TENSAI® Software


Eta Compute’s TENSAI® Flow, the software suite in the TENSAI Platform, de-risks development, allows developers to quickly verify feasibility and proof of concept, and enables seamless design from concept to firmware, speeding the creation of machine learning applications in IoT and low power edge devices. It includes a neural network compiler, a Neural Network Zoo, and middleware comprising FreeRTOS, Hardware Abstraction Level (HAL) and frameworks for sensors, as well as IoT/cloud enablement.

TENSAI Compiler

The TENSAI Flow exclusive compiler delivers the best optimization for neural networks running on Eta Compute’s device as well as the industry’s best power efficiency.

Several examples with documentation are included that show how to use the compiler and Neural Network Zoo.

TENSAI Neural Network Zoo

Our unique Neural Network Zoo accelerates and simplifies development with ready-to-use networks for the most common use cases. These include motion, image, and sound classification. Developers simply train the networks with their data.

TENSAI Middleware, Frameworks, and Firmware

The middleware makes dual core programming seamless by eliminating the need to write customized code to take full advantage of Eta Compute’s multicore architecture.

TENSAI middleware includes an RTOS Kernel, by default FreeRTOS, a HAL for peripherals, and a sensor framework along with drivers for popular sensors. In addition, an executor framework to manage distributed processing and inter-processor communication between the two cores in our TENSAI SoC, and a neural network framework that help the mapping of neural networks to the executor API are provided.

The middleware sits on top of Eta Compute’s chip support package firmware, a software layer that abstracts the hardware registers but is not hardware dependent.

The TENSAI middleware comes with a Project Generator to generate project based on software components and required functions.

Application Examples

With the insight from TENSAI Flow’s real-world applications, developers can see the potential of neural sensor processors precisely in terms of energy efficiency and performance in a variety of field tested examples with unmatched proficiency while preserving total flexibility.