#include "op_plugin/AclOpsInterface.h"
#include "op_plugin/OpApiInterface.h"
#include "op_plugin/utils/op_api_common.h"
namespace op_api {
using npu_preparation = at_npu::native::OpPreparation;
std::tuple<at::Tensor, at::Tensor, at::Tensor> native_batch_norm_backward(
const at::Tensor& grad_out, const at::Tensor& self, const c10::optional<at::Tensor>& weight_opt,
const c10::optional<at::Tensor>& running_mean_opt, const c10::optional<at::Tensor>& running_var_opt,
const c10::optional<at::Tensor>& save_mean_opt, const c10::optional<at::Tensor>& save_invstd_opt, bool train,
double eps, std::array<bool, 3> grad_input_mask) {
DO_COMPATIBILITY(aclnnBatchNormBackward,
acl_op::native_batch_norm_backward(grad_out, self, weight_opt, running_mean_opt, running_var_opt,
save_mean_opt, save_invstd_opt, train, eps, grad_input_mask));
at::Tensor grad_input;
at::Tensor grad_weight;
at::Tensor grad_bias;
if (grad_input_mask[0]) {
grad_input = npu_preparation::apply_tensor_without_format(self.sizes(), self.options());
}
if (grad_input_mask[1]) {
grad_weight = npu_preparation::apply_tensor_without_format({self.size(1)}, self.options().dtype(at::kFloat));
}
if (grad_input_mask[2]) {
grad_bias = npu_preparation::apply_tensor_without_format({self.size(1)}, self.options().dtype(at::kFloat));
}
EXEC_NPU_CMD(aclnnBatchNormBackward, grad_out, self, weight_opt, running_mean_opt, running_var_opt, save_mean_opt,
save_invstd_opt, train, eps, grad_input_mask, grad_input, grad_weight, grad_bias);
return std::make_tuple(grad_input, grad_weight, grad_bias);
}
}