Closed AndreasMadsen closed 7 years ago
sg_ce says it returns:
sg_ce
"A 1-D Tensor with the same shape as tensor".
Tensor
tensor
(_actually it return an n-D Tensor that has shape tensor.get_shape()[:-1]_.)
tensor.get_shape()[:-1]
sg_train says in takes an:
sg_train
A 0-D Tensor containing the value to minimize.
but somehow this supports an n-D tensor. See for example your ByteNet implementation: https://github.com/buriburisuri/ByteNet/blob/master/train.py#L103L106 sg_train also calls np.mean internally: https://github.com/buriburisuri/sugartensor/blob/master/sugartensor/sg_train.py#L339
np.mean
I'm not really sure that the intended behaviour is, but I would definitely like if sg_train continued to able to take a scalar value.
This is solved after implementing tower support.
sg_ce
says it returns:(_actually it return an n-D
Tensor
that has shapetensor.get_shape()[:-1]
_.)sg_train
says in takes an:but somehow this supports an n-D
tensor
. See for example your ByteNet implementation: https://github.com/buriburisuri/ByteNet/blob/master/train.py#L103L106sg_train
also callsnp.mean
internally: https://github.com/buriburisuri/sugartensor/blob/master/sugartensor/sg_train.py#L339I'm not really sure that the intended behaviour is, but I would definitely like if
sg_train
continued to able to take a scalar value.