wehub-resource-sync
119 lines
4.3 KiB
C++
119 lines
4.3 KiB
C++
// Copyright (c) 2022 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/prior_box_kernel.h"
|
|
|
|
#include "paddle/phi/backends/xpu/enforce_xpu.h"
|
|
#include "paddle/phi/core/kernel_registry.h"
|
|
#include "paddle/phi/kernels/full_kernel.h"
|
|
#include "paddle/phi/kernels/funcs/eigen/common.h"
|
|
|
|
namespace phi {
|
|
|
|
template <typename T, typename Context>
|
|
void PriorBoxKernel(const Context& dev_ctx,
|
|
const DenseTensor& input,
|
|
const DenseTensor& image,
|
|
const std::vector<float>& min_sizes,
|
|
const std::vector<float>& max_sizes,
|
|
const std::vector<float>& aspect_ratios,
|
|
const std::vector<float>& variances,
|
|
bool flip,
|
|
bool clip,
|
|
float step_w,
|
|
float step_h,
|
|
float offset,
|
|
bool min_max_aspect_ratios_order,
|
|
DenseTensor* out,
|
|
DenseTensor* var) {
|
|
if (input.numel() == 0 || image.numel() == 0) {
|
|
Full<T, Context>(dev_ctx, out->dims(), 0, out);
|
|
Full<T, Context>(dev_ctx, var->dims(), 0, var);
|
|
return;
|
|
}
|
|
|
|
std::vector<float> new_aspect_ratios;
|
|
ExpandAspectRatios(aspect_ratios, flip, &new_aspect_ratios);
|
|
|
|
T new_step_w = static_cast<T>(step_w);
|
|
T new_step_h = static_cast<T>(step_h);
|
|
T new_offset = static_cast<T>(offset);
|
|
|
|
auto img_width = image.dims()[3];
|
|
auto img_height = image.dims()[2];
|
|
|
|
auto feature_width = input.dims()[3];
|
|
auto feature_height = input.dims()[2];
|
|
|
|
T step_width, step_height;
|
|
if (new_step_w == 0 || new_step_h == 0) {
|
|
step_width = static_cast<T>(img_width) / feature_width;
|
|
step_height = static_cast<T>(img_height) / feature_height;
|
|
} else {
|
|
step_width = new_step_w;
|
|
step_height = new_step_h;
|
|
}
|
|
|
|
int64_t num_priors = new_aspect_ratios.size() * min_sizes.size();
|
|
if (max_sizes.size() > 0) {
|
|
num_priors += max_sizes.size();
|
|
}
|
|
|
|
dev_ctx.template Alloc<T>(out);
|
|
dev_ctx.template Alloc<T>(var);
|
|
|
|
auto boxes_data = out->data<T>();
|
|
auto var_data = var->data<T>();
|
|
xpu::VectorParam<float> aspect_ratios_param{
|
|
new_aspect_ratios.data(),
|
|
static_cast<int64_t>(new_aspect_ratios.size()),
|
|
nullptr};
|
|
xpu::VectorParam<float> min_sizes_param{
|
|
min_sizes.data(), static_cast<int64_t>(min_sizes.size()), nullptr};
|
|
xpu::VectorParam<float> max_sizes_param{
|
|
max_sizes.data(), static_cast<int64_t>(max_sizes.size()), nullptr};
|
|
|
|
int ret = xpu::gen_prior_box(dev_ctx.x_context(),
|
|
boxes_data,
|
|
aspect_ratios_param,
|
|
min_sizes_param,
|
|
max_sizes_param,
|
|
feature_height,
|
|
feature_width,
|
|
img_height,
|
|
img_width,
|
|
new_offset,
|
|
step_height,
|
|
step_width,
|
|
clip,
|
|
min_max_aspect_ratios_order);
|
|
PADDLE_ENFORCE_XDNN_SUCCESS(ret, "gen_prior_box");
|
|
|
|
int64_t box_num = feature_height * feature_width * num_priors;
|
|
int64_t vlen = variances.size();
|
|
std::vector<T> var_cpu(vlen * box_num);
|
|
for (int64_t i = 0; i < box_num; ++i) {
|
|
std::copy(variances.begin(), variances.end(), var_cpu.begin() + i * vlen);
|
|
}
|
|
dev_ctx.Wait();
|
|
PADDLE_ENFORCE_XPU_SUCCESS(xpu_memcpy(var_data,
|
|
var_cpu.data(),
|
|
var_cpu.size() * sizeof(T),
|
|
XPUMemcpyKind::XPU_HOST_TO_DEVICE));
|
|
}
|
|
|
|
} // namespace phi
|
|
|
|
PD_REGISTER_KERNEL(prior_box, XPU, ALL_LAYOUT, phi::PriorBoxKernel, float) {}
|