eIQ Toolkit for MCU - Getting Started Labs

[#1]

The tools previously bundled as part of eIQ Toolkit are now released as standalone packages and eIQ Toolkit will no longer be updated after the eIQ Toolkit v1.17 in Q3 2025. Going forward the tools previously included in eIQ Toolkit can now be found at:

• eIQ Neutron SDK now contains the latest versions of the Neutron Converter tool
• eIQ Time Series Studio can now be found in a standalone package
• eIQ Model Creator provides an option for vision based model creation
• eIQ AI Toolkit will provide model optimization functionality (Coming Soon)
• Netron provides TFLite model viewing functionality

This article will remain up for existing users.

---------

eIQ Toolkit enables machine learning development with an intuitive GUI (named eIQ Portal) and development workflow tools, along with command line host tool options as part of the eIQ ML software development environment. Developers can create, optimize, debug and export ML models, as well as import datasets and models, rapidly train and deploy neural network models and ML workloads.

The eIQ Portal provides output TensorFlow Lite models that seamlessly feed into eIQ inference engines like TensorFlow Lite and TensorFlow Lite for Microcontrollers. Using a tool called Model Runner, eIQ Toolkit can also generate runtime insights to help optimize neural network architectures on i.MX RT and i.MX devices.

These labs go over how to use eIQ Portal. It is recommended to do them in the following order:

• Data Import Lab
• Model Runner Lab

The labs are written for using a FRDM-MCXN947 and i.MX RT1170-EVK, but other eIQ supported devices can be used as well. 

• MCX N
• i.MX RT1050
• i.MX RT1060
• i.MX RT1064
• i.MX RT1160
• i.MX RT1170
• i.MX RT1180
• i.MX RT500
• i.MX RT600

For details on the Time Series Studio tool please see the Time Series Studio lab guides.

For