#include "op_plugin/AclOpsInterface.h"
#include "op_plugin/utils/OpAdapter.h"
namespace acl_op {
using npu_preparation = at_npu::native::OpPreparation;
namespace {
inline void upsample_linear1d_backward_check(const at::Tensor &grad_output, at::IntArrayRef output_size,
at::IntArrayRef input_size)
{
TORCH_CHECK(output_size.size() == 1, "It is expected output_size equals to 1, but got size ", output_size.size(),
OPS_ERROR(ErrCode::PARAM));
TORCH_CHECK(input_size.size() == 3, "It is expected input_size equals to 3, but got size ", input_size.size(),
OPS_ERROR(ErrCode::PARAM));
TORCH_CHECK(grad_output.dim() >= 3, "grad_output dim must larger than 3 ", grad_output.sizes(),
OPS_ERROR(ErrCode::PARAM));
int64_t output_width = grad_output.size(2);
int64_t input_width = input_size[2];
TORCH_CHECK(output_width > 0 && input_width > 0,
"Input and output sizes should be greater than 0, but got input (W: ", input_width,
") and output (W: ", output_width, ")" + OPS_ERROR(ErrCode::VALUE));
}
at::Tensor &upsample_linear1d_backward_out_nocheck(at::Tensor &result, const at::Tensor &grad_output,
at::IntArrayRef input_size, bool align_corners,
c10::optional<double> scales)
{
c10::SmallVector<float, N> sc = {};
TORCH_CHECK(input_size.size() == 3 && input_size[2] != 0, "It is expected input_size equals to 3, but got size ",
input_size.size(), OPS_ERROR(ErrCode::PARAM));
TORCH_CHECK(grad_output.dim() >= 3, "grad_output dim must larger than 3 ", grad_output.sizes(),
OPS_ERROR(ErrCode::PARAM));
if (scales.has_value()) {
sc.push_back(scales.value());
} else {
float temp = float(grad_output.size(3)) / float(input_size[2]);
sc.push_back(temp);
}
string coordinate_transformation_mode = align_corners ? "align_corners" : "half_pixel";
at_npu::native::OpCommand cmd;
cmd.Name("ResizeGradD")
.Input(grad_output, "grads")
.Output(result, "y")
.Attr("original_size", input_size)
.Attr("scales", sc)
.Attr("coordinate_transformation_mode", coordinate_transformation_mode)
.Attr("mode", static_cast<string>("linear"))
.Run();
return result;
}
}
at::Tensor upsample_linear1d_backward(const at::Tensor &grad_output, at::IntArrayRef output_size,
at::IntArrayRef input_size, bool align_corners, c10::optional<double> scales)
{
upsample_linear1d_backward_check(grad_output, output_size, input_size);
at::Tensor grad_output_cp = grad_output;
if (grad_output.scalar_type() != at::ScalarType::Float) {
grad_output_cp = at_npu::native::custom_ops::_npu_dtype_cast(grad_output_cp, at::ScalarType::Float);
}
int64_t N = grad_output_cp.size(0);
int64_t C = grad_output_cp.size(1);
int64_t W = input_size[2];
c10::SmallVector<int64_t, SIZE> output_sizes = {N, C, W};
auto grad_output_4dim = grad_output_cp.unsqueeze(2);
at::Tensor result = npu_preparation::apply_tensor(grad_output_cp, output_sizes);
upsample_linear1d_backward_out_nocheck(result, grad_output_4dim, input_size, align_corners, scales);
if (result.dtype() != grad_output.dtype()) {
result = result.to(grad_output.dtype());
}
return result;
}
}