We really recommend that fans of Python and NumPy give this one a look over! Posted in Arduino Hacks, Microcontrollers Tagged fft, matrix, microcontroller, micropython, numpy, python, ulab.

This NumPy version performs admirably, clocking in at around 28.77 ns per element -- almost two times faster than the pure Python rendition. Comparison established -- we have a clear winner. However, ...

This is new: TensorFlow 2.18 integrates the current version 2.0 of NumPy and, with Hermetic CUDA, will no longer require local CUDA libraries during the build.