wehub-resource-sync
122 lines
4.1 KiB
C++
122 lines
4.1 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/gumbel_softmax_kernel.h"
|
|
|
|
#include "paddle/phi/core/kernel_registry.h"
|
|
#include "paddle/phi/kernels/funcs/axis_utils.h"
|
|
#include "paddle/phi/kernels/funcs/math_function.h"
|
|
#include "paddle/phi/kernels/impl/gumbel_softmax_kernel_impl.h"
|
|
|
|
namespace phi {
|
|
|
|
template <typename T>
|
|
struct GumbleNoiseGenerator<CPUContext, T> {
|
|
static void Transform(const CPUContext& dev_ctx,
|
|
const T* input_data,
|
|
T* output_data,
|
|
int size_to_axis,
|
|
int size_from_axis,
|
|
const float temperature) {
|
|
// generate uniform random number
|
|
const int64_t size = static_cast<int64_t>(size_to_axis) * size_from_axis;
|
|
std::uniform_real_distribution<T> dist(0.00001, 1);
|
|
auto engine = dev_ctx.GetGenerator()->GetCPUEngine();
|
|
DenseTensor random_tensor;
|
|
random_tensor.Resize({size});
|
|
auto* random_data = dev_ctx.template Alloc<T>(&random_tensor);
|
|
for (int64_t i = 0; i < size; ++i) {
|
|
random_data[i] = dist(*engine);
|
|
}
|
|
|
|
// generate gumbel noise
|
|
DDim dim_2d{size_to_axis, size_from_axis};
|
|
auto gumbel_noise_eigen = EigenMatrix<T>::From(random_tensor, dim_2d);
|
|
gumbel_noise_eigen = -(((-(gumbel_noise_eigen.log())).log()));
|
|
|
|
// add noise
|
|
for (int64_t i = 0; i < size_to_axis * size_from_axis; i++) {
|
|
output_data[i] = (input_data[i] + random_data[i]) / temperature;
|
|
}
|
|
}
|
|
};
|
|
|
|
template <typename T>
|
|
struct OneHotGenerator<CPUContext, T> {
|
|
static void Transform(const CPUContext& dev_ctx,
|
|
const DenseTensor& x,
|
|
DenseTensor* out,
|
|
int axis) {
|
|
DenseTensor index;
|
|
std::vector<int> index_dim;
|
|
const auto rank = x.dims().size();
|
|
const int size_to_axis = funcs::SizeToAxis(axis, x.dims());
|
|
const int size_from_axis = funcs::SizeFromAxis(axis, x.dims());
|
|
const int size_out_axis = funcs::SizeOutAxis(axis, x.dims());
|
|
|
|
for (int i = 0; i < x.dims().size(); i++) {
|
|
if (i != axis) index_dim.push_back(static_cast<int>(x.dims().Get()[i]));
|
|
}
|
|
DDim index_ddim(index_dim.data(), rank - 1);
|
|
index.Resize(index_ddim);
|
|
auto* index_data = dev_ctx.template Alloc<int>(&index);
|
|
|
|
#define CALL_ARG_MINMAX_FUNCTOR(rank) \
|
|
ArgMaxFunctor<CPUContext, T, rank> functor##rank; \
|
|
functor##rank(dev_ctx, *out, &index, axis);
|
|
switch (out->dims().size()) {
|
|
case 1:
|
|
CALL_ARG_MINMAX_FUNCTOR(1);
|
|
break;
|
|
case 2:
|
|
CALL_ARG_MINMAX_FUNCTOR(2);
|
|
break;
|
|
case 3:
|
|
CALL_ARG_MINMAX_FUNCTOR(3);
|
|
break;
|
|
case 4:
|
|
CALL_ARG_MINMAX_FUNCTOR(4);
|
|
break;
|
|
case 5:
|
|
CALL_ARG_MINMAX_FUNCTOR(5);
|
|
break;
|
|
case 6:
|
|
CALL_ARG_MINMAX_FUNCTOR(6);
|
|
break;
|
|
default:
|
|
PADDLE_ENFORCE_LE(
|
|
out->dims().size(),
|
|
6,
|
|
errors::InvalidArgument("gumbel_softmax operator doesn't supports "
|
|
"tensors whose ranks are greater "
|
|
"than 6 in CPU mode."));
|
|
break;
|
|
#undef CALL_ARG_MINMAX_FUNCTOR
|
|
}
|
|
|
|
funcs::set_constant(dev_ctx, out, static_cast<T>(0.0));
|
|
for (int i = 0; i < size_to_axis; i++) {
|
|
for (int j = 0; j < size_out_axis; j++) {
|
|
*(out->data<T>() + i * size_from_axis + j +
|
|
index_data[i * size_out_axis + j] * size_out_axis) = 1.0;
|
|
}
|
|
}
|
|
}
|
|
};
|
|
|
|
} // namespace phi
|
|
|
|
PD_REGISTER_KERNEL(
|
|
gumbel_softmax, CPU, ALL_LAYOUT, phi::GumbelSoftmaxKernel, float, double) {}
|