mindspore-Ecosystem
/
mindspore
21
0
Fork 27
Code Issues Pull Requests 6 Projects Releases Wiki Activity

!45035 Code Cleanings 1102 

Merge pull request !45035 from gaoshuanglong/clean_code_1031
This commit is contained in:
i-robot 2022-11-03 12:13:57 +00:00 committed by Gitee
parent 0d90568241 37074876d2
commit 1242055472
No known key found for this signature in database
GPG Key ID: 173E9B9CA92EEF8F
9 changed files with 35 additions and 29 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
mindspore/ccsrc/plugin/device/cpu/kernel/bincount_cpu_kernel.cc
View File

@ -36,9 +36,9 @@ void BincountCpuKernelMod::InitKernel(const CNodePtr &kernel_node) {
} }
template <typename T_in, typename T_out> template <typename T_in, typename T_out>
void BincountTask(const std::vector<AddressPtr> &inputs, const std::vector<AddressPtr> &workspaces, void BincountTask(const std::vector<AddressPtr> &inputs, const std::vector<AddressPtr> &,
const std::vector<AddressPtr> &outputs, const std::vector<int64_t> &input_arr_sizes, int32_t num_bins, const std::vector<AddressPtr> &outputs, const std::vector<int64_t> &input_arr_sizes, int32_t num_bins,
const std::vector<int64_t> &input_weights_sizes, const std::vector<int64_t> &output_sizes) { const std::vector<int64_t> &input_weights_sizes, const std::vector<int64_t> &) {
auto bin_array = static_cast<T_in *>(inputs[0]->addr); auto bin_array = static_cast<T_in *>(inputs[0]->addr);
auto output_data = static_cast<T_out *>(outputs[0]->addr); auto output_data = static_cast<T_out *>(outputs[0]->addr);
const size_t data_num = SizeOf(input_arr_sizes); const size_t data_num = SizeOf(input_arr_sizes);

6
mindspore/ccsrc/plugin/device/cpu/kernel/combined_non_max_suppression_cpu_kernel.cc
View File

@ -187,9 +187,9 @@ size_t CombinedNonMaxSuppressionCpuKernelMod::nms_perbath(float *boxes, float *s
int *valid_detection) { int *valid_detection) {
int box_size = num_bath_ * num_detection_ * sizeof(float) * multiplier; int box_size = num_bath_ * num_detection_ * sizeof(float) * multiplier;
int score_size = num_bath_ * num_detection_ * sizeof(float); int score_size = num_bath_ * num_detection_ * sizeof(float);
void(memset_s(nmsed_boxes, box_size, 0.0, box_size)); void(memset_s(nmsed_boxes, box_size, 0, box_size));
void(memset_s(nmsed_scores, score_size, 0.0, score_size)); void(memset_s(nmsed_scores, score_size, 0, score_size));
void(memset_s(nmsed_class, score_size, 0.0, score_size)); void(memset_s(nmsed_class, score_size, 0, score_size));
void(memset_s(valid_detection, sizeof(int) * num_bath_, 0, sizeof(int) * num_bath_)); void(memset_s(valid_detection, sizeof(int) * num_bath_, 0, sizeof(int) * num_bath_));
const float box_min = 0.0; const float box_min = 0.0;
const float box_max = 1.0; const float box_max = 1.0;

2
mindspore/ccsrc/plugin/device/cpu/kernel/eigen/eigen_common_utils.h
View File

@ -86,7 +86,7 @@ struct TTypes {
class EigenTensor { class EigenTensor {
public: public:
EigenTensor() = delete; EigenTensor() = delete;
EigenTensor(ShapeVector &shape, void *data_ptr) : tensor_shape(shape), tensor_data_ptr(data_ptr) {} EigenTensor(const ShapeVector &shape, void *data_ptr) : tensor_shape(shape), tensor_data_ptr(data_ptr) {}
EigenTensor(std::vector<size_t> &shape, void *data_ptr) : tensor_data_ptr(data_ptr) { EigenTensor(std::vector<size_t> &shape, void *data_ptr) : tensor_data_ptr(data_ptr) {
for (size_t dim : shape) { for (size_t dim : shape) {
(void)tensor_shape.emplace_back(static_cast<int64_t>(dim)); (void)tensor_shape.emplace_back(static_cast<int64_t>(dim));

18
mindspore/ccsrc/plugin/device/cpu/kernel/geqrf_cpu_kernel.cc
View File

@ -100,19 +100,19 @@ std::unique_ptr<T[]> GeqrfCpuKernelMod::Larf(size_t m, size_t n, T *x, T *tau, s
} }
template <typename T> template <typename T>
void GeqrfCpuKernelMod::Geqrf(size_t num_m, size_t num_n, T *x, T *tau) { void GeqrfCpuKernelMod::Geqrf(size_t num_m_, size_t num_n_, T *x, T *tau) {
if (num_m < 0 || num_n < 0) { if (num_m_ < 0 || num_n_ < 0) {
return; return;
} }
size_t k = std::min(num_m, num_n); size_t k = std::min(num_m_, num_n_);
T one = static_cast<T>(1); T one = static_cast<T>(1);
std::unique_ptr<T[]> workspace = std::make_unique<T[]>(num_n); std::unique_ptr<T[]> workspace = std::make_unique<T[]>(num_n_);
for (size_t i = 0; i < k; i++) { for (size_t i = 0; i < k; i++) {
Larfg<T>(num_m - i, i, i, x, tau + i); Larfg<T>(num_m_ - i, i, i, x, tau + i);
T aii = *(x + i * num_n + i); T aii = *(x + i * num_n_ + i);
*(x + i * num_n + i) = one; *(x + i * num_n_ + i) = one;
workspace = Larf<T>(num_m - i, num_n - i - 1, x, tau + i, std::move(workspace), i, i + 1); workspace = Larf<T>(num_m_ - i, num_n_ - i - 1, x, tau + i, std::move(workspace), i, i + 1);
*(x + i * num_n + i) = aii; *(x + i * num_n_ + i) = aii;
} }
} }

2
mindspore/ccsrc/plugin/device/cpu/kernel/matrix_exp_cpu_kernel.cc
View File

@ -215,7 +215,7 @@ bool MatrixExpCpuKernelMod::LaunchKernel(const std::vector<kernel::AddressPtr> &
return true; return true;
} }
void MatrixExpCpuKernelMod::TyepChangeForFp16(int64_t i, int64_t m, int64_t size_mm, mindspore::Float16 *input_x, void MatrixExpCpuKernelMod::TyepChangeForFp16(int64_t i, int64_t m, int64_t size_mm, const mindspore::Float16 *input_x,
mindspore::Float16 *output_y) const { mindspore::Float16 *output_y) const {
typedef Eigen::Matrix<float, Eigen::Dynamic, Eigen::Dynamic, Eigen::RowMajor> MatrixXd; typedef Eigen::Matrix<float, Eigen::Dynamic, Eigen::Dynamic, Eigen::RowMajor> MatrixXd;
MatrixXd I(m, m); MatrixXd I(m, m);

2
mindspore/ccsrc/plugin/device/cpu/kernel/matrix_exp_cpu_kernel.h
View File

@ -66,7 +66,7 @@ class MatrixExpCpuKernelMod : public NativeCpuKernelMod, public MatchKernelHelpe
bool LaunchKernel(const std::vector<kernel::AddressPtr> &inputs, const std::vector<kernel::AddressPtr> &workspace, bool LaunchKernel(const std::vector<kernel::AddressPtr> &inputs, const std::vector<kernel::AddressPtr> &workspace,
const std::vector<kernel::AddressPtr> &outputs); const std::vector<kernel::AddressPtr> &outputs);
void TyepChangeForFp16(int64_t i, int64_t m, int64_t size_mm, mindspore::Float16 *input_x, void TyepChangeForFp16(int64_t i, int64_t m, int64_t size_mm, const mindspore::Float16 *input_x,
mindspore::Float16 *output_y) const; mindspore::Float16 *output_y) const;
template <typename T> template <typename T>

2
mindspore/ccsrc/plugin/device/cpu/kernel/non_deterministic_ints_cpu_kernel.cc
View File

@ -59,7 +59,7 @@ bool NonDeterministicIntsCPUKernelMod::LaunchKernel(const std::vector<AddressPtr
size_t output_elem_num = outputs[0]->size / sizeof(T1); size_t output_elem_num = outputs[0]->size / sizeof(T1);
auto task = [output](size_t start, size_t end) { auto task = [output](size_t start, size_t end) {
auto max_data = std::numeric_limits<T1>::max(); auto max_data = std::numeric_limits<T1>::max();
std::default_random_engine seed(time(0)); std::default_random_engine seed(time(nullptr));
std::uniform_int_distribution<T1> u(-max_data, max_data); std::uniform_int_distribution<T1> u(-max_data, max_data);
for (size_t i = start; i < end; ++i) { for (size_t i = start; i < end; ++i) {
output[i] = u(seed); output[i] = u(seed);

13
mindspore/ccsrc/plugin/device/cpu/kernel/resize_bicubic_cpu_kernel.cc
View File

@ -122,10 +122,13 @@ class CachedInterpolationCalculator {
switch (new_indices_hand) { switch (new_indices_hand) {
case 0: case 0:
indexes_[0] = x_0; indexes_[0] = x_0;
break;
case 1: case 1:
indexes_[1] = x_1; indexes_[1] = x_1;
break;
case caseid2: case caseid2:
indexes_[kIndex2] = x_2; indexes_[kIndex2] = x_2;
break;
case caseid3: case caseid3:
indexes_[kIndex3] = x_3; indexes_[kIndex3] = x_3;
break; break;
@ -198,10 +201,10 @@ inline void GetWeightsAndIndices(const float scale, const int64_t out_loc, const
} }
} }
static void ComputeXWeightsAndIndices(const ResizerState &resizer_state, const bool half_pixel_centers, static void ComputeXWeightsAndIndices(const ResizerState &resizer_state, const bool half_pixel_centers_,
std::vector<WeightsAndIndices> *x_wais) { std::vector<WeightsAndIndices> *x_wais) {
CachedInterpolationCalculator calc; CachedInterpolationCalculator calc;
if (half_pixel_centers) { if (half_pixel_centers_) {
for (int64_t x = 0; x < resizer_state.out_width; ++x) { for (int64_t x = 0; x < resizer_state.out_width; ++x) {
GetWeightsAndIndices<HalfPixelScaler, true>(resizer_state.width_scale, x, resizer_state.in_width, GetWeightsAndIndices<HalfPixelScaler, true>(resizer_state.width_scale, x, resizer_state.in_width,
&(*x_wais)[static_cast<size_t>(x)]); &(*x_wais)[static_cast<size_t>(x)]);
@ -314,10 +317,10 @@ std::vector<float> CalSwitch(const WeightsAndIndices &x_wai, std::vector<float>
} }
template <typename T1, typename T2> template <typename T1, typename T2>
inline void interpolate_with_caching(const T1 *input_data, const ResizerState &RS, const bool half_pixel_centers, inline void interpolate_with_caching(const T1 *input_data, const ResizerState &RS, const bool half_pixel_centers_,
T2 output_data) { T2 output_data) {
std::vector<WeightsAndIndices> x_wais(RS.out_width); std::vector<WeightsAndIndices> x_wais(RS.out_width);
ComputeXWeightsAndIndices(RS, half_pixel_centers, &x_wais); ComputeXWeightsAndIndices(RS, half_pixel_centers_, &x_wais);
const int64_t in_row_width = RS.in_width * RS.channels; const int64_t in_row_width = RS.in_width * RS.channels;
const int64_t in_batch_width = RS.in_height * in_row_width; const int64_t in_batch_width = RS.in_height * in_row_width;
const T1 *input_b_ptr = input_data; const T1 *input_b_ptr = input_data;
@ -327,7 +330,7 @@ inline void interpolate_with_caching(const T1 *input_data, const ResizerState &R
for (int64_t b = 0; b < RS.batch_size; ++b, input_b_ptr += in_batch_width) { for (int64_t b = 0; b < RS.batch_size; ++b, input_b_ptr += in_batch_width) {
for (int64_t y = 0; y < RS.out_height; ++y, output_y_ptr += RS.out_width * RS.channels) { for (int64_t y = 0; y < RS.out_height; ++y, output_y_ptr += RS.out_width * RS.channels) {
WeightsAndIndices y_wai; WeightsAndIndices y_wai;
if (half_pixel_centers) { if (half_pixel_centers_) {
GetWeightsAndIndices<HalfPixelScaler, true>(RS.height_scale, y, RS.in_height, &y_wai); GetWeightsAndIndices<HalfPixelScaler, true>(RS.height_scale, y, RS.in_height, &y_wai);
} else { } else {
GetWeightsAndIndices<LegacyScaler, false>(RS.height_scale, y, RS.in_height, &y_wai); GetWeightsAndIndices<LegacyScaler, false>(RS.height_scale, y, RS.in_height, &y_wai);

15
mindspore/ccsrc/plugin/device/cpu/kernel/resize_bicubic_grad_cpu_kernel.cc
View File

@ -109,10 +109,13 @@ class CachedInterpolationCalculator {
switch (new_indices_hand) { switch (new_indices_hand) {
case 0: case 0:
indexes_[0] = x_0; indexes_[0] = x_0;
break;
case 1: case 1:
indexes_[1] = x_1; indexes_[1] = x_1;
break;
case caseid2: case caseid2:
indexes_[kIndex2] = x_2; indexes_[kIndex2] = x_2;
break;
case caseid3: case caseid3:
indexes_[kIndex3] = x_3; indexes_[kIndex3] = x_3;
break; break;
@ -186,10 +189,10 @@ inline void GetWeightsAndIndicesGrad(const float scale, const int64_t out_loc, c
} }
} }
static void ComputeGradientXWeightsAndIndices(const ResizerGradState &RGS, const bool half_pixel_centers, static void ComputeGradientXWeightsAndIndices(const ResizerGradState &RGS, const bool half_pixel_centers_,
std::vector<WeightsAndIndices> *x_wais) { std::vector<WeightsAndIndices> *x_wais) {
CachedInterpolationCalculator calc; CachedInterpolationCalculator calc;
if (half_pixel_centers) { if (half_pixel_centers_) {
for (int64_t x = 0; x < RGS.resized_width; ++x) { for (int64_t x = 0; x < RGS.resized_width; ++x) {
GetWeightsAndIndicesGrad<HalfPixelScalerGrad, true>(RGS.width_scale, x, RGS.original_width, GetWeightsAndIndicesGrad<HalfPixelScalerGrad, true>(RGS.width_scale, x, RGS.original_width,
&(*x_wais)[static_cast<size_t>(x)]); &(*x_wais)[static_cast<size_t>(x)]);
@ -284,10 +287,10 @@ void CalNonUtil(const ResizerGradState &RGS, const bool half_pixel_centers,
} }
template <typename T> template <typename T>
inline void ResizeBicubicGrad(const float *input_grad, const ResizerGradState &RGS, const bool half_pixel_centers, inline void ResizeBicubicGrad(const float *input_grad, const ResizerGradState &RGS, const bool half_pixel_centers_,
T *output_grad) { T *output_grad) {
std::vector<WeightsAndIndices> x_wais(RGS.resized_width); std::vector<WeightsAndIndices> x_wais(RGS.resized_width);
ComputeGradientXWeightsAndIndices(RGS, half_pixel_centers, &x_wais); ComputeGradientXWeightsAndIndices(RGS, half_pixel_centers_, &x_wais);
const bool flag = true; const bool flag = true;
bool utils_flag = false; bool utils_flag = false;
if (RGS.original_width * RGS.original_height * RGS.channels * RGS.batch_size >= kParallelDataNum) { if (RGS.original_width * RGS.original_height * RGS.channels * RGS.batch_size >= kParallelDataNum) {
@ -297,13 +300,13 @@ inline void ResizeBicubicGrad(const float *input_grad, const ResizerGradState &R
for (int64_t b = 0; b < RGS.batch_size; ++b) { for (int64_t b = 0; b < RGS.batch_size; ++b) {
auto task = [&](int64_t start, int64_t end) { auto task = [&](int64_t start, int64_t end) {
for (int64_t y = start; y < end; ++y) { for (int64_t y = start; y < end; ++y) {
ResizeCommomCalc(RGS, half_pixel_centers, x_wais, flag, input_grad, output_grad, b, y); ResizeCommomCalc(RGS, half_pixel_centers_, x_wais, flag, input_grad, output_grad, b, y);
} }
}; };
CPUKernelUtils::ParallelFor(task, static_cast<size_t>(RGS.resized_height)); CPUKernelUtils::ParallelFor(task, static_cast<size_t>(RGS.resized_height));
} }
} else { } else {
CalNonUtil(RGS, half_pixel_centers, x_wais, flag, input_grad, output_grad); CalNonUtil(RGS, half_pixel_centers_, x_wais, flag, input_grad, output_grad);
} }
} }