Hello!
I am trying to recreate the inference part of my model-the quantized math- to recreate on an microcontroller. I can’t use normal methods.
Any tips for figuring out which implementation I gotta use?
Also, I am not wrapping my head around a fused activation function. Yall got any good links that explain it like I’m a kid?
Thanks!
Hi @Theo_Benge, For making inference using python you can use tf.lite.Interpreter to load a model and run an inference. Please refer to this document for example. Thank You.