Skip to content

P
Paddle

BaiXuePrincess / Paddle
与 Fork 源项目一致

Fork自 PaddlePaddle / Paddle

通知 1
Star 1
Fork 0
P
Paddle
提交 a81fabd3 编写于 作者: D dengkaipeng
浏览文件
操作

fix doc errors. test=develop

上级 cf06e50f
隐藏空白更改
内联 并排
Showing with 154 addition and 149 deletion
paddle/fluid/operators/math/pooling.cc
浏览文件 @ a81fabd3
...@@ -19,16 +19,6 @@ namespace paddle { ...@@ -19,16 +19,6 @@ namespace paddle {
namespace operators { namespace operators {
namespace math { namespace math {
static inline int ADAPT_START_INDEX(int ph, int input_size, int output_size) {
return static_cast<int>(
floor(static_cast<float>(ph * input_size) / output_size));
}
static inline int ADAPT_END_INDEX(int ph, int input_size, int output_size) {
return static_cast<int>(
ceil(static_cast<float>((ph + 1) * input_size) / output_size));
}
/* /*
* All tensors are in NCHW format. * All tensors are in NCHW format.
* Ksize, strides, paddings are two elements. These two elements represent * Ksize, strides, paddings are two elements. These two elements represent
...@@ -67,8 +57,8 @@ class Pool2dFunctor<platform::CPUDeviceContext, PoolProcess, T> { ...@@ -67,8 +57,8 @@ class Pool2dFunctor<platform::CPUDeviceContext, PoolProcess, T> {
for (int c = 0; c < output_channels; ++c) { for (int c = 0; c < output_channels; ++c) {
for (int ph = 0; ph < output_height; ++ph) { for (int ph = 0; ph < output_height; ++ph) {
if (adaptive) { if (adaptive) {
hstart = ADAPT_START_INDEX(ph, input_height, output_height); hstart = AdaptStartIndex(ph, input_height, output_height);
hend = ADAPT_END_INDEX(ph, input_height, output_height); hend = AdaptEndIndex(ph, input_height, output_height);
} else { } else {
hstart = ph * stride_height - padding_height; hstart = ph * stride_height - padding_height;
hend = std::min(hstart + ksize_height, input_height); hend = std::min(hstart + ksize_height, input_height);
...@@ -76,8 +66,8 @@ class Pool2dFunctor<platform::CPUDeviceContext, PoolProcess, T> { ...@@ -76,8 +66,8 @@ class Pool2dFunctor<platform::CPUDeviceContext, PoolProcess, T> {
} }
for (int pw = 0; pw < output_width; ++pw) { for (int pw = 0; pw < output_width; ++pw) {
if (adaptive) { if (adaptive) {
wstart = ADAPT_START_INDEX(pw, input_width, output_width); wstart = AdaptStartIndex(pw, input_width, output_width);
wend = ADAPT_END_INDEX(pw, input_width, output_width); wend = AdaptEndIndex(pw, input_width, output_width);
} else { } else {
wstart = pw * stride_width - padding_width; wstart = pw * stride_width - padding_width;
wend = std::min(wstart + ksize_width, input_width); wend = std::min(wstart + ksize_width, input_width);
...@@ -144,8 +134,8 @@ class Pool2dGradFunctor<platform::CPUDeviceContext, PoolProcess, T> { ...@@ -144,8 +134,8 @@ class Pool2dGradFunctor<platform::CPUDeviceContext, PoolProcess, T> {
for (int c = 0; c < output_channels; ++c) { for (int c = 0; c < output_channels; ++c) {
for (int ph = 0; ph < output_height; ++ph) { for (int ph = 0; ph < output_height; ++ph) {
if (adaptive) { if (adaptive) {
hstart = ADAPT_START_INDEX(ph, input_height, output_height); hstart = AdaptStartIndex(ph, input_height, output_height);
hend = ADAPT_END_INDEX(ph, input_height, output_height); hend = AdaptEndIndex(ph, input_height, output_height);
} else { } else {
hstart = ph * stride_height - padding_height; hstart = ph * stride_height - padding_height;
hend = std::min(hstart + ksize_height, input_height); hend = std::min(hstart + ksize_height, input_height);
...@@ -153,8 +143,8 @@ class Pool2dGradFunctor<platform::CPUDeviceContext, PoolProcess, T> { ...@@ -153,8 +143,8 @@ class Pool2dGradFunctor<platform::CPUDeviceContext, PoolProcess, T> {
} }
for (int pw = 0; pw < output_width; ++pw) { for (int pw = 0; pw < output_width; ++pw) {
if (adaptive) { if (adaptive) {
wstart = ADAPT_START_INDEX(pw, input_width, output_width); wstart = AdaptStartIndex(pw, input_width, output_width);
wend = ADAPT_END_INDEX(pw, input_width, output_width); wend = AdaptEndIndex(pw, input_width, output_width);
} else { } else {
wstart = pw * stride_width - padding_width; wstart = pw * stride_width - padding_width;
wend = std::min(wstart + ksize_width, input_width); wend = std::min(wstart + ksize_width, input_width);
...@@ -319,8 +309,8 @@ class Pool3dFunctor<platform::CPUDeviceContext, PoolProcess, T> { ...@@ -319,8 +309,8 @@ class Pool3dFunctor<platform::CPUDeviceContext, PoolProcess, T> {
for (int c = 0; c < output_channels; ++c) { for (int c = 0; c < output_channels; ++c) {
for (int pd = 0; pd < output_depth; ++pd) { for (int pd = 0; pd < output_depth; ++pd) {
if (adaptive) { if (adaptive) {
dstart = ADAPT_START_INDEX(pd, input_depth, output_depth); dstart = AdaptStartIndex(pd, input_depth, output_depth);
dend = ADAPT_END_INDEX(pd, input_depth, output_depth); dend = AdaptEndIndex(pd, input_depth, output_depth);
} else { } else {
dstart = pd * stride_depth - padding_depth; dstart = pd * stride_depth - padding_depth;
dend = std::min(dstart + ksize_depth, input_depth); dend = std::min(dstart + ksize_depth, input_depth);
...@@ -328,8 +318,8 @@ class Pool3dFunctor<platform::CPUDeviceContext, PoolProcess, T> { ...@@ -328,8 +318,8 @@ class Pool3dFunctor<platform::CPUDeviceContext, PoolProcess, T> {
} }
for (int ph = 0; ph < output_height; ++ph) { for (int ph = 0; ph < output_height; ++ph) {
if (adaptive) { if (adaptive) {
hstart = ADAPT_START_INDEX(ph, input_height, output_height); hstart = AdaptStartIndex(ph, input_height, output_height);
hend = ADAPT_END_INDEX(ph, input_height, output_height); hend = AdaptEndIndex(ph, input_height, output_height);
} else { } else {
hstart = ph * stride_height - padding_height; hstart = ph * stride_height - padding_height;
hend = std::min(hstart + ksize_height, input_height); hend = std::min(hstart + ksize_height, input_height);
...@@ -337,8 +327,8 @@ class Pool3dFunctor<platform::CPUDeviceContext, PoolProcess, T> { ...@@ -337,8 +327,8 @@ class Pool3dFunctor<platform::CPUDeviceContext, PoolProcess, T> {
} }
for (int pw = 0; pw < output_width; ++pw) { for (int pw = 0; pw < output_width; ++pw) {
if (adaptive) { if (adaptive) {
wstart = ADAPT_START_INDEX(pw, input_width, output_width); wstart = AdaptStartIndex(pw, input_width, output_width);
wend = ADAPT_END_INDEX(pw, input_width, output_width); wend = AdaptEndIndex(pw, input_width, output_width);
} else { } else {
wstart = pw * stride_width - padding_width; wstart = pw * stride_width - padding_width;
wend = std::min(wstart + ksize_width, input_width); wend = std::min(wstart + ksize_width, input_width);
...@@ -417,8 +407,8 @@ class Pool3dGradFunctor<platform::CPUDeviceContext, PoolProcess, T> { ...@@ -417,8 +407,8 @@ class Pool3dGradFunctor<platform::CPUDeviceContext, PoolProcess, T> {
for (int c = 0; c < output_channels; ++c) { for (int c = 0; c < output_channels; ++c) {
for (int pd = 0; pd < output_depth; ++pd) { for (int pd = 0; pd < output_depth; ++pd) {
if (adaptive) { if (adaptive) {
dstart = ADAPT_START_INDEX(pd, input_depth, output_depth); dstart = AdaptStartIndex(pd, input_depth, output_depth);
dend = ADAPT_END_INDEX(pd, input_depth, output_depth); dend = AdaptEndIndex(pd, input_depth, output_depth);
} else { } else {
dstart = pd * stride_depth - padding_depth; dstart = pd * stride_depth - padding_depth;
dend = std::min(dstart + ksize_depth, input_depth); dend = std::min(dstart + ksize_depth, input_depth);
...@@ -426,8 +416,8 @@ class Pool3dGradFunctor<platform::CPUDeviceContext, PoolProcess, T> { ...@@ -426,8 +416,8 @@ class Pool3dGradFunctor<platform::CPUDeviceContext, PoolProcess, T> {
} }
for (int ph = 0; ph < output_height; ++ph) { for (int ph = 0; ph < output_height; ++ph) {
if (adaptive) { if (adaptive) {
hstart = ADAPT_START_INDEX(ph, input_height, output_height); hstart = AdaptStartIndex(ph, input_height, output_height);
hend = ADAPT_END_INDEX(ph, input_height, output_height); hend = AdaptEndIndex(ph, input_height, output_height);
} else { } else {
hstart = ph * stride_height - padding_height; hstart = ph * stride_height - padding_height;
hend = std::min(hstart + ksize_height, input_height); hend = std::min(hstart + ksize_height, input_height);
...@@ -435,8 +425,8 @@ class Pool3dGradFunctor<platform::CPUDeviceContext, PoolProcess, T> { ...@@ -435,8 +425,8 @@ class Pool3dGradFunctor<platform::CPUDeviceContext, PoolProcess, T> {
} }
for (int pw = 0; pw < output_width; ++pw) { for (int pw = 0; pw < output_width; ++pw) {
if (adaptive) { if (adaptive) {
wstart = ADAPT_START_INDEX(pw, input_width, output_width); wstart = AdaptStartIndex(pw, input_width, output_width);
wend = ADAPT_END_INDEX(pw, input_width, output_width); wend = AdaptEndIndex(pw, input_width, output_width);
} else { } else {
wstart = pw * stride_width - padding_width; wstart = pw * stride_width - padding_width;
wend = std::min(wstart + ksize_width, input_width); wend = std::min(wstart + ksize_width, input_width);
...@@ -615,8 +605,8 @@ class MaxPool2dWithIndexFunctor<platform::CPUDeviceContext, T1, T2> { ...@@ -615,8 +605,8 @@ class MaxPool2dWithIndexFunctor<platform::CPUDeviceContext, T1, T2> {
for (int c = 0; c < output_channels; ++c) { for (int c = 0; c < output_channels; ++c) {
for (int ph = 0; ph < output_height; ++ph) { for (int ph = 0; ph < output_height; ++ph) {
if (adaptive) { if (adaptive) {
hstart = ADAPT_START_INDEX(ph, input_height, output_height); hstart = AdaptStartIndex(ph, input_height, output_height);
hend = ADAPT_END_INDEX(ph, input_height, output_height); hend = AdaptEndIndex(ph, input_height, output_height);
} else { } else {
hstart = ph * stride_height - padding_height; hstart = ph * stride_height - padding_height;
hend = std::min(hstart + ksize_height, input_height); hend = std::min(hstart + ksize_height, input_height);
...@@ -624,8 +614,8 @@ class MaxPool2dWithIndexFunctor<platform::CPUDeviceContext, T1, T2> { ...@@ -624,8 +614,8 @@ class MaxPool2dWithIndexFunctor<platform::CPUDeviceContext, T1, T2> {
} }
for (int pw = 0; pw < output_width; ++pw) { for (int pw = 0; pw < output_width; ++pw) {
if (adaptive) { if (adaptive) {
wstart = ADAPT_START_INDEX(pw, input_width, output_width); wstart = AdaptStartIndex(pw, input_width, output_width);
wend = ADAPT_END_INDEX(pw, input_width, output_width); wend = AdaptEndIndex(pw, input_width, output_width);
} else { } else {
wstart = pw * stride_width - padding_width; wstart = pw * stride_width - padding_width;
wend = std::min(wstart + ksize_width, input_width); wend = std::min(wstart + ksize_width, input_width);
...@@ -753,8 +743,8 @@ class MaxPool3dWithIndexFunctor<platform::CPUDeviceContext, T1, T2> { ...@@ -753,8 +743,8 @@ class MaxPool3dWithIndexFunctor<platform::CPUDeviceContext, T1, T2> {
for (int c = 0; c < output_channels; ++c) { for (int c = 0; c < output_channels; ++c) {
for (int pd = 0; pd < output_depth; ++pd) { for (int pd = 0; pd < output_depth; ++pd) {
if (adaptive) { if (adaptive) {
dstart = ADAPT_START_INDEX(pd, input_depth, output_depth); dstart = AdaptStartIndex(pd, input_depth, output_depth);
dend = ADAPT_END_INDEX(pd, input_depth, output_depth); dend = AdaptEndIndex(pd, input_depth, output_depth);
} else { } else {
dstart = pd * stride_depth - padding_depth; dstart = pd * stride_depth - padding_depth;
dend = std::min(dstart + ksize_depth, input_depth); dend = std::min(dstart + ksize_depth, input_depth);
...@@ -762,8 +752,8 @@ class MaxPool3dWithIndexFunctor<platform::CPUDeviceContext, T1, T2> { ...@@ -762,8 +752,8 @@ class MaxPool3dWithIndexFunctor<platform::CPUDeviceContext, T1, T2> {
} }
for (int ph = 0; ph < output_height; ++ph) { for (int ph = 0; ph < output_height; ++ph) {
if (adaptive) { if (adaptive) {
hstart = ADAPT_START_INDEX(ph, input_height, output_height); hstart = AdaptStartIndex(ph, input_height, output_height);
hend = ADAPT_END_INDEX(ph, input_height, output_height); hend = AdaptEndIndex(ph, input_height, output_height);
} else { } else {
hstart = ph * stride_height - padding_height; hstart = ph * stride_height - padding_height;
hend = std::min(hstart + ksize_height, input_height); hend = std::min(hstart + ksize_height, input_height);
...@@ -771,8 +761,8 @@ class MaxPool3dWithIndexFunctor<platform::CPUDeviceContext, T1, T2> { ...@@ -771,8 +761,8 @@ class MaxPool3dWithIndexFunctor<platform::CPUDeviceContext, T1, T2> {
} }
for (int pw = 0; pw < output_width; ++pw) { for (int pw = 0; pw < output_width; ++pw) {
if (adaptive) { if (adaptive) {
wstart = ADAPT_START_INDEX(pw, input_width, output_width); wstart = AdaptStartIndex(pw, input_width, output_width);
wend = ADAPT_END_INDEX(pw, input_width, output_width); wend = AdaptEndIndex(pw, input_width, output_width);
} else { } else {
wstart = pw * stride_width - padding_width; wstart = pw * stride_width - padding_width;
wend = std::min(wstart + ksize_width, input_width); wend = std::min(wstart + ksize_width, input_width);
......
paddle/fluid/operators/math/pooling.cu
浏览文件 @ a81fabd3
...@@ -21,18 +21,6 @@ namespace paddle { ...@@ -21,18 +21,6 @@ namespace paddle {
namespace operators { namespace operators {
namespace math { namespace math {
__device__ __forceinline__ int ADAPT_START_INDEX(int ph, int input_size,
int output_size) {
return static_cast<int>(
floor(static_cast<double>(ph * input_size) / output_size));
}
__device__ __forceinline__ int ADAPT_END_INDEX(int ph, int input_size,
int output_size) {
return static_cast<int>(
ceil(static_cast<double>((ph + 1) * input_size) / output_size));
}
template <typename PoolProcess, typename T> template <typename PoolProcess, typename T>
__global__ void KernelPool2D(const int nthreads, const T* input_data, __global__ void KernelPool2D(const int nthreads, const T* input_data,
const int channels, const int input_height, const int channels, const int input_height,
...@@ -52,11 +40,11 @@ __global__ void KernelPool2D(const int nthreads, const T* input_data, ...@@ -52,11 +40,11 @@ __global__ void KernelPool2D(const int nthreads, const T* input_data,
int hstart, hend; int hstart, hend;
int wstart, wend; int wstart, wend;
if (adaptive) { if (adaptive) {
hstart = ADAPT_START_INDEX(ph, input_height, output_height); hstart = AdaptStartIndex(ph, input_height, output_height);
hend = ADAPT_END_INDEX(ph, input_height, output_height); hend = AdaptEndIndex(ph, input_height, output_height);
wstart = ADAPT_START_INDEX(pw, input_width, output_width); wstart = AdaptStartIndex(pw, input_width, output_width);
wend = ADAPT_END_INDEX(pw, input_width, output_width); wend = AdaptEndIndex(pw, input_width, output_width);
} else { } else {
hstart = ph * stride_height - padding_height; hstart = ph * stride_height - padding_height;
hend = min(hstart + ksize_height, input_height); hend = min(hstart + ksize_height, input_height);
...@@ -91,28 +79,29 @@ __global__ void KernelPool2DGrad( ...@@ -91,28 +79,29 @@ __global__ void KernelPool2DGrad(
PoolProcess pool_process, bool exclusive, bool adaptive, T* input_grad) { PoolProcess pool_process, bool exclusive, bool adaptive, T* input_grad) {
for (int index = blockIdx.x * blockDim.x + threadIdx.x; index < nthreads; for (int index = blockIdx.x * blockDim.x + threadIdx.x; index < nthreads;
index += blockDim.x * gridDim.x) { index += blockDim.x * gridDim.x) {
int offsetW = index % input_width + padding_width; int w_offset = index % input_width + padding_width;
int offsetH = (index / input_width) % input_height + padding_height; int h_offset = (index / input_width) % input_height + padding_height;
int offsetC = (index / input_width / input_height) % channels; int offsetC = (index / input_width / input_height) % channels;
int batch_idx = index / input_width / input_height / channels; int batch_idx = index / input_width / input_height / channels;
int phstart, phend; int phstart, phend;
int pwstart, pwend; int pwstart, pwend;
if (adaptive) { if (adaptive) {
phstart = offsetH * output_height / input_height; phstart = h_offset * output_height / input_height;
phend = phend =
min((offsetH + 1) * output_height / input_height + 1, output_height); min((h_offset + 1) * output_height / input_height + 1, output_height);
pwstart = offsetW * output_width / input_width; pwstart = w_offset * output_width / input_width;
pwend = min((offsetW + 1) * output_width / input_width + 1, output_width); pwend =
min((w_offset + 1) * output_width / input_width + 1, output_width);
} else { } else {
phstart = (offsetH < ksize_height) phstart = (h_offset < ksize_height)
? 0 ? 0
: (offsetH - ksize_height) / stride_height + 1; : (h_offset - ksize_height) / stride_height + 1;
pwstart = (offsetW < ksize_width) pwstart = (w_offset < ksize_width)
? 0 ? 0
: (offsetW - ksize_width) / stride_width + 1; : (w_offset - ksize_width) / stride_width + 1;
phend = min(offsetH / stride_height + 1, output_height); phend = min(h_offset / stride_height + 1, output_height);
pwend = min(offsetW / stride_width + 1, output_width); pwend = min(w_offset / stride_width + 1, output_width);
} }
T gradient = 0; T gradient = 0;
T input = input_data[index]; T input = input_data[index];
...@@ -414,14 +403,14 @@ __global__ void KernelPool3D( ...@@ -414,14 +403,14 @@ __global__ void KernelPool3D(
int hstart, hend; int hstart, hend;
int wstart, wend; int wstart, wend;
if (adaptive) { if (adaptive) {
dstart = ADAPT_START_INDEX(pd, input_depth, output_depth); dstart = AdaptStartIndex(pd, input_depth, output_depth);
dend = ADAPT_END_INDEX(pd, input_depth, output_depth); dend = AdaptEndIndex(pd, input_depth, output_depth);
hstart = ADAPT_START_INDEX(ph, input_height, output_height); hstart = AdaptStartIndex(ph, input_height, output_height);
hend = ADAPT_END_INDEX(ph, input_height, output_height); hend = AdaptEndIndex(ph, input_height, output_height);
wstart = ADAPT_START_INDEX(pw, input_width, output_width); wstart = AdaptStartIndex(pw, input_width, output_width);
wend = ADAPT_END_INDEX(pw, input_width, output_width); wend = AdaptEndIndex(pw, input_width, output_width);
} else { } else {
dstart = pd * stride_depth - padding_depth; dstart = pd * stride_depth - padding_depth;
hstart = ph * stride_height - padding_height; hstart = ph * stride_height - padding_height;
...@@ -464,9 +453,9 @@ __global__ void KernelPool3DGrad( ...@@ -464,9 +453,9 @@ __global__ void KernelPool3DGrad(
bool exclusive, bool adaptive, T* input_grad) { bool exclusive, bool adaptive, T* input_grad) {
for (int index = blockIdx.x * blockDim.x + threadIdx.x; index < nthreads; for (int index = blockIdx.x * blockDim.x + threadIdx.x; index < nthreads;
index += blockDim.x * gridDim.x) { index += blockDim.x * gridDim.x) {
int offsetW = index % input_width + padding_width; int w_offset = index % input_width + padding_width;
int offsetH = (index / input_width) % input_height + padding_height; int h_offset = (index / input_width) % input_height + padding_height;
int offsetD = int d_offset =
(index / input_width / input_height) % input_depth + padding_depth; (index / input_width / input_height) % input_depth + padding_depth;
int offsetC = (index / input_width / input_height / input_depth) % channels; int offsetC = (index / input_width / input_height / input_depth) % channels;
int batch_idx = index / input_width / input_height / input_depth / channels; int batch_idx = index / input_width / input_height / input_depth / channels;
...@@ -475,26 +464,28 @@ __global__ void KernelPool3DGrad( ...@@ -475,26 +464,28 @@ __global__ void KernelPool3DGrad(
int phstart, phend; int phstart, phend;
int pwstart, pwend; int pwstart, pwend;
if (adaptive) { if (adaptive) {
pdstart = offsetD * output_depth / input_depth; pdstart = d_offset * output_depth / input_depth;
pdend = min((offsetD + 1) * output_depth / input_depth + 1, output_depth); pdend =
phstart = offsetH * output_height / input_height; min((d_offset + 1) * output_depth / input_depth + 1, output_depth);
phstart = h_offset * output_height / input_height;
phend = phend =
min((offsetH + 1) * output_height / input_height + 1, output_height); min((h_offset + 1) * output_height / input_height + 1, output_height);
pwstart = offsetW * output_width / input_width; pwstart = w_offset * output_width / input_width;
pwend = min((offsetW + 1) * output_width / input_width + 1, output_width); pwend =
min((w_offset + 1) * output_width / input_width + 1, output_width);
} else { } else {
pdstart = (offsetD < ksize_depth) pdstart = (d_offset < ksize_depth)
? 0 ? 0
: (offsetD - ksize_depth) / stride_depth + 1; : (d_offset - ksize_depth) / stride_depth + 1;
phstart = (offsetH < ksize_height) phstart = (h_offset < ksize_height)
? 0 ? 0
: (offsetH - ksize_height) / stride_height + 1; : (h_offset - ksize_height) / stride_height + 1;
pwstart = (offsetW < ksize_width) pwstart = (w_offset < ksize_width)
? 0 ? 0
: (offsetW - ksize_width) / stride_width + 1; : (w_offset - ksize_width) / stride_width + 1;
pdend = min((offsetD) / stride_depth + 1, output_depth); pdend = min((d_offset) / stride_depth + 1, output_depth);
phend = min((offsetH) / stride_height + 1, output_height); phend = min((h_offset) / stride_height + 1, output_height);
pwend = min((offsetW) / stride_width + 1, output_width); pwend = min((w_offset) / stride_width + 1, output_width);
} }
T gradient = 0; T gradient = 0;
...@@ -795,11 +786,11 @@ __global__ void KernelMaxPool2dWithIdx( ...@@ -795,11 +786,11 @@ __global__ void KernelMaxPool2dWithIdx(
int hstart, hend; int hstart, hend;
int wstart, wend; int wstart, wend;
if (adaptive) { if (adaptive) {
hstart = ADAPT_START_INDEX(ph, input_height, output_height); hstart = AdaptStartIndex(ph, input_height, output_height);
hend = ADAPT_END_INDEX(ph, input_height, output_height); hend = AdaptEndIndex(ph, input_height, output_height);
wstart = ADAPT_START_INDEX(pw, input_width, output_width); wstart = AdaptStartIndex(pw, input_width, output_width);
wend = ADAPT_END_INDEX(pw, input_width, output_width); wend = AdaptEndIndex(pw, input_width, output_width);
} else { } else {
hstart = ph * stride_height - padding_height; hstart = ph * stride_height - padding_height;
hend = min(hstart + ksize_height, input_height); hend = min(hstart + ksize_height, input_height);
...@@ -837,35 +828,36 @@ __global__ void KernelMaxPool2DWithIdxGrad( ...@@ -837,35 +828,36 @@ __global__ void KernelMaxPool2DWithIdxGrad(
T1* input_grad) { T1* input_grad) {
for (int index = blockIdx.x * blockDim.x + threadIdx.x; index < nthreads; for (int index = blockIdx.x * blockDim.x + threadIdx.x; index < nthreads;
index += blockDim.x * gridDim.x) { index += blockDim.x * gridDim.x) {
int offsetW = index % input_width; int w_offset = index % input_width;
int offsetH = (index / input_width) % input_height; int h_offset = (index / input_width) % input_height;
int offsetC = (index / input_width / input_height) % channels; int offsetC = (index / input_width / input_height) % channels;
int batch_idx = index / input_width / input_height / channels; int batch_idx = index / input_width / input_height / channels;
int phstart, phend; int phstart, phend;
int pwstart, pwend; int pwstart, pwend;
if (adaptive) { if (adaptive) {
phstart = offsetH * output_height / input_height; phstart = h_offset * output_height / input_height;
phend = phend =
min((offsetH + 1) * output_height / input_height + 1, output_height); min((h_offset + 1) * output_height / input_height + 1, output_height);
pwstart = offsetW * output_width / input_width; pwstart = w_offset * output_width / input_width;
pwend = min((offsetW + 1) * output_width / input_width + 1, output_width); pwend =
min((w_offset + 1) * output_width / input_width + 1, output_width);
} else { } else {
phstart = phstart =
(offsetH + padding_height < ksize_height) (h_offset + padding_height < ksize_height)
? 0 ? 0
: (offsetH + padding_height - ksize_height) / stride_height + 1; : (h_offset + padding_height - ksize_height) / stride_height + 1;
pwstart = pwstart =
(offsetW + padding_width < ksize_width) (w_offset + padding_width < ksize_width)
? 0 ? 0
: (offsetW + padding_width - ksize_width) / stride_width + 1; : (w_offset + padding_width - ksize_width) / stride_width + 1;
phend = phend =
min((offsetH + padding_height) / stride_height + 1, output_height); min((h_offset + padding_height) / stride_height + 1, output_height);
pwend = min((offsetW + padding_width) / stride_width + 1, output_width); pwend = min((w_offset + padding_width) / stride_width + 1, output_width);
} }
T1 gradient = 0; T1 gradient = 0;
int input_current_featuremap_idx = offsetH * input_width + offsetW; int input_current_featuremap_idx = h_offset * input_width + w_offset;
int output_idx = int output_idx =
(batch_idx * channels + offsetC) * output_height * output_width; (batch_idx * channels + offsetC) * output_height * output_width;
...@@ -1000,14 +992,14 @@ __global__ void KernelMaxPool3DWithIdx( ...@@ -1000,14 +992,14 @@ __global__ void KernelMaxPool3DWithIdx(
int hstart, hend; int hstart, hend;
int wstart, wend; int wstart, wend;
if (adaptive) { if (adaptive) {
dstart = ADAPT_START_INDEX(pd, input_depth, output_depth); dstart = AdaptStartIndex(pd, input_depth, output_depth);
dend = ADAPT_END_INDEX(pd, input_depth, output_depth); dend = AdaptEndIndex(pd, input_depth, output_depth);
hstart = ADAPT_START_INDEX(ph, input_height, output_height); hstart = AdaptStartIndex(ph, input_height, output_height);
hend = ADAPT_END_INDEX(ph, input_height, output_height); hend = AdaptEndIndex(ph, input_height, output_height);
wstart = ADAPT_START_INDEX(pw, input_width, output_width); wstart = AdaptStartIndex(pw, input_width, output_width);
wend = ADAPT_END_INDEX(pw, input_width, output_width); wend = AdaptEndIndex(pw, input_width, output_width);
} else { } else {
dstart = pd * stride_depth - padding_depth; dstart = pd * stride_depth - padding_depth;
hstart = ph * stride_height - padding_height; hstart = ph * stride_height - padding_height;
...@@ -1051,9 +1043,9 @@ __global__ void KernelMaxPool3DWithIdxGrad( ...@@ -1051,9 +1043,9 @@ __global__ void KernelMaxPool3DWithIdxGrad(
const int padding_width, bool adaptive, T1* input_grad) { const int padding_width, bool adaptive, T1* input_grad) {
for (int index = blockIdx.x * blockDim.x + threadIdx.x; index < nthreads; for (int index = blockIdx.x * blockDim.x + threadIdx.x; index < nthreads;
index += blockDim.x * gridDim.x) { index += blockDim.x * gridDim.x) {
int offsetW = index % input_width; int w_offset = index % input_width;
int offsetH = (index / input_width) % input_height; int h_offset = (index / input_width) % input_height;
int offsetD = (index / input_width / input_height) % input_depth; int d_offset = (index / input_width / input_height) % input_depth;
int offsetC = (index / input_width / input_height / input_depth) % channels; int offsetC = (index / input_width / input_height / input_depth) % channels;
int batch_idx = index / input_width / input_height / input_depth / channels; int batch_idx = index / input_width / input_height / input_depth / channels;
...@@ -1061,35 +1053,37 @@ __global__ void KernelMaxPool3DWithIdxGrad( ...@@ -1061,35 +1053,37 @@ __global__ void KernelMaxPool3DWithIdxGrad(
int phstart, phend; int phstart, phend;
int pwstart, pwend; int pwstart, pwend;
if (adaptive) { if (adaptive) {
pdstart = offsetD * output_depth / input_depth; pdstart = d_offset * output_depth / input_depth;
pdend = min((offsetD + 1) * output_depth / input_depth + 1, output_depth); pdend =
phstart = offsetH * output_height / input_height; min((d_offset + 1) * output_depth / input_depth + 1, output_depth);
phstart = h_offset * output_height / input_height;
phend = phend =
min((offsetH + 1) * output_height / input_height + 1, output_height); min((h_offset + 1) * output_height / input_height + 1, output_height);
pwstart = offsetW * output_width / input_width; pwstart = w_offset * output_width / input_width;
pwend = min((offsetW + 1) * output_width / input_width + 1, output_width); pwend =
min((w_offset + 1) * output_width / input_width + 1, output_width);
} else { } else {
pdstart = pdstart =
(offsetD + padding_depth < ksize_depth) (d_offset + padding_depth < ksize_depth)
? 0 ? 0
: (offsetD + padding_depth - ksize_depth) / stride_depth + 1; : (d_offset + padding_depth - ksize_depth) / stride_depth + 1;
phstart = phstart =
(offsetH + padding_height < ksize_height) (h_offset + padding_height < ksize_height)
? 0 ? 0
: (offsetH + padding_height - ksize_height) / stride_height + 1; : (h_offset + padding_height - ksize_height) / stride_height + 1;
pwstart = pwstart =
(offsetW + padding_width < ksize_width) (w_offset + padding_width < ksize_width)
? 0 ? 0
: (offsetW + padding_width - ksize_width) / stride_width + 1; : (w_offset + padding_width - ksize_width) / stride_width + 1;
pdend = min((offsetD + padding_depth) / stride_depth + 1, output_depth); pdend = min((d_offset + padding_depth) / stride_depth + 1, output_depth);
phend = phend =
min((offsetH + padding_height) / stride_height + 1, output_height); min((h_offset + padding_height) / stride_height + 1, output_height);
pwend = min((offsetW + padding_width) / stride_width + 1, output_width); pwend = min((w_offset + padding_width) / stride_width + 1, output_width);
} }
T1 gradient = 0; T1 gradient = 0;
int input_current_feature_map_idx = int input_current_feature_map_idx =
(offsetD * input_height + offsetH) * input_width + offsetW; (d_offset * input_height + h_offset) * input_width + w_offset;
int output_idx = (batch_idx * channels + offsetC) * output_depth * int output_idx = (batch_idx * channels + offsetC) * output_depth *
output_height * output_width; output_height * output_width;
mask += output_idx; mask += output_idx;
......
paddle/fluid/operators/math/pooling.h
浏览文件 @ a81fabd3
...@@ -68,6 +68,18 @@ class AvgPoolGrad { ...@@ -68,6 +68,18 @@ class AvgPoolGrad {
} }
}; };
/* used for adaptive pool to calculate start and end index of each divided grid
*/
HOSTDEVICE inline int AdaptStartIndex(int ph, int input_size, int output_size) {
return static_cast<int>(
floor(static_cast<double>(ph * input_size) / output_size));
}
HOSTDEVICE inline int AdaptEndIndex(int ph, int input_size, int output_size) {
return static_cast<int>(
ceil(static_cast<double>((ph + 1) * input_size) / output_size));
}
/* /*
* \brief Getting pooling results, and calculating gradient. * \brief Getting pooling results, and calculating gradient.
* *
......
python/paddle/fluid/layers/nn.py
浏览文件 @ a81fabd3
...@@ -2506,7 +2506,7 @@ def adaptive_pool2d(input, ...@@ -2506,7 +2506,7 @@ def adaptive_pool2d(input,
pool_size, pool_size,
pool_type="max", pool_type="max",
require_index=False, require_index=False,
use_cudnn=True, use_cudnn=False,
name=None): name=None):
""" """
${comment} ${comment}
...@@ -2521,7 +2521,7 @@ def adaptive_pool2d(input, ...@@ -2521,7 +2521,7 @@ def adaptive_pool2d(input,
pool_type: ${pooling_type_comment} pool_type: ${pooling_type_comment}
require_index (bool): If true, the index of max pooling point along with outputs. require_index (bool): If true, the index of max pooling point along with outputs.
it cannot be set in average pooling type. it cannot be set in average pooling type.
use_cudnn (bool): ${use_cudnn_comment} use_cudnn (bool, default False): adaptive pool currently not supported in cudnn.
name (str|None): A name for this layer(optional). If set None, the name (str|None): A name for this layer(optional). If set None, the
layer will be named automatically. layer will be named automatically.
...@@ -2531,6 +2531,7 @@ def adaptive_pool2d(input, ...@@ -2531,6 +2531,7 @@ def adaptive_pool2d(input,
Raises: Raises:
ValueError: 'pool_type' is not 'max' nor 'avg'. ValueError: 'pool_type' is not 'max' nor 'avg'.
ValueError: 'use_cudnn' is not a bool value. ValueError: 'use_cudnn' is not a bool value.
ValueError: adaptive pool currently not supported in cudnn.
ValueError: invalid setting 'require_index' true when 'pool_type' is 'avg'. ValueError: invalid setting 'require_index' true when 'pool_type' is 'avg'.
ValueError: 'pool_size' should be a list or tuple with length as 2. ValueError: 'pool_size' should be a list or tuple with length as 2.
...@@ -2540,11 +2541,11 @@ def adaptive_pool2d(input, ...@@ -2540,11 +2541,11 @@ def adaptive_pool2d(input,
data = fluid.layers.data( data = fluid.layers.data(
name='data', shape=[3, 32, 32], dtype='float32') name='data', shape=[3, 32, 32], dtype='float32')
conv2d = fluid.layers.pool2d( pool_out = fluid.layers.adaptive_pool2d(
input=data, input=data,
pool_size=[3, 3], pool_size=[3, 3],
pool_type='max', pool_type='max',
require_index=True) require_index=False)
""" """
if pool_type not in ["max", "avg"]: if pool_type not in ["max", "avg"]:
raise ValueError( raise ValueError(
...@@ -2565,6 +2566,9 @@ def adaptive_pool2d(input, ...@@ -2565,6 +2566,9 @@ def adaptive_pool2d(input,
if not isinstance(use_cudnn, bool): if not isinstance(use_cudnn, bool):
raise ValueError("use_cudnn should be True or False.") raise ValueError("use_cudnn should be True or False.")
if use_cudnn:
raise ValueError("adaptive pool currently not supported in cudnn.")
if pool_type == "max": if pool_type == "max":
l_type = 'max_pool2d_with_index' l_type = 'max_pool2d_with_index'
else: else:
...@@ -2590,7 +2594,7 @@ def adaptive_pool2d(input, ...@@ -2590,7 +2594,7 @@ def adaptive_pool2d(input,
"adaptive": True, "adaptive": True,
}) })
return pool_out return (pool_out, mask) if require_index else pool_out
@templatedoc(op_type="pool3d") @templatedoc(op_type="pool3d")
...@@ -2598,7 +2602,7 @@ def adaptive_pool3d(input, ...@@ -2598,7 +2602,7 @@ def adaptive_pool3d(input,
pool_size, pool_size,
pool_type="max", pool_type="max",
require_index=False, require_index=False,
use_cudnn=True, use_cudnn=False,
name=None): name=None):
""" """
${comment} ${comment}
...@@ -2613,7 +2617,7 @@ def adaptive_pool3d(input, ...@@ -2613,7 +2617,7 @@ def adaptive_pool3d(input,
pool_type: ${pooling_type_comment} pool_type: ${pooling_type_comment}
require_index (bool): If true, the index of max pooling point along with outputs. require_index (bool): If true, the index of max pooling point along with outputs.
it cannot be set in average pooling type. it cannot be set in average pooling type.
use_cudnn (bool): ${use_cudnn_comment} use_cudnn (bool, default False): adaptive pool currently not supported in cudnn.
name (str|None): A name for this layer(optional). If set None, the name (str|None): A name for this layer(optional). If set None, the
layer will be named automatically. layer will be named automatically.
...@@ -2623,6 +2627,7 @@ def adaptive_pool3d(input, ...@@ -2623,6 +2627,7 @@ def adaptive_pool3d(input,
Raises: Raises:
ValueError: 'pool_type' is not 'max' nor 'avg'. ValueError: 'pool_type' is not 'max' nor 'avg'.
ValueError: 'use_cudnn' is not a bool value. ValueError: 'use_cudnn' is not a bool value.
ValueError: adaptive pool currently not supported in cudnn.
ValueError: invalid setting 'require_index' true when 'pool_type' is 'avg'. ValueError: invalid setting 'require_index' true when 'pool_type' is 'avg'.
ValueError: 'pool_size' should be a list or tuple with length as 2. ValueError: 'pool_size' should be a list or tuple with length as 2.
...@@ -2632,7 +2637,7 @@ def adaptive_pool3d(input, ...@@ -2632,7 +2637,7 @@ def adaptive_pool3d(input,
data = fluid.layers.data( data = fluid.layers.data(
name='data', shape=[3, 32, 32], dtype='float32') name='data', shape=[3, 32, 32], dtype='float32')
conv2d = fluid.layers.pool2d( pool_out, mask = fluid.layers.adaptive_pool3d(
input=data, input=data,
pool_size=[3, 3], pool_size=[3, 3],
pool_type='max', pool_type='max',
...@@ -2657,6 +2662,9 @@ def adaptive_pool3d(input, ...@@ -2657,6 +2662,9 @@ def adaptive_pool3d(input,
if not isinstance(use_cudnn, bool): if not isinstance(use_cudnn, bool):
raise ValueError("use_cudnn should be True or False.") raise ValueError("use_cudnn should be True or False.")
if use_cudnn:
raise ValueError("adaptive pool currently not supported in cudnn.")
if pool_type == "max": if pool_type == "max":
l_type = 'max_pool3d_with_index' l_type = 'max_pool3d_with_index'
else: else:
...@@ -2682,7 +2690,7 @@ def adaptive_pool3d(input, ...@@ -2682,7 +2690,7 @@ def adaptive_pool3d(input,
"adaptive": True, "adaptive": True,
}) })
return pool_out return (pool_out, mask) if require_index else pool_out
def batch_norm(input, def batch_norm(input,
......
python/paddle/fluid/tests/unittests/test_layers.py
浏览文件 @ a81fabd3
...@@ -237,23 +237,24 @@ class TestBook(unittest.TestCase): ...@@ -237,23 +237,24 @@ class TestBook(unittest.TestCase):
program = Program() program = Program()
with program_guard(program): with program_guard(program):
x = layers.data(name='x', shape=[3, 224, 224], dtype='float32') x = layers.data(name='x', shape=[3, 224, 224], dtype='float32')
self.assertIsNotNone(
layers.adaptive_pool2d(
x, [3, 3], require_index=True))
self.assertIsNotNone( self.assertIsNotNone(
layers.adaptive_pool2d( layers.adaptive_pool2d(
x, [3, 3], pool_type='avg')) x, [3, 3], pool_type='avg'))
pool, mask = layers.adaptive_pool2d(x, [3, 3], require_index=True)
self.assertIsNotNone(pool)
self.assertIsNotNone(mask)
def test_adaptive_pool3d(self): def test_adaptive_pool3d(self):
program = Program() program = Program()
with program_guard(program): with program_guard(program):
x = layers.data(name='x', shape=[3, 244, 224, 224], dtype='float32') x = layers.data(name='x', shape=[3, 244, 224, 224], dtype='float32')
self.assertIsNotNone(
layers.adaptive_pool3d(
x, [3, 3, 3], require_index=True))
self.assertIsNotNone( self.assertIsNotNone(
layers.adaptive_pool3d( layers.adaptive_pool3d(
x, [3, 3, 3], pool_type='avg')) x, [3, 3, 3], pool_type='avg'))
pool, mask = layers.adaptive_pool3d(
x, [3, 3, 3], require_index=True)
self.assertIsNotNone(pool)
self.assertIsNotNone(mask)
def test_lstm_unit(self): def test_lstm_unit(self):
program = Program() program = Program()
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或