140 lines
4.8 KiB
C++
140 lines
4.8 KiB
C++
// Copyright (c) 2024 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/kernels/put_along_axis_kernel.h"
|
|
|
|
#include "paddle/common/layout.h"
|
|
#include "paddle/phi/backends/xpu/enforce_xpu.h"
|
|
#include "paddle/phi/core/kernel_registry.h"
|
|
|
|
namespace phi {
|
|
|
|
int64_t get_reduction_mode(const std::string& reduce) {
|
|
if (reduce == "assign") {
|
|
return 0;
|
|
} else if (reduce == "add") {
|
|
return 1;
|
|
} else if (reduce == "multiply" || reduce == "mul") {
|
|
return 2;
|
|
} else if (reduce == "mean") {
|
|
return 3;
|
|
} else if (reduce == "amax") {
|
|
return 4;
|
|
} else if (reduce == "amin") {
|
|
return 5;
|
|
} else {
|
|
PADDLE_THROW(errors::InvalidArgument(
|
|
"can not support reduce: '%s' for put_along_axis kernel, only "
|
|
"support reduce op: 'add', 'assign', 'mul', 'mean', 'amin', 'amax' and "
|
|
"'multiply', the "
|
|
"default reduce "
|
|
"op is 'assign' ",
|
|
reduce));
|
|
}
|
|
}
|
|
|
|
template <typename T, typename Context>
|
|
void PutAlongAxisKernel(const Context& dev_ctx,
|
|
const DenseTensor& x,
|
|
const DenseTensor& index,
|
|
const DenseTensor& value,
|
|
int axis,
|
|
const std::string& reduce,
|
|
bool include_self,
|
|
DenseTensor* out) {
|
|
out->Resize(x.dims());
|
|
dev_ctx.template Alloc<T>(out);
|
|
|
|
if (x.numel() == 0 || index.numel() == 0) return;
|
|
|
|
const auto& index_dtype = index.dtype();
|
|
bool index_dtype_match =
|
|
index_dtype == DataType::INT32 || index_dtype == DataType::INT64;
|
|
PADDLE_ENFORCE_EQ(index_dtype_match,
|
|
true,
|
|
errors::InvalidArgument(
|
|
"Input(Index) holds the wrong type, it holds %s, but "
|
|
"desires to be %s or %s",
|
|
DataTypeToString(index_dtype),
|
|
DataTypeToString(DataType::INT32),
|
|
DataTypeToString(DataType::INT64)));
|
|
|
|
auto input_dtype = x.dtype();
|
|
std::vector<int64_t> x_shape = vectorize<int64_t>(x.dims());
|
|
std::vector<int64_t> index_shape = vectorize<int64_t>(index.dims());
|
|
std::vector<int64_t> value_shape = vectorize<int64_t>(value.dims());
|
|
using XPUType = typename XPUTypeTrait<T>::Type;
|
|
int64_t reduce_mode = get_reduction_mode(reduce);
|
|
|
|
bool invalid_input =
|
|
(input_dtype == DataType::INT32 || input_dtype == DataType::INT64) &&
|
|
(!include_self || reduce_mode > 2);
|
|
PADDLE_ENFORCE_EQ(invalid_input,
|
|
false,
|
|
errors::InvalidArgument(
|
|
"Only support include_self = true and reduce mode: "
|
|
"'add', 'assign' and 'multiply' for int32/int64"));
|
|
|
|
PADDLE_ENFORCE_EQ(index.dims().size(),
|
|
value.dims().size(),
|
|
errors::InvalidArgument(
|
|
"The input(Index) and the input(Value) must have same "
|
|
"rank, but received Index rank is %d, Value rank is %d",
|
|
index.dims().size(),
|
|
value.dims().size()));
|
|
|
|
if (index_dtype == DataType::INT32) {
|
|
int ret = xpu::paddle_put_along_axis(
|
|
dev_ctx.x_context(),
|
|
reinterpret_cast<const XPUType*>(x.data<T>()),
|
|
reinterpret_cast<const XPUType*>(value.data<T>()),
|
|
index.data<int32_t>(),
|
|
reinterpret_cast<XPUType*>(out->data<T>()),
|
|
x_shape,
|
|
value_shape,
|
|
index_shape,
|
|
axis,
|
|
reduce_mode,
|
|
include_self);
|
|
PADDLE_ENFORCE_XDNN_SUCCESS(ret, "paddle_put_along_axis");
|
|
|
|
} else {
|
|
int ret = xpu::paddle_put_along_axis(
|
|
dev_ctx.x_context(),
|
|
reinterpret_cast<const XPUType*>(x.data<T>()),
|
|
reinterpret_cast<const XPUType*>(value.data<T>()),
|
|
index.data<int64_t>(),
|
|
reinterpret_cast<XPUType*>(out->data<T>()),
|
|
x_shape,
|
|
value_shape,
|
|
index_shape,
|
|
axis,
|
|
reduce_mode,
|
|
include_self);
|
|
PADDLE_ENFORCE_XDNN_SUCCESS(ret, "paddle_put_along_axis");
|
|
}
|
|
}
|
|
|
|
} // namespace phi
|
|
|
|
PD_REGISTER_KERNEL(put_along_axis,
|
|
XPU,
|
|
ALL_LAYOUT,
|
|
phi::PutAlongAxisKernel,
|
|
float,
|
|
int64_t,
|
|
int,
|
|
phi::float16,
|
|
phi::bfloat16) {}
|