|
Amazon SageMaker Canvas delivers small business analysts with a visible interface to solve enterprise troubles working with device mastering (ML) with no producing a one line of code. Since we released SageMaker Canvas in 2021, many users have asked us for an increased, seamless collaboration encounter that permits knowledge scientists to share experienced designs with their company analysts with a few straightforward clicks.
Today, I’m energized to announce that you can now carry ML types constructed anywhere into SageMaker Canvas and produce predictions.
New – Deliver Your Personal Product into SageMaker Canvas
As a information scientist or ML practitioner, you can now seamlessly share models developed anyplace, inside of or exterior Amazon SageMaker, with your enterprise groups. This removes the weighty lifting for your engineering teams to develop a independent software or person interface to share ML products and collaborate in between the various sections of your group. As a company analyst, you can now leverage ML designs shared by your info researchers inside minutes to generate predictions.
Enable me display you how this works in observe!
In this illustration, I share an ML design that has been experienced to detect consumers that are potentially at risk of churning with my advertising and marketing analyst. Very first, I sign up the design in the SageMaker model registry. SageMaker model registry allows you catalog versions and control model versions. I build a model group known as 2022-purchaser-churn-product-group and then pick Build product edition to register my model.
To sign up your product, present the locale of the inference impression in Amazon ECR, as properly as the locale of your design.tar.gz file in Amazon S3. You can also incorporate product endpoint recommendations and extra model information and facts. When you’ve registered your model, find the design version and find Share.
You can now pick the SageMaker Canvas person profile(s) in just the similar SageMaker area you want to share your design with. Then, give additional design information, these as details about training and validation datasets, the ML difficulty type, and design output information. You can also insert a take note for the SageMaker Canvas consumers you share the model with.
Likewise, you can now also share models skilled in SageMaker Autopilot and models accessible in SageMaker JumpStart with SageMaker Canvas buyers.
The organization analysts will obtain an in-app notification in SageMaker Canvas that a product has been shared with them, alongside with any notes you added.
My advertising analyst can now open, evaluate, and start out applying the model to crank out ML predictions in SageMaker Canvas.
Decide on Batch prediction to make ML predictions for an full dataset or Single prediction to develop predictions for a one enter. You can down load the results in a .csv file.
New – Improved Product Sharing and Collaboration from SageMaker Canvas with SageMaker Studio Buyers
We also improved the sharing and collaboration capabilities from SageMaker Canvas with data science and ML groups. As a organization analyst, you can now pick out which SageMaker Studio consumer profile(s) you want to share your common-develop styles with.
Your knowledge researchers or ML practitioners will obtain a related in-application notification in SageMaker Studio the moment a model has been shared with them, alongside with any notes from you. In addition to just examining the model, SageMaker Studio end users can now also, if required, update the data transformations in SageMaker Facts Wrangler, retrain the model in SageMaker Autopilot, and share back the up-to-date product. SageMaker Studio consumers can also advise an alternate model from the listing of versions in SageMaker Autopilot.
After SageMaker Studio customers share back a model, you receive an additional notification in SageMaker Canvas that an current design has been shared back with you. This collaboration concerning enterprise analysts and data researchers will enable democratize ML across businesses by bringing transparency to automated conclusions, constructing belief, and accelerating ML deployments.
Now Available
The enhanced, seamless collaboration capabilities for Amazon SageMaker Canvas, which includes the ability to bring your ML versions designed everywhere, are offered these days in all AWS Locations where SageMaker Canvas is offered with no alterations to the current SageMaker Canvas pricing.
Start off collaborating and convey your ML product to Amazon SageMaker Canvas now!
— Antje