(1, 2, 4)instead of
all_readyis a programmer helper function. The purpose is to check inputs whether could be accepted by current iDeep primitive or not, not to check whether dimension is valid for iDeep
mdarrayor not. So, it is no needed to check
bis initialized. Here is a mistake in Intel-Chainer release_v3. https://github.com/intel/chainer/blob/release_v3/chainer/links/connection/linear.py#L128
(2, 4). Programmer could adjust
all_ready. What do you think of that?
[fengyuan, chainer] e.g. Explicitly define
supported_ndim, if an iDeep primitive just supports
if intel64.inputs_all_ready((x, y, z, ), supported_ndim = (1, )):
intel64.ideep.some_primitive(x, y ,z)
Variable.to_intel64, whereas intel-chainer does not use it in
numpy>=1.13.0would be nice.
ideep4py.basic_copyto(dst, src)which seems like an equivalent for
numpy.copyto(dst, src). Is this
basic_copytoconsidered stable? Can we use this interface? (context: chainer/chainer#5009)