There is no standard notation, so if you explain your notation it is just fine to use it. Note that you don't need such a notation if you only present one architecture. In that case, simply state the number of units per layer: "the architecture has three input neurons, one neuron in the hidden layer and one output neuron".

One suggestion, I'd go with letters: "each architecture presented is indicated as NAMEn-h-m, where n is the number of input neurons, h is the number of hidden neurons and m is the number of output neurons."

Good luck!

Hi Eludire,

There is no really standard but there are some identified design patterns. Refer to the neural network zoo by following:

http://www.asimovinstitute.org/neural-network-zoo/

Neverthless, some working groups emerge for providing interoperability solutions addressing the Neural Network Exchange Format subject. Refer to::

• https://thenewstack.io/open-neural-network-exchange-brings-interoperability-to-machine-learning-frameworks/ jointly with https://onnx.ai/
• https://www.anandtech.com/show/12176/khronos-group-launches-nnef-provisional-spec jointly with https://www.khronos.org/nnef/ and https://www.khronos.org/registry/NNEF/specs/1.0/nnef-1.0.pdf

Regards