def backward(ctx, grad_out):
# type: (Any, torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]
r"""
Parameters
----------
grad_out : torch.Tensor
(B, c, n) tensor with gradients of ouputs
Returns
-------
grad_features : torch.Tensor
(B, c, m) tensor with gradients of features
None
None
"""
idx, weight, features = ctx.saved_tensors
m = features.size(2)
grad_features = _ext.three_interpolate_grad(
grad_out.contiguous(), idx, weight, m
)
return grad_features, torch.zeros_like(idx), torch.zeros_like(weight)
def backward(
ctx, grad_out: torch.Tensor
) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
r"""
Parameters
----------
grad_out : torch.Tensor
(B, c, n) tensor with gradients of ouputs
Returns
-------
grad_features : torch.Tensor
(B, c, m) tensor with gradients of features
None
None
"""
idx, weight, m = ctx.three_interpolate_for_backward
B, c, n = grad_out.size()
#grad_features = Variable(torch.cuda.FloatTensor(B, c, m).zero_())
#grad_out_data = grad_out.data.contiguous()
#pointnet2.three_interpolate_grad_wrapper(
# B, c, n, m, grad_out_data, idx, weight, grad_features.data
#)
grad_features = _ext.three_interpolate_grad(grad_out.contiguous(), idx,
weight, m)
return grad_features, None, None
def backward(ctx, grad_out):
# type: (Any, torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]
r"""
Parameters
----------
grad_out : torch.Tensor
(B, c, n) tensor with gradients of ouputs
Returns
-------
grad_features : torch.Tensor
(B, c, m) tensor with gradients of features
None
None
"""
idx, weight, m = ctx.three_interpolate_for_backward
grad_features = _ext.three_interpolate_grad(grad_out.contiguous(), idx,
weight, m)
return grad_features, None, None