eIQ® ML Software Development Environment

The NXP® eIQ® machine learning (ML) software development environment enables the use of ML algorithms on NXP EdgeVerse™ microcontrollers and microprocessors, including i.MX RT crossover MCUs, and i.MX family application processors. eIQ ML software includes a ML workflow tool called eIQ Toolkit, along with inference engines, neural network compilers and optimized libraries. This software leverages open-source and proprietary technologies and is fully integrated into our MCUXpresso SDK and Yocto development environments, allowing you to develop complete system-level applications with ease.

Machine Learning Workflow Tools

eIQ Toolkit

  • Enables graph-level profiling capability with runtime insights to help optimize neural network architectures
  • Eases ML development with eIQ Portal and command line host tools

eIQ Extensions New

  • Enhanced add-on features and capabilities
  • Model IP protection, training data enhancements, advanced quantization and more

Application Software Pack: ML-Based System State Monitor

  • Develop and deploy neural networks on MCU-based systems
  • Step-by-step guide for state monitoring and failure identification solutions in sensing applications

Supported ML Inference Engines

DeepViewRT™ New

  • Proprietary inference engine for i.MX RT crossover MCUs and i.MX 8 series applications processors

TensorFlow™ Lite Micro

  • Faster and smaller inference engine than the traditional TensorFlow Lite
  • Optimized for running machine learning models i.MX RT crossover MCUs

TensorFlow™ Lite

  • Flexible inference engine for embedded applications
  • Targeting Arm® Cortex®-A, and Cortex-M processor cores and Verisilicon GPUs and NPUs

CMSIS-NN

  • Kernels to create a static, performance- and memory-optimized inference engine for Cortex-M processor cores

Glow

  • Neural Network compiler targeted at i.MX RT Crossover MCUs
  • NXP optimized library integration to deliver high-performance, memory-efficient inferencing

Intellectual Property Aspects of ML

This white paper sets out to add clarity to the legal background and challenges for the IP aspects of ML for data collection and model building and highlights tools NXP offers to help protect ML IP investments.

Read the whitepaper

eIQ Software Ecosystem

eIQ ML Software Development Environment Image

DeepViewRT

A platform-optimized runtime inference engine that enables compact code size for resource-constrained devices.

eIQ-Enabled Devices:
i.MX RT1050, i.MX RT1060, i.MX RT1064, i.MX RT1160, i.MX RT1170, i.MX 8M Plus, i.MX 8M, i.MX 8M Nano, i.MX 8M Nano UL, i.MX 8M Mini.

TensorFlow Lite

Faster and smaller than TensorFlow, TensorFlow Lite enables inference at the edge with lower latency and smaller binary size using open source libraries.

eIQ-Enabled Devices:
i.MX 8M Plus, i.MX 8M, i.MX 8M Nano, i.MX 8M Nano UL, i.MX 8M Mini.

TensorFlow Lite for Microcontrollers

Faster and smaller than TensorFlow, TF Micro is optimized for running machine learning models on resource constrained devices.

eIQ-Enabled Devices:
i.MX RT500, i.MX RT600, i.MX RT1050, i.MX RT1060, i.MX RT1064, i.MX RT1160, i.MX RT1170, i.MX 8M Plus.

Glow Compiler

A machine learning compiler that enables ahead-of-time compilation by converting neural networks into object files, which are then converted into binary images for increased performance and smaller memory footprint.

eIQ-Enabled Devices:
i.MX RT600, i.MX RT1050, i.MX RT1060, i.MX RT1064, i.MX RT1160, i.MX RT1170.

ONNX Runtime

Cross-platform inference and training ML accelerator.

eIQ-Enabled Devices:
i.MX 8M Plus, i.MX 8M, i.MX 8M Nano, i.MX 8M Nano UL, i.MX 8M Mini.

AI and Machine Learning Training Academy

On-demand training modules to help you with your ML application development.

Learn more

Design Resources

Developing AI Solutions at the Edge

Webinar

Watch our webinar recording with Arcturus, Arm, Siemens and Synopsys and learn how to build and secure an Edge AI solution from IP cores, to SoC, to OS, to algorithm, to software applications.

eIQ Machine Learning Software Environment Demo

Video

Image recognition for label identification using the i.MX RT1060 and a TensorFlow Lite model.

Machine Learning Goes Mainstream - eIQ ML Software Development Environment

Video

NXP’s eIQ Machine Learning software development platform makes Machine Learning at the edge possible for all levels of developers - from those just getting started, to the ML experts.® NN for object detection acceleration.

Additional Documents

The Age of Edge Machine Learning Democratization

Discover five key factors developers need to consider when choosing a processing solution for their edge ML projects in this whitepaper from ABI Research.

Sign in to read the whitepaper

Handwritten Digit Recognition Using TensorFlow Lite Micro on i.MX RT Devices

This application note focuses on handwritten digit recognition on embedded systems through deep learning, using i.MX RT MCUs, MCUXpresso SDK and eIQ technology.

Read the application note

eIQ Machine Learning (ML) Software Development Environment

eIQ software leverages inference engines, neural network compilers, optimized libraries and open-source technology allowing AI and ML enablement on edge nodes.

Read the factsheet

Enabling eIQ Core NPU Delegates for i.MX Android Applications

This Application Note describes the deployment of the eIQ’s LiteRT inference engine (formerly known as TensorFlow Lite), with NPU acceleration support on Android OS for i.MX Applications Processors.

Read the application note