Machine Learning (ML) is turning into a prerequisite for embedded designers who are trying to create or enhance a wide range of products. With its new MPLAB® Machine Learning Development Suite, Microchip Technology (Nasdaq: MCHP) has introduced a comprehensive, integrated workflow for speedy ML model development. This software toolbox can be used to quickly and effectively implement an ML inference across Microchip’s portfolio of microcontrollers (MCUs) and microprocessors (MPUs).
According to Rodger Richey, vice president of Microchip’s Development Systems business unit, “Machine Learning is the new normal for embedded controllers, and utilizing it at the edge allows a product to be efficient, more secure, and use less power than systems that rely on cloud communication for processing.” For the first time, Microchip’s special integrated solution for embedded engineers supports 8- and 16-bit devices in addition to 32-bit MCUs and MPUs, facilitating effective product development.
To help with decision-making, machine learning (ML) employs a variety of algorithmic techniques to extract patterns from massive data sets. Compared to manual processing, it is often faster, simpler to update, and more accurate. Customers of Microchip will use this tool, as one example, to make it possible for predictive maintenance programs to precisely foresee probable problems with machinery used in a range of industrial, manufacturing, consumer, and automotive applications.
The MPLAB Machine Learning Development Suite aids engineers in creating ML models with a minimum environmental impact. The toolkit, which uses AutoML to generate models, removes time-consuming, repetitive operations like extraction, training, validation, and testing. Additionally, it offers model optimizations that adhere to the MPU and MCU’s memory limitations.
The new toolkit offers a comprehensive solution that can be quickly deployed by people with little to no ML programming experience when used in conjunction with the MPLAB X Integrated Development Environment (IDE), which can reduce the expense of hiring data scientists. Additionally, it is advanced enough for more seasoned ML designers to manage.
The option to import a model from TensorFlow Lite and use it in any MPLAB Harmony v3 project is also provided by Microchip. MPLAB Harmony v3 is a fully integrated embedded software development framework that offers adaptable and interoperable software modules to make it easier to develop value-added features and shorten the time it takes for a product to reach the market. Additionally, the VectorBloxTM Accelerator Software Development Kit (SDK) provides PolarFire® FPGAs with the most power-efficient Convolutional Neural Network (CNN)-based AI/ML inference.
The tools required for building and optimizing edge products that conduct ML inference are provided by the MPLAB Machine Learning Development Suite. Learn more about optimizing the development process, reducing expenses, and accelerating time to market using Microchip’s user-friendly ML tools by visiting the company’s page on machine learning solutions.
Cost and accessibility
Depending on the license, prices change. The MPLAB Machine Learning Development Suite is trial-available in a free edition. Contact a Microchip sales representative at www.microchipdirect.com for more details or to make a purchase.
