The Arduino development board allows for thousands upon thousands of different projects; the limit is practically only the maker's imagination, although it does have some physical limitations, such as memory and processing power. However, there are increasingly more products and projects to further extend its capabilities, as is the case with the new release from AIfES.
Now, thanks to this project created by Fraunhofer IMS for Arduino, this open source board will feature a artificial intelligence (AI) framework programmed in C, using the standard GNU GCC compiler libraries. Users will now be able to add AIfES to their Arduino project and integrate it from the library manager from the IDE to use it in your developments, providing the possibility of using machine learning algorithms even in small microcontrollers such as a board Arduino UNO 8-bit.
This will allow developers to create a multitude of IoT (Internet of Things) devices that are more independent of the cloud and can be smarter, while also respecting your privacy, since functions can be run offline from the Arduino board without relying on remote services. Furthermore, the AIfES project has been released under the GNU GPLv3 license , making it completely free, although a paid license is available for commercial projects.
AIfES is very similar to and compatible with Python ML frameworks such as TensorFlow, Keras, and PyTorch, but its functionality is somewhat more limited. However, this released version already supports Feedforward Neural Networks (FNNs) and also allows the activation of integrated functions like ReLU, Sigmoid, and Softmax. Furthermore, the developers are working on a future implementation of ConvNet (Convolutional Neural Networks), which will likely be released soon.
It also includes some common training algorithms , such as SGD (Gradient Descent Optimizer) and Adam Optimizer, among others. In other words, for an 8-bit MCU, it's not bad at all…