Machine learning on structured data at speed and scale
AutoML Tables enables your entire team of data scientists, analysts, and developers to automatically build and deploy state-of-the-art machine learning models on structured data at massively increased speed and scale. Transform your business by leveraging your enterprise data to tackle mission-critical tasks like supply chain management, fraud detection, lead conversion optimization, and increasing customer lifetime value.
Increases model quality
Produce state-of-the art models with one click. AutoML Tables automatically handles a wide range of tabular data primitives — such as numbers, classes, strings, timestamps, and lists — and also helps you detect and handle missing values, outliers, and other common data issues. Leverage the best of Google’s model zoo for ML on structured data all in one place.
Our codeless interface guides users through the full end-to-end machine learning lifecycle, making it easy for anyone on your team to build models and reliably incorporate them into broader applications. We also provide extensive input data and model behavior explainability features, along with guardrails to prevent common mistakes.
Easy-to-deploy and scale models
AutoML Tables uses Google’s low-latency serving infrastructure, which makes deploying machine learning models extremely easy, regardless of production workload volume and global reach.
Flexible user options
To meet all user preferences, we offer the flexibility to use AutoML Tables in an API or notebook environment.
AutoML Tables reduces the time it takes to go from raw data to top-quality, production-ready machine learning models from months to just a few days.
AutoML Tables doesn’t require a large annual licensing fee. It’s priced based on compute and memory usage, so you’ll only get charged for what you actually use.
The speed, precision, and scale of AutoML Tables allowed Fox Sports to create an entirely new experience for millions of cricket fans across Australia. By training our model on the multiple variables of historical cricket matches we could predict when wickets would fall 5 minutes before it happened on the pitch. This new feature became a fundamental part of our marketing strategy through integrating AutoML with App Engine and Cloud Dataflow to transform every customer touchpoint. We’ve showcased cricket like never before with user engagement up 140% vs industry averages and the marketing for this activity delivering 150% more subscribers per dollar spent by communicating to fans in the right place at the right time.Chris Pocock, Marketing Director, Fox Sports
AutoML Tables lets us anticipate how players will behave in our Harry Potter game. With over a million unique users, within three days after the app is installed, we can begin to predict their game behavior. Accordingly, we can use Google AdWords to target the players we think will become paying users on our platform, for more efficient ad spend and higher revenue generation.Joshua Clark, Lead Data Scientist, Jam City
AutoML Tables has allowed Indiana University to experiment with our existing work, which has led to some impressive results. The accuracy of out-of-sample predictions from this turn-key solution were strong, considering the lack of structure on the models. AutoML quickly provided analysis and insights to data that would have taken weeks to otherwise implement. The intuitive web interface and the direct integration with BigQuery rapidly accelerated the training and deployment of predictive models at institutional scale. IU plans to continue to leverage this service for future analytics initiatives to increase speed and scale of IU projects.Ben Motz, Faculty Fellow for Academic Analytics, University Information Technology Services, Indiana University
AutoML Tables and your business
Leverage your enterprise data to address a whole host of mission-critical challenges.
Maximize your revenue
Better predict customer demand so you can preemptively fill gaps in your portfolio and maximize your revenue by optimizing product distribution, promotions, and pricing.
Optimize your portfolio
Foresee and optimize your policyholder portfolio’s risk and return by zeroing in on the potential for large claims and likelihood of fraud.
Understand your customer
What’s your average customer’s lifetime value? Make the most of your marketing spend by using AutoML Tables to estimate predicted purchasing value, volume, frequency, lead conversion probability, and churn likelihood.
Maintain your equipment
Proactively anticipate asset, device, and equipment breakdowns to ensure your fleet operates at optimal performance with minimal costs.
|Compute and memory usage|
|Training||6 hours of free one-time use + $19.32 per hour
(92 n1-standard-4 equivalent machines used in parallel)
|Batch prediction||6 hours of free one-time use + $1.16 per hour
(5.5 n1-standard-4 equivalent machines used in parallel)
|Online prediction||$0.21 per hour
(1 n1-standard-4 equivalent machine)
|Deployment||$0.005 per GiB-hour x 9 machines
(model replicated to 9 machines for low latency serving purposes)
For more detailed information, please view the pricing guide.
AutoML Tables beginner’s guide
AutoML Tables documentation
Ready to try AutoML Tables?
Automatically build and deploy state-of-the-art machine learning models on structured data.
This product is in beta. For more information on our product launch stages, see here.