How has your experience with CuPy + Keras been?
CuPy, part of Chainer, is a drop-in replacement for Numpy with GPU acceleration. & Chainer gives you free gradients! https://cupy.chainer.org/
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
can’t one get similar experience using pytorch?
-
I guess if should be closer to replace PyCuda, but I cannot find a comparison anywhere ...
End of conversation
New conversation -
-
-
TFW I first heard the words "drop-in replacement for Numpy"pic.twitter.com/9CP050Fws7
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
What about CuPy vs PyCuda?
-
Maybe cupy is more focused on numpy compatibility so that you can use it as if it is numpy.
- Show replies
New conversation -
-
-
Nice, is there any comparison yet?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Are you claiming that everything Numpy is able to do today can be done with CuPy?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This can already be done using minpy, just replace import numpy as np with import minpy.numpy as nphttps://github.com/dmlc/minpy
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
How is the performance ? Includes optimizations like parallelism etc. ?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.