Crate microflow

A robust and efficient TinyML inference engine for embedded systems.

Modules

activation

buffer

ops

quantize

tensor

Attribute Macros

model

The entry point of MicroFlow. This attribute-like procedural macro can be placed on structs to implement the predict() function based on the given model. The macro takes as input the path of the model, which must be in the TensorFlow Lite format (.tflite).