This project aims to develop a machine learning model capable of performing handwritten digit recognition on embedded systems such as the Raspberry Pi Pico.
The following components have been implemented:
- Raw data reading/writing
- Data pre-processing
- Feature extraction
sudo picocom -b 115200 /dev/ttyACM0- Topological-preserving downsampling
- Performance optimizations (processing and memory)
- add a .gitignore