Ambarella Broadens Services Using The Amazon SageMaker Neo
Ambarella is an artificial intelligence (AI) vision silicon company recently collaborated with Amazon, specifically AWS on single-click machine learning for edge applications. It was recently announced that Ambarella and Amazon Web Services, Inc. (AWS) customers can now use the Amazon SageMaker Neo to train machine learning (ML) models once and run them on any device equipped with an Ambarella CVflow-powered AI vision system on chip (SoC).
“Ambarella is in mass production today with CVflow AI vision processors for the home monitoring, enterprise video security, and automotive markets,” said Chris Day, vice president of marketing and business development for Ambarella. “The ability to select an Ambarella SoC and compile a trained ML model with a single click is a powerful tool that makes it possible for our customers to rapidly bring the next generation of AI-enabled products to market.”
Until now, developers had to manually optimize ML models for devices based on Ambarella AI vision SoCs. This step could add considerable delays and errors to the application development process. Ambarella and AWS collaborated to simplify the process by integrating the Ambarella toolchain with the Amazon SageMaker Neo cloud service. Now, developers can simply bring their trained models to Amazon SageMaker Neo and automatically optimize the model for Ambarella CVflow-powered SoCs.
Customers can build an ML model using MXNet, TensorFlow, PyTorch, or XGBoost and train the model using Amazon SageMaker in the cloud or on their local machine. Then, they can upload the model to their AWS account and use Amazon SageMaker Neo to optimize the model for Ambarella SoCs. They can choose CV25, CV22, or CV2 as the compilation target. Amazon SageMaker Neo compiles the trained model into an executable that is optimized for Ambarella’s CVflow neural network accelerator. The compiler applies a series of optimizations that can make the model run up to 2x faster on the Ambarella SoC. Customers can download the compiled model and deploy it to their fleet of Ambarella-equipped devices. The optimized model runs in the Amazon SageMaker Neo runtime purpose-built for Ambarella SoCs and available for the Ambarella SDK. The Amazon SageMaker Neo runtime occupies less than 10x the disk and memory footprint of TensorFlow, MXNet, or PyTorch, making it much more efficient to deploy ML models on connected cameras.
AWS has the broadest and deepest set of ML and AI services focused on solving some of the toughest challenges facing developers. Amazon SageMaker is a fully managed service that provides every developer and data scientist with the ability to build, train, and deploy machine learning models quickly,” said Bratin Saha, vice president, Machine Learning & Engines, Amazon Web Services, Inc. “We’re excited that VIVOTEK is using SageMaker Neo to simplify the deployment of ML models at the edge on Ambarella CVflow-powered IP cameras.”
