DEEPCRAFT™ Model Converter

Convert, optimize and validate AI models for Infineon microcontrollers

About DEEPCRAFT™ Model Converter

The DEEPCRAFT™ Model Converter converts, optimizes and validates AI models to run on the edge.

Models are converted so they can run on the edge
Models are optimized, which means they will become highly compact and high-performance models, designed to run on Infineon MCUs
Models are validated with a dataset to check that the generated code matches the performance in the original models. Validation is done on the desktop

The DEEPCRAFT™ Model Converter has an easy-to-use Graphical User Interface (GUI) and runs on Windows, Linux and Mac. It includes a Command Line Interface (CLI) to enable integration into Continuous Integration (CI)/Continuous Delivery (CD) pipelines and workflows.

How it works

The DEEPCRAFT™ Model Converter supports the Bring Your Own Model concept (BYOM), which allows users to bring their own AI model and run it on edge devices. The application allows for converting, optimizing and validating AI models to run on the edge, and more specifically, on Infineon MCUs. The application helps machine learning engineers work more efficiently by streamlining the processes for running their models on edge devices.

The DEEPCRAFT™ Model Converter has an easy-to-use Graphical User Interface (GUI) and runs on Windows, Linux and Mac.

It includes a Command Line Interface to enable easy integration into Continuous Integration/Continuous Delivery (CI/CD) pipelines and other scripted workflows. This means that a lot of work can be automated.

DEEPCRAFT™ Model Converter is targeted toward machine learning engineers and embedded engineers who want to convert and optimize proprietary or open-source AI models and run them on the edge.

Running Edge AI on Infineon MCUs

Infineon offers a large range of MCUs that can run Edge AI models. Initially The DEEPCRAFT™ Model Converter will support conversion to the following hardware platforms, with support for more Infineon MCUs to come: