search

wavefrontshaping / complexPyTorch

A high-level toolbox for using complex valued neural networks in PyTorch
MIT License
610 stars 148 forks source link

About operating efficiency and convergence #22

Closed echocatzh closed 2 years ago

echocatzh commented 2 years ago

I found some problems when using ComplexGRUCell a. the complex-valued operator runs much lower than the real-valued operator, using such a complex-valued structure is much slower b. the convergence may not be easily controlled, is there any paper supporting the point like complex-valued RNN is better than real-valued RNN in specific cases?

wavefrontshaping commented 2 years ago

complexPyTorch is deprecated as PyTorch now natively supports complex values. You are better of using standard modules.

echocatzh commented 2 years ago

thx for your advice.