Fix dtype bug for loss_scale and weight_decay.

1.Change dtype of scale to dtype of grad in loss_scale.py;
2.Change dtype of weight_decay to dtype of weight in optimizer.py.
This commit is contained in:
seatea 2020-03-30 11:46:53 +08:00
parent 930a1fb0a8
commit 6c03542eec
2 changed files with 2 additions and 2 deletions

View File

@ -84,7 +84,7 @@ apply_decay = C.MultitypeFuncGraph("apply_decay")
def _tensor_apply_decay(weight_decay, if_apply, weight, gradient):
"""Get grad with weight_decay."""
if if_apply:
return op_add((gradient, weight * F.scalar_to_array(weight_decay)))
return op_add((gradient, weight * weight_decay))
return gradient

View File

@ -32,7 +32,7 @@ reciprocal = P.Reciprocal()
@_grad_scale.register("Tensor", "Tensor")
def tensor_grad_scale(scale, grad):
return grad * reciprocal(scale)
return grad * F.cast(reciprocal(scale), F.dtype(grad))
class DynamicLossScaleUpdateCell(Cell):