wehub-resource-sync
66 lines
2.2 KiB
C++
66 lines
2.2 KiB
C++
// Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
|
|
//
|
|
// Licensed under the Apache License, Version 2.0 (the "License");
|
|
// you may not use this file except in compliance with the License.
|
|
// You may obtain a copy of the License at
|
|
//
|
|
// http://www.apache.org/licenses/LICENSE-2.0
|
|
//
|
|
// Unless required by applicable law or agreed to in writing, software
|
|
// distributed under the License is distributed on an "AS IS" BASIS,
|
|
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
// See the License for the specific language governing permissions and
|
|
// limitations under the License.
|
|
|
|
#include "paddle/phi/api/ext/dispatch.h"
|
|
#include "paddle/phi/backends/cpu/cpu_context.h"
|
|
#include "paddle/phi/core/kernel_registry.h"
|
|
#include "paddle/phi/kernels/cpu/elementwise.h"
|
|
#include "paddle/phi/kernels/impl/elementwise_kernel_impl.h"
|
|
|
|
namespace phi {
|
|
|
|
template <typename T, typename Context>
|
|
void DivideRawKernel(const Context& dev_ctx,
|
|
const DenseTensor& x,
|
|
const DenseTensor& y,
|
|
int axis,
|
|
DenseTensor* out) {
|
|
// allocate memory for out
|
|
dev_ctx.template Alloc<T>(out);
|
|
if (x.dims() == y.dims() && std::is_floating_point<T>::value) {
|
|
SameDimsElementwiseCompute<SameDimsDivideFunctor<CPUContext, T>>()(
|
|
dev_ctx, x, y, out);
|
|
} else {
|
|
auto x_dims = x.dims();
|
|
auto y_dims = y.dims();
|
|
if (x_dims.size() >= y_dims.size()) {
|
|
funcs::ElementwiseCompute<funcs::DivideFunctor<T>, T>(
|
|
dev_ctx, x, y, funcs::DivideFunctor<T>(), out, axis);
|
|
} else {
|
|
funcs::ElementwiseCompute<funcs::InverseDivideFunctor<T>, T>(
|
|
dev_ctx, x, y, funcs::InverseDivideFunctor<T>(), out, axis);
|
|
}
|
|
}
|
|
}
|
|
|
|
} // namespace phi
|
|
|
|
using complex64 = phi::complex64;
|
|
using complex128 = phi::complex128;
|
|
|
|
// NOTE(chenweihang): using bfloat16 will cause redefine with xpu bfloat16
|
|
// using bfloat16 = phi::bfloat16;
|
|
|
|
PD_REGISTER_KERNEL(divide_raw,
|
|
CPU,
|
|
ALL_LAYOUT,
|
|
phi::DivideRawKernel,
|
|
float,
|
|
double,
|
|
int,
|
|
int64_t,
|
|
bool,
|
|
complex64,
|
|
complex128) {}
|