync course of. A affirmation message seems in any case information have been uploaded. You’ll be able to alter the code above to satisfy your individual enter folder and Amazon S3 location wants. When you do not wish to create a folder, while you enter the terminal, you may simply click on Change Listing (cd) command copies all move information throughout your SageMaker Studio Basic file system to Amazon S3, no matter their unique folder.

After you add the information to Amazon S3, you should utilize the Amazon S3 console to confirm that the information have been copied. Within the following screenshot, the unique three move information are seen within the S3 bucket.

Importing Knowledge Wrangler move information into SageMaker Canvas

To import your move file into SageMaker Canvas, observe these steps:

1. Within the SageMaker Studio console, Knowledge Wrangler Within the navigation pane.
2. select Importing a Dataflow.
   
3. for Choose a knowledge supply and click on select Amazon S3.
4. for Enter S3 endpointEnter the Amazon S3 location that you just used earlier to repeat the information from SageMaker Studio to Amazon S3. goYou can too navigate to the Amazon S3 location utilizing the next browsers:
5. Choose the move file you wish to import, Import.
   

After you import the file, the SageMaker Knowledge Wrangler web page updates to point out the newly imported file, as proven within the following screenshot.

Utilizing SageMaker Canvas to rework information with SageMaker Knowledge Wrangler

Choose one of many flows (on this instance, titanic.move) to begin the SageMaker Knowledge Wrangler transformation.

Now you can add evaluation and transformation to your dataflows utilizing both a visible interface (Speed up Knowledge Prep for ML with Amazon SageMaker Canvas) or a pure language interface (Discover and put together information in pure language utilizing new options in Amazon SageMaker Canvas).

Whenever you’re blissful together with your information, choose the plus signal Create a mannequinor choose export Export the dataset to construct and use an ML mannequin.

Different migration strategies

This submit described learn how to migrate SageMaker Knowledge Wrangler move information from a SageMaker Studio Basic atmosphere utilizing Amazon S3. Part 3: (Optionally available) Migrate information from Studio Basic to Studio offers a second technique to switch move information utilizing an area machine. Moreover, you may obtain a single move file from the SageMaker Studio tree management to your native machine and manually import it into SageMaker Canvas. Select the tactic that fits your wants and use case.

cleansing

Whenever you’re completed, shut down any SageMaker Knowledge Wrangler purposes which might be operating in SageMaker Studio Basic. To save lots of prices, you can even delete the move information from the SageMaker Studio Basic file browser, which is an Amazon Elastic File System (Amazon EFS) quantity. You can too delete the intermediate information in Amazon S3. As soon as the move information have been imported into SageMaker Canvas, the information copied to Amazon S3 are not wanted.

You’ll be able to log off of SageMaker Canvas when you’re completed together with your work, after which relaunch it when you’re prepared to make use of it once more.

Conclusion

Migrating your current SageMaker Knowledge Wrangler flows to SageMaker Canvas is a simple course of, permitting you to make use of the superior information preparation you have already developed whereas making the most of the end-to-end low-code, no-code ML workflow of SageMaker Canvas. By following the steps outlined on this submit, you may seamlessly migrate your information wrangling artifacts to your SageMaker Canvas atmosphere, streamlining your ML tasks and empowering enterprise analysts and non-technical customers to construct and deploy fashions extra effectively.

Strive SageMaker Canvas at this time and expertise the ability of a unified platform for information preparation, mannequin constructing, and deployment.

In regards to the Creator

Charles Laughlin Charles is a Principal AI Specialist at Amazon Internet Companies (AWS). Charles holds an MSc in Provide Chain Administration and a PhD in Knowledge Science. Charles works on the Amazon SageMaker service staff, integrating analysis and buyer suggestions into the service roadmap. In his work, he works day by day with varied AWS clients, serving to them rework their companies with innovative AWS applied sciences and thought management.

Dan Schinreich He’s a Senior Product Supervisor at Amazon SageMaker, specializing in increasing our no-code/low-code choices. He’s captivated with making ML and generative AI extra accessible to assist clear up powerful issues. Exterior of labor, he enjoys enjoying hockey, scuba diving, and studying sci-fi novels.

Phuong Nguyen He’s a Senior Product Supervisor at AWS with 15 years of expertise constructing customer-centric, data-driven merchandise, the place he leads ML information preparation for SageMaker Canvas and SageMaker Knowledge Wrangler.

Davide Galittelli I am a Specialist Options Architect for AI/ML in EMEA. I am primarily based in Brussels and work intently with purchasers throughout the Benelux. I have been a developer since I used to be younger, beginning to code on the age of seven. I began studying AI/ML after graduating college and have been hooked ever since. Get verified