AWS offers a comprehensive platform for machine learning (ML) services and internet of things (IoT) software services which can be leveraged and deployed on TI’s analytics processors like the TDA4x family of processors. Training and optimizing ML models require massive computing resources, so it is a natural fit for the cloud.
AWS SageMaker is a suite of services and tools providing every developer and data scientist with the ability to prepare and label data sets, build models, train, optimize, deploy and, monitor\maintain machine-learning models at scale.
Amazon SageMaker Neo service optimizes machine-learning models already built with DarkNet, Keras, MXNet, PyTorch, TensorFlow, TensorFlow-Lite, ONNX, or XGBoost for inference on TI Jacinto processor. SageMaker Neo optimizes the trained model and compiles it into an executable, offloading subgraphs to C7x/MMA for accelerated execution with TIDL and generating code and running on the ARM core for layers that are not supported by TIDL. Sagemaker Neo enables TVM and Neo-AI DLR as the top-level inference API for user applications.
AWS IoT Greengrass is an IoT open source edge runtime and cloud service that provides a secure way to seamlessly connect TI Jacinto powered devices to any AWS service as well as to third-party services. Greengrass enables inference results and data to be sent back to AWS where data can be stored or further analyzed.
Amazon SageMaker Edge Manager facilitates easy management and performance monitoring of ML models on a fleet of TI Jacinto powered edge devices.
