Releases: Fraunhofer-IMS/AIfES_for_Arduino
Releases · Fraunhofer-IMS/AIfES_for_Arduino
2.2.0 - AIfES for Arduino
Release Notes
With this update AIfES supports Convolutional Neural Networks (CNNs).
New Features:
- Convolutional Neural Nets with following layers:
- Conv2D
- Flatten
- Maxpool
- BatchNorm
- Transposed weights for Dense layers (speed improvement)
Bug Fixes:
- Loss reductions (mean and sum, new default is mean, which is different to previous release)
- Bug fixed in loss calculation in MSE
- Bug fixed in AISTORAGE-Wrapper
- Fixing dytpe bug in inference function for output dtype
- Fix compiler warnings
2.1.1 - AIfES for Arduino
- Bugfixes in aifes_config.h and aifes_config.cpp
- Error correction in documentation
2.1.0 - AIfES for Arduino
Add new features and bugfixes to the Arduino compatible port of the AIfES framework.
What is new:
- Add integer quantization support for 8-bit (Q7) and 32-bit (Q31) integers
- Add AVR controller specific functions to work with data directly from the program memory
- Extend Arm CMSIS support for Cortex controllers
- Print output and logs default to the Serial.print() function
- Refactor configurations
2.0.0 - AIfES for Arduino
First public release of the Arduino compatible port of the AIfES ML framework.