Machine learning solutions from Microchip Technology
The power of machine learning (ML) is only as good as the data inputs and the data training regiment. And, it can be extremely difficult to get a ML training model built from scratch. In this article, explore the wide range of premium machine learning models, products, systems, and solutions available from Microchip.
Explore the cutting-edge world of Microchip Technology Machine Learning, where you are empowered to create and implement advanced models effortlessly. Whether you're venturing into the realm of Microcontroller Units (MCUs) and Microprocessor Units (MPUs) or seeking specialized tools for image classification and video applications, this comprehensive suite of solutions has you covered.
Build Your Own Model
MCU/MPU Development with MPLAB® Machine Learning Development suite
Embark on your machine learning journey with Microchip's MPLAB® Machine Learning Development suite, seamlessly integrated as a plugin into MPLAB® X IDE. This all-encompassing solution streamlines the entire process—from data collection to model testing—culminating in a tailored knowledge pack for Microchip MCUs/MPUs.
These meticulously designed Machine Learning Evaluation toolkits specifically cater to inertial measurement unit (IMU) applications, focusing on vibration and sensor data analysis. Explore the following exceptional platforms:
- Features the SAMD21G18 Arm® Cortex®-M0+ based 32-bit MCU.
- Equipped with an on-board debugger (nEDBG), ATECC608A CryptoAuthentication™ secure element IC, and ATWINC1510 Wi-Fi® network controller.
- Boasts the SAMD21G18 Arm Cortex-M0+ based 32-bit MCU.
- Includes an on-board debugger (nEDBG), ATECC608A CryptoAuthentication secure element IC, and ATWINC1510 Wi-Fi network controller.
Curiosity Nano Evaluation Kit
Bring Your Own Model
If you have a pre-trained DNN model you can use either Microchip MPU or FPGA based on your use case.
For Audio/Image/Lower Frame Rate Video ML Applications (MPUs):
- Convert TensorFlow models to TensorFlow Lite models using standard APIs
- Utilize MPLAB Harmony V3 to integrate the ML run-time engine (TensorFlow Lite models) and peripherals seamlessly
For Low Power and High Frame Rate Video Applications (FPGAs):
- Microchip FPGAs offer a niche solution for demanding applications
- Leverage the VectorBlox™ Accelerator SDK for easy conversion of high-level Deep Neural Networks to TensorFlow Lite, even without prior FPGA design experience
Evaluation tool kit
Getting Started
- Step 1: Install the VectorBlox™ Software Development Kit (SDK) and read the Programmer’s Guide
- Step 2: Choose between the PolarFire® FPGA or the PolarFire SoC FPGA family
- Step 3: Choose the corresponding video kit and go to the respective GitHub location for instructions to set up the VectorBlox SDK environment:
- Step 4: Download Libero® SoC Design Suite 2023.1
- Step 5: Generate and merge* the Libero SoC Design Suite Silver License and the CoreVectorBlox License
*Refer to section 7.3 of the Libero software quick-start guide to learn how to merge these licenses.
Accelerate your machine learning endeavors with Microchip Technology. Experience the power and ease-of-use of the evaluation kits and unlock the potential of intelligent computing.
Evaluation tool kit
Reference designs for MCU/MPUs:
Reference designs for FPGAs:
Additional Resources:
Article Tags
