Microchip: Machine learning toolkit for MCUs from 8bit to 32bit
“Machine Learning is the new normal for embedded controllers, and utilising it at the edge allows a product to be more secure and use less power than systems that rely on cloud communication for processing,” according to Microchip v-p of development systems Rodger Richey.
To be used in combination with the company’s MPLAB X IDE (integrated development environment), the toolkit can be used to implemented artificial intelligence based functions “by those with little to no ML programming knowledge”, claimed the company. “It is also sophisticated enough for more experienced ML designers to control.”
There are three versions of the machine learning kit: free, which must not be used for production, Standard and Pro – see table below
___noise___ 1000
Examples in the user guide include: SAMD21 fan state condition monitoring, AVR DA fan state condition monitoring and SAMD21 gesture recognition.
Optionally, a model from TensorFlow Lite can be used in any MPLAB Harmony v3 project.
| Evaluation License | Standard License | Pro License | |
|---|---|---|---|
| Price | Free | $89/month | $1,990/year |
| Supported | 8, 16 and 32bit MCUs | ||
| Data | 1Gbyte | up to 10Gbyte | |
| Labels | 2500 | unlimited | |
| Output | library | source code | |
| AutoML CPU Time | 5hr/month | 10hr/month | 250hr |
| Application | evaluation only | production | |
The freely viewable on-line user guide includes a step-by-step tutorial on how to use tools within the development kit.
View more : IGBT modules | LCD displays | Electronic Components
Source from: https://www.shunlongwei.com/microchip-machine-learning-toolkit-for-mcus-from-8bit-to-32bit/