Microchip Technology machine learning involves selecting software and hardware kits, reference designs, and silicon platforms where a simplified, easy-to-use environment with advanced performance is required to build a machine-learning model.
Machine learning algorithms available from Microchip are easy to use for collecting and organising data, training neural networks, or implementing optimised edge solutions.
Mouser Electronics carries an extensive Microchip portfolio of silicon devices that includes microcontrollers (MCUs), microprocessors (MPUs), and field-programmable gate arrays (FPGAs). The software toolkits allow the use of popular machine learning frameworks, including TensorFlow, Keras, Caffe, and many others covered by the ONNX umbrella, and those found within TinyML and TensorFlow Lite.
Microchip’s EV18H79A SAMD21 Machine Learning Evaluation Kit with TDK six-axis MEMS features the SAMD21G18 Arm Cortex-M0+ based 32-bit MCU with on-board debugger, an ATECC608A CryptoAuthentication secure element IC and the fully certified ATWINC1510 Wi-Fi network controller.
Figure 1 – Microchip EV45Y33A.
The EV45Y33A SAMD21 Machine Learning Evaluation Kit from Microchip utilises the same MCU, CryptoAuthentication secure element IC and the ATWINC1510 Wi-Fi network controller, but replaces the TDK 6-axis MEMS for a BOSCH inertial measurement unit (IMU).
In addition, both kits have a Microchip MCP9808 high-accuracy temperature and a light sensor on board. Either kit will enable designers to collect data to be used for training and creating machine learning models.
Both kits provide the user with an improved development time and a local learning system which decreases the prevalence of false positives or negatives.
The combination of hardware and software in both these available kits enables users to design various applications, including high-performance AI acceleration cards, self-driving cars, security and surveillance, electronic fences, augmented and virtual reality headsets, drones, robots, satellite imagery, and communication systems.
Designing with PCIe Spectrum Concepts
Editor's Choice Interconnection
PCI Express has become the backbone of modern high-performance systems with each new generation promising higher bandwidth, but that performance comes with a cost.
Read more...Understanding two key tools for cleaner serial data Altron Arrow
Editor's Choice Analogue, Mixed Signal, LSI
Understanding how pre-emphasis and equalisation works, and when to use one over the other, is critical when designing reliable high-speed systems.
Read more...Telemetry powering South Africa’s industry Interlynx-SA
Editor's Choice
As South Africa’s economy evolves, industries are under increasing pressure to improve operational efficiency and cut costs. Telemetry is becoming a vital component of industrial strategy, allowing companies to harness real-time data to optimise processes and reduce waste.
Read more...Engineering in a world that cannot assume connectivity Technews Publishing
Editor's Choice News
Across industrial automation, networking, and defence systems, engineers are rediscovering the importance of resilience and autonomy in an increasingly connected world.
Read more...Designing IoT devices for deterministic LPWAN environments
Editor's Choice Telecoms, Datacoms, Wireless, IoT
Built on Ultra Narrow Band communication technology, the Sigfox network focuses on low power, wide area M2M connectivity rather than maximising data throughput.
Read more...Driving excellence in electronics manufacturing Jemstech
Editor's Choice Manufacturing / Production Technology, Hardware & Services
Jemstech’s reputation for disciplined execution and client-focused service has earned it strong loyalty from companies operating in demanding industries.
Read more...In sync with the line Testerion
Editor's Choice Manufacturing / Production Technology, Hardware & Services
In modern SMT lines, stencil printing must meet two requirements at the same time: it has to ensure a reproducible solder paste volume, while adhering to the specified line cycle time.
While every effort has been made to ensure the accuracy of the information contained herein, the publisher and its agents cannot be held responsible for any errors contained, or any loss incurred as a result. Articles published do not necessarily reflect the views of the publishers. The editor reserves the right to alter or cut copy. Articles submitted are deemed to have been cleared for publication. Advertisements and company contact details are published as provided by the advertiser. Technews Publishing (Pty) Ltd cannot be held responsible for the accuracy or veracity of supplied material.
Figure 2 – Supervised learning flow diagram.
printer friendly version