Machine learning at the edge
This product has been released to the market and is available for purchase. For some products, newer alternatives may be available.
Our processors specialize in enabling machine learning inference at the edge, which helps reduce latency, decrease network bandwidth requirements, and address security and reliability concerns. Our processors incorporate highly efficient hardware accelerators to help you design intelligent applications within low power budgets.
With machine learning, engineers can design solutions to help companies improve energy efficiency in buildings (e.g. smart thermostats and lighting), increase automation throughout factories (e.g. industrial robots and automated guided vehicles), and deliver an enhanced driving experience in vehicles (e.g. forward collision warning systems, adaptive cruise control).
How to get started
- BeagleBone AI- Low cost machine learning evaluation module that is supported by the community.
- J721EXSOMXEVM- Our newest and most capable evaluation module for machine learning and computer vision applications, equipped with the latest C7x DSP core—which uses mixed matrix multiplication—and two C66x DSP cores.
- TMDSID572- Industrial development kit based on the Sitara™ AM5729 processor with machine learning capabilities.
Download Processor SDK for the hardware you selected:
Review the following machine learning reference design demonstrating Texas Instruments Deep Learning (TIDL) software framework to help you get started quickly.
Other helpful resources:
TI Deep Learning (TIDL) software framework
The TIDL software framework leverages a highly optimized neural network implementation on TI processors, making use of hardware accelerators on the device. The TIDL software framework is a set of open-source Linux software packages and tools that enables offloading of deep learning inference to the Embedded Vision Engine (EVE) subsystem, the C66x DSP subsystem, or both. The TIDL software framework is available as part of our free Processor SDK.
Bringing machine learning to embedded systems (white paper)