I have noticed that almost all tutorials take the number of neurons as a power of 2. Is there any proper mathematical and well-proven reason for that?
If you sometimes change it to some other odd value, then we get a very long and weird error calling about a dozen things making the traceback of a page. Is there any reason for that?
I had tried it on a text predicting RNN with some GRU and LSTM layers mixed (bidirectional). I changed the no. of neuron units and It also resulted in an error. So, any ideas/theories?