chainer icon indicating copy to clipboard operation
chainer copied to clipboard

Support complex types

Open unnonouno opened this issue 7 years ago • 8 comments

Currently Chainer assumes integers and floating point numbers. Now CuPy supports complex numbers. It's time to support complex numbers in Chainer.

  • [ ] Support complex numbers in gradient check
  • [ ] Support complex numbers in various functions

unnonouno avatar Feb 13 '18 05:02 unnonouno

I'm curious about this, do you already have use-cases in mind or is that to broaden the capabilities of Chainer? I see your related ongoing PR, does that target the reproduction of a specific paper?

m3at avatar Feb 26 '18 06:02 m3at

Yes. Some research papers use complex numbers or FFT. And i think it can be applied for speech processing. I want to improve capability of Chainer.

  • https://arxiv.org/abs/1511.06464
  • https://arxiv.org/abs/1611.00035
  • http://www.aclweb.org/anthology/D16-1044

unnonouno avatar Feb 27 '18 02:02 unnonouno

Understood. Thanks for the papers, I didn't know about the last two, seems very interesting! Looking forward to use it in Chainer 👍

m3at avatar Feb 27 '18 04:02 m3at

You should definitely look into https://arxiv.org/pdf/1701.00392.pdf as well.

hknerdgn avatar Mar 02 '18 03:03 hknerdgn

Thank you for a link!! Please thumb up this issue if you guys need this feature.

unnonouno avatar Mar 06 '18 02:03 unnonouno

Should Chainer support "complex32" whose real/imag part is float16? It seems NumPy hasn't yet support "complex32".

toslunar avatar Jun 01 '18 06:06 toslunar

IMO, no, especially for the first implementation.

unnonouno avatar Jun 03 '18 13:06 unnonouno

I would like to work on this. Could anyone guide me with the process? Thanks!

shivam7569 avatar Oct 03 '19 20:10 shivam7569