p19647583
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

!14681 update example 

From: @zhujingxuan
Reviewed-by: @wangchengyuan,@zhanghaibo5
Signed-off-by: @wangchengyuan
This commit is contained in:
mindspore-ci-bot 2021-04-07 15:31:12 +08:00 committed by Gitee
parent a3fc997c4e 2fcca3c357
commit 698a301927
16 changed files with 143 additions and 116 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
mindspore/lite/micro/example/mnist_x86/CMakeLists.txt
View File

@ -22,12 +22,10 @@ endif()
if(PLATFORM_ARM64) if(PLATFORM_ARM64)
add_compile_definitions(ENABLE_ARM64) add_compile_definitions(ENABLE_ARM64)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -march=armv8.2-a+dotprod")
endif() endif()
if(PLATFORM_ARM32) if(PLATFORM_ARM32)
add_compile_definitions(ENABLE_ARM32) add_compile_definitions(ENABLE_ARM32)
add_definitions(-mfloat-abi=softfp -mfpu=neon)
endif() endif()
set(CMAKE_C_FLAGS "${CMAKE_ENABLE_C99} ${CMAKE_C_FLAGS}") set(CMAKE_C_FLAGS "${CMAKE_ENABLE_C99} ${CMAKE_C_FLAGS}")

2
mindspore/lite/micro/example/mnist_x86/benchmark/benchmark.cc
View File

@ -111,7 +111,7 @@ int main(int argc, const char **argv) {
} }
lite::Context *context = nullptr; lite::Context *context = nullptr;
if (argc >= 5) { if (argc >= 6) {
// config benchmark context // config benchmark context
context = new (std::nothrow) lite::Context(); context = new (std::nothrow) lite::Context();
if (context == nullptr) { if (context == nullptr) {

0
mindspore/lite/micro/example/mnist_x86/benchmark/load_input.c Executable file → Normal file
View File

0
mindspore/lite/micro/example/mnist_x86/benchmark/load_input.h Executable file → Normal file
View File

BIN
mindspore/lite/micro/example/mnist_x86/mnist_input.bin
View File

Binary file not shown.

20
mindspore/lite/micro/example/mnist_x86/src/CMakeLists.txt
View File

@ -8,8 +8,9 @@ endif()
get_filename_component(PKG_PATH ${PKG_PATH} ABSOLUTE BASE_DIR ${CMAKE_CURRENT_BINARY_DIR}) get_filename_component(PKG_PATH ${PKG_PATH} ABSOLUTE BASE_DIR ${CMAKE_CURRENT_BINARY_DIR})
set(OP_LIB ${PKG_PATH}/tools/codegen/operator_library/lib/libops.a) set(OP_LIB ${PKG_PATH}/inference/lib/libmindspore-lite.a)
set(OP_HEADER_PATH ${PKG_PATH}/tools/codegen/operator_library/include) set(WRAPPER_LIB ${PKG_PATH}/tools/codegen/lib/libwrapper.a)
set(OP_HEADER_PATH ${PKG_PATH}/tools/codegen/include)
set(HEADER_PATH ${PKG_PATH}/inference) set(HEADER_PATH ${PKG_PATH}/inference)
message("operator lib path: ${OP_LIB}") message("operator lib path: ${OP_LIB}")
@ -20,6 +21,13 @@ add_compile_definitions(NOT_USE_STL)
include_directories(${OP_HEADER_PATH}) include_directories(${OP_HEADER_PATH})
include_directories(${HEADER_PATH}) include_directories(${HEADER_PATH})
if(NOT PLATFORM_ARM32 AND NOT PLATFORM_ARM64)
include_directories(${PKG_PATH}/tools/codegen/third_party/include)
include_directories(${PKG_PATH}/tools/codegen/third_party/include/CMSIS/Core/Include)
include_directories(${PKG_PATH}/tools/codegen/third_party/include/CMSIS/DSP/Include)
include_directories(${PKG_PATH}/tools/codegen/third_party/include/CMSIS/NN/Include)
endif()
include(net.cmake) include(net.cmake)
option(PLATFORM_ARM64 "build android arm64" OFF) option(PLATFORM_ARM64 "build android arm64" OFF)
@ -32,12 +40,10 @@ endif()
if(PLATFORM_ARM64) if(PLATFORM_ARM64)
add_compile_definitions(ENABLE_ARM64) add_compile_definitions(ENABLE_ARM64)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -march=armv8.2-a+dotprod")
endif() endif()
if(PLATFORM_ARM32) if(PLATFORM_ARM32)
add_compile_definitions(ENABLE_ARM32) add_compile_definitions(ENABLE_ARM32)
add_definitions(-mfloat-abi=softfp -mfpu=neon)
endif() endif()
set(CMAKE_C_FLAGS "${CMAKE_ENABLE_C99} ${CMAKE_C_FLAGS}") set(CMAKE_C_FLAGS "${CMAKE_ENABLE_C99} ${CMAKE_C_FLAGS}")
@ -64,11 +70,17 @@ function(create_library)
COMMAND rm -rf tmp COMMAND rm -rf tmp
COMMAND mkdir tmp COMMAND mkdir tmp
COMMAND cd tmp && ar -x ${OP_LIB} COMMAND cd tmp && ar -x ${OP_LIB}
COMMAND cd tmp && ar -x ${WRAPPER_LIB}
COMMAND echo "raw static library ${library_name} size:" COMMAND echo "raw static library ${library_name} size:"
COMMAND ls -lh ${library_name} COMMAND ls -lh ${library_name}
COMMAND mv ${library_name} ./tmp && cd tmp && ar -x ${library_name} COMMAND mv ${library_name} ./tmp && cd tmp && ar -x ${library_name}
COMMENT "unzip raw static library ${library_name}" COMMENT "unzip raw static library ${library_name}"
) )
if(NOT PLATFORM_ARM32 AND NOT PLATFORM_ARM64)
set(CMSIS_LIB ${PKG_PATH}/tools/codegen/third_party/lib/libcmsis_nn.a)
add_custom_command(TARGET net POST_BUILD COMMAND cd tmp && ar -x ${CMSIS_LIB})
endif()
foreach(object_file ${OP_SRC}) foreach(object_file ${OP_SRC})
add_custom_command(TARGET net POST_BUILD COMMAND mv ./tmp/${object_file} .) add_custom_command(TARGET net POST_BUILD COMMAND mv ./tmp/${object_file} .)
endforeach() endforeach()

7
mindspore/lite/micro/example/mnist_x86/src/model.h
View File

@ -1,6 +1,6 @@
/** /**
* Copyright 2020 Huawei Technologies Co., Ltd * Copyright 2021 Huawei Technologies Co., Ltd
* *
* Licensed under the Apache License, Version 2.0 (the "License"); * Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License. * you may not use this file except in compliance with the License.
@ -46,12 +46,11 @@ class MModel : public Model {
Model *Model::Import(const char *model_buf, size_t size) { Model *Model::Import(const char *model_buf, size_t size) {
MS_NULLPTR_IF_NULL(model_buf); MS_NULLPTR_IF_NULL(model_buf);
MModel *model = new (std::nothrow) MModel();
MS_NULLPTR_IF_NULL(model);
if (size == 0) { if (size == 0) {
delete model;
return nullptr; return nullptr;
} }
MModel *model = new (std::nothrow) MModel();
MS_NULLPTR_IF_NULL(model);
model->buf = reinterpret_cast<char *>(malloc(size)); model->buf = reinterpret_cast<char *>(malloc(size));
if (model->buf == nullptr) { if (model->buf == nullptr) {
delete model; delete model;

BIN
mindspore/lite/micro/example/mnist_x86/src/net.bin
View File

Binary file not shown.

159
mindspore/lite/micro/example/mnist_x86/src/net.c
View File

@ -37,12 +37,12 @@ int CopyOutputsData(void **outputs, int num) {
if (num != 1) { if (num != 1) {
return RET_ERROR; return RET_ERROR;
} }
memcpy(outputs[0], g_Buffer+56, 40); memcpy(outputs[0], g_Buffer + 32, 40);
return RET_OK; return RET_OK;
} }
int GetBufferSize() { int GetBufferSize() {
return 40032; return 39248;
} }
int SetBuffer( void *buffer) { int SetBuffer( void *buffer) {
if (buffer == NULL) { if (buffer == NULL) {
@ -62,108 +62,107 @@ void FreeResource() {
} }
void Inference() { void Inference() {
{ {
DoQuantizeFp32ToInt8((float *)(g_Input0), (int8_t *)(g_Buffer+0), 0.007874015718698501587, 0, 784, false); memset((int16_t *)(g_Buffer + 10144), 0, 2048);
} memset((int16_t *)(g_Buffer + 12192), 0, 256);
{ memset((int *)(g_Buffer + 12448), 0, 6144);
memset((int16_t *)(g_Buffer+10928), 0, 2048); memset((int8_t *)(g_Buffer + 18592), 0, 8112);
memset((int16_t *)(g_Buffer+12976), 0, 256); memset((int16_t *)(g_Buffer + 26704), 0, 12544);
memset((int *)(g_Buffer+13232), 0, 6144); QuantArg conv_param__quant_arg_in[1] = {{0.003921568859368562698, -128}};
memset((int8_t *)(g_Buffer+19376), 0, 8112); QuantArg conv_param__quant_arg_w[12] = {{0.005689438898116350174, 0}, {0.006241692230105400085, 0}, {0.007301395758986473083, 0}, {0.005148916970938444138, 0}, {0.005132303573191165924, 0}, {0.004976313561201095581, 0}, {0.00564815988764166832, 0}, {0.002269793068990111351, 0}, {0.0030086529441177845, 0}, {0.005234404932707548141, 0}, {0.007580270525068044662, 0}, {0.004589735530316829681, 0}};
memset((int16_t *)(g_Buffer+27488), 0, 12544); QuantArg conv_param__quant_arg_out[1] = {{0.01811622083187103271, 17}};
QuantArg conv_param__quant_arg_in[1] = {{0.007874015718698501587, 0}}; double conv_param__real_multiplier[12] = {0.001231577267748737653, 0.001351122051282624588, 0.00158051323770531417, 0.001114571969708069233, 0.001110975704014940469, 0.001077209041359399825, 0.001222641776980984765, 0.0004913359221160916793, 0.0006512749113606706042, 0.001133077320583530554, 0.001640880438584302065, 0.0009935275121536731122};
QuantArg conv_param__quant_arg_w[12] = {{0.003238174133002758026, -6}, {0.003890725085511803627, -8}, {0.003394871251657605171, -7}, {0.001685356837697327137, -127}, {0.004322394262999296188, 1}, {0.002274985425174236298, -56}, {0.003617759561166167259, 17}, {0.004447745624929666519, 23}, {0.004683905746787786484, 26}, {0.004021023400127887726, 24}, {0.005650237202644348145, 11}, {0.001966834301128983498, -84}};
QuantArg conv_param__quant_arg_out[1] = {{0.01778890006244182587, 0}};
double conv_param__real_multiplier[12] = {0.001433333970799530351, 0.001722176774828924938, 0.00150269379968211614, 0.0007460003866156953226, 0.001913249346122961134, 0.001006991503636309139, 0.001601352314486244018, 0.001968734305210294733, 0.002073267527210802957, 0.00177985160945266568, 0.002501001060249878095, 0.0008705926067589928779};
int conv_param__left_shift[12] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}; int conv_param__left_shift[12] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
int conv_param__right_shift[12] = {-9, -9, -9, -10, -9, -9, -9, -8, -8, -9, -8, -10}; int conv_param__right_shift[12] = {-9, -9, -9, -9, -9, -9, -9, -10, -10, -9, -9, -9};
int conv_param__quant_multiplier[12] = {1575967367, 1893553389, 1652229306, 1640472199, 2103639903, 1107198867, 1760705490, 1082323130, 1139790877, 1956967540, 1374939873, 1914453388}; int conv_param__quant_multiplier[12] = {1354133526, 1485574406, 1737792683, 1225484841, 1221530705, 1184403867, 1344308850, 1080459119, 1432168676, 1245831689, 1804167122, 1092395052};
int conv_param__out_act_min[1] = {0}; int conv_param__out_act_min[1] = {-128};
int conv_param__out_act_max[1] = {127}; int conv_param__out_act_max[1] = {127};
ConvQuantArg conv_param__conv_quant_arg = {(RoundingMode)(1), 2, conv_param__quant_arg_in, conv_param__quant_arg_w, conv_param__quant_arg_out, conv_param__real_multiplier, conv_param__left_shift, conv_param__right_shift, conv_param__quant_multiplier, conv_param__out_act_min, conv_param__out_act_max, 1, 12, 1, 2}; ConvQuantArg conv_param__conv_quant_arg = {(RoundingMode)(2), 2, conv_param__quant_arg_in, conv_param__quant_arg_w, conv_param__quant_arg_out, conv_param__real_multiplier, conv_param__left_shift, conv_param__right_shift, conv_param__quant_multiplier, conv_param__out_act_min, conv_param__out_act_max, 1, 12, 1, 2};
int thread_num = MSMIN(g_thread_num, 26); int thread_num = MSMIN(g_thread_num, 26);
ConvParameter conv_param_ = {{ "", 35, g_thread_num}, conv_param__conv_quant_arg, 3, 3, 1, 1, 1, 1, 0, 0, 0, 0, 1, 0, 1, 28, 28, 1, 1, 26, 26, 12, thread_num, 0, 0, (PadMode)(2), (ActType)(1), 0, 0, 0}; ConvParameter conv_param_ = {{ "", true, 35, g_thread_num, 0}, conv_param__conv_quant_arg, 3, 3, 1, 1, 1, 1, 0, 0, 0, 0, 1, 0, 1, 28, 28, 1, 1, 26, 26, 12, thread_num, 0, 0, (PadMode)(2), (ActType)(0), 0, 0, 0};
PackInputToC8Int8((int8_t *)(g_Buffer+0), (int16_t *)(g_Buffer+27488), &conv_param_); PackInputToC8Int8((int8_t *)(g_Input0), (int16_t *)(g_Buffer + 26704), &conv_param_);
Conv3x3Int8((int16_t *)(g_Buffer+27488), g_Weight10, g_Weight11, (int8_t *)(g_Buffer+784), (int16_t *)(g_Buffer+10928), (int16_t *)(g_Buffer+12976), (int *)(g_Buffer+13232), (int8_t *)(g_Buffer+19376), 0, &conv_param_); Conv3x3Int8((int16_t *)(g_Buffer + 26704), g_Weight10, g_Weight11, (int8_t *)(g_Buffer + 0), (int16_t *)(g_Buffer + 10144), (int16_t *)(g_Buffer + 12192), (int *)(g_Buffer + 12448), (int8_t *)(g_Buffer + 18592), 0, &conv_param_);
PackNC4HW4ToNHWCInt8((int8_t *)(g_Buffer+19376), (int8_t *)(g_Buffer+784), 1, 676, 12); PackNC4HW4ToNHWCInt8((int8_t *)(g_Buffer + 18592), (int8_t *)(g_Buffer + 0), 1, 676, 12);
} }
{ {
static QuantArg pooling_parameter_quant_in = {0.01778890006244182587, 0}; static QuantArg pooling_parameter_quant_in = {0.01811622083187103271, 17};
static QuantArg pooling_parameter_quant_out = {0.01778890006244182587, 0}; static QuantArg pooling_parameter_quant_out = {0.01811622083187103271, 17};
static QuantArg *pooling_parameter_quant[2] = { &pooling_parameter_quant_in, &pooling_parameter_quant_out}; static QuantArg *pooling_parameter_quant[2] = { &pooling_parameter_quant_in, &pooling_parameter_quant_out};
const PoolingParameter pooling_parameter = {{ "", 92, g_thread_num}, (PoolMode)(1), (RoundMode)(2), (PadMode)(2), (ActType)(0), 0, false, 2, 2, 2, 2, 26, 26, 1, 12, 13, 13, 1, 12, 0, 0, 0, 0, 0, pooling_parameter_quant, false}; const PoolingParameter pooling_parameter = {{ "", true, 92, g_thread_num, 0}, (PoolMode)(1), (RoundMode)(2), (PadMode)(2), (ActType)(0), 0, false, 2, 2, 2, 2, 26, 26, 1, 12, 13, 13, 1, 12, 0, 0, 0, 0, 0, pooling_parameter_quant, false};
MaxPoolingInt8((int8_t *)(g_Buffer+784), (int8_t *)(g_Buffer+8896), (PoolingParameter *)&pooling_parameter, 0); MaxPoolingInt8((int8_t *)(g_Buffer + 0), (int8_t *)(g_Buffer + 8112), (PoolingParameter *)&pooling_parameter, 0);
} }
{ {
memset((int16_t *)(g_Buffer+10928), 0, 4096); memset((int16_t *)(g_Buffer + 10144), 0, 4096);
memset((int16_t *)(g_Buffer+15024), 0, 256); memset((int16_t *)(g_Buffer + 14240), 0, 256);
memset((int *)(g_Buffer+15280), 0, 6144); memset((int *)(g_Buffer + 14496), 0, 6144);
memset((int8_t *)(g_Buffer+21424), 0, 1452); memset((int8_t *)(g_Buffer + 20640), 0, 1452);
memset((int16_t *)(g_Buffer+22876), 0, 5408); memset((int16_t *)(g_Buffer + 22092), 0, 5408);
QuantArg conv_param__quant_arg_in[1] = {{0.01778890006244182587, 0}}; QuantArg conv_param__quant_arg_in[1] = {{0.01811622083187103271, 17}};
QuantArg conv_param__quant_arg_w[12] = {{0.005374609492719173431, 33}, {0.005837683100253343582, 22}, {0.004709810949862003326, -15}, {0.003726204857230186462, 27}, {0.00318551529198884964, -8}, {0.003453079145401716232, 50}, {0.004045850131660699844, -9}, {0.003903790842741727829, 30}, {0.004003710579127073288, -10}, {0.00560879148542881012, 27}, {0.005486610345542430878, -23}, {0.003554018214344978333, 4}}; QuantArg conv_param__quant_arg_w[12] = {{0.006381968967616558075, 0}, {0.005092236679047346115, 0}, {0.004954888485372066498, 0}, {0.007594361435621976852, 0}, {0.006317862775176763535, 0}, {0.004739056341350078583, 0}, {0.004733041394501924515, 0}, {0.005125139374285936356, 0}, {0.005773660261183977127, 0}, {0.007067613303661346436, 0}, {0.00728381425142288208, 0}, {0.004714466165751218796, 0}};
QuantArg conv_param__quant_arg_out[1] = {{0.07183934003114700317, 0}}; QuantArg conv_param__quant_arg_out[1] = {{0.118615470826625824, 31}};
double conv_param__real_multiplier[12] = {0.001330863973520378732, 0.001445530533608141606, 0.001166246148374064893, 0.0009226850783705293785, 0.0007887991893445710223, 0.0008550534992628172192, 0.001001835847923064193, 0.0009666590447744700769, 0.0009914011740411567478, 0.001388852288199173826, 0.00135859773990280961, 0.0008800481219728497088}; double conv_param__real_multiplier[12] = {0.0009747224012760375951, 0.0007777407468524931162, 0.0007567634496453238277, 0.001159891919861241348, 0.0009649314419479496259, 0.0007237992569070154231, 0.0007228806183814449719, 0.0007827659621256170689, 0.0008818150205007141765, 0.001079441365823280083, 0.001112461807995879974, 0.0007200436103814696152};
int conv_param__left_shift[12] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}; int conv_param__left_shift[12] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
int conv_param__right_shift[12] = {-9, -9, -9, -10, -10, -10, -9, -10, -9, -9, -9, -10}; int conv_param__right_shift[12] = {-10, -10, -10, -9, -10, -10, -10, -10, -10, -9, -9, -10};
int conv_param__quant_multiplier[12] = {1463300414, 1589377630, 1282301201, 2029005945, 1734587761, 1880282530, 1101530164, 2125705720, 1090057119, 1527059240, 1493794012, 1935246286}; int conv_param__quant_multiplier[12] = {2143437228, 1710269989, 1664140425, 1275314653, 2121906681, 1591651398, 1589631291, 1721320554, 1939131737, 1186858333, 1223164693, 1583392644};
int conv_param__out_act_min[1] = {0}; int conv_param__out_act_min[1] = {-128};
int conv_param__out_act_max[1] = {127}; int conv_param__out_act_max[1] = {127};
ConvQuantArg conv_param__conv_quant_arg = {(RoundingMode)(1), 2, conv_param__quant_arg_in, conv_param__quant_arg_w, conv_param__quant_arg_out, conv_param__real_multiplier, conv_param__left_shift, conv_param__right_shift, conv_param__quant_multiplier, conv_param__out_act_min, conv_param__out_act_max, 1, 12, 1, 2}; ConvQuantArg conv_param__conv_quant_arg = {(RoundingMode)(1), 2, conv_param__quant_arg_in, conv_param__quant_arg_w, conv_param__quant_arg_out, conv_param__real_multiplier, conv_param__left_shift, conv_param__right_shift, conv_param__quant_multiplier, conv_param__out_act_min, conv_param__out_act_max, 1, 12, 1, 2};
int thread_num = MSMIN(g_thread_num, 11); int thread_num = MSMIN(g_thread_num, 11);
ConvParameter conv_param_ = {{ "", 35, g_thread_num}, conv_param__conv_quant_arg, 3, 3, 1, 1, 1, 1, 0, 0, 0, 0, 1, 0, 1, 13, 13, 12, 1, 11, 11, 12, thread_num, 0, 0, (PadMode)(2), (ActType)(1), 0, 0, 0}; ConvParameter conv_param_ = {{ "", true, 35, g_thread_num, 0}, conv_param__conv_quant_arg, 3, 3, 1, 1, 1, 1, 0, 0, 0, 0, 1, 0, 1, 13, 13, 12, 1, 11, 11, 12, thread_num, 0, 0, (PadMode)(2), (ActType)(0), 0, 0, 0};
PackInputToC8Int8((int8_t *)(g_Buffer+8896), (int16_t *)(g_Buffer+22876), &conv_param_); PackInputToC8Int8((int8_t *)(g_Buffer + 8112), (int16_t *)(g_Buffer + 22092), &conv_param_);
Conv3x3Int8((int16_t *)(g_Buffer+22876), g_Weight12, g_Weight13, (int8_t *)(g_Buffer+0), (int16_t *)(g_Buffer+10928), (int16_t *)(g_Buffer+15024), (int *)(g_Buffer+15280), (int8_t *)(g_Buffer+21424), 0, &conv_param_); Conv3x3Int8((int16_t *)(g_Buffer + 22092), g_Weight12, g_Weight13, (int8_t *)(g_Buffer + 0), (int16_t *)(g_Buffer + 10144), (int16_t *)(g_Buffer + 14240), (int *)(g_Buffer + 14496), (int8_t *)(g_Buffer + 20640), 0, &conv_param_);
PackNC4HW4ToNHWCInt8((int8_t *)(g_Buffer+21424), (int8_t *)(g_Buffer+0), 1, 121, 12); PackNC4HW4ToNHWCInt8((int8_t *)(g_Buffer + 20640), (int8_t *)(g_Buffer + 0), 1, 121, 12);
} }
{ {
static QuantArg pooling_parameter_quant_in = {0.07136065512895584106, 0}; static QuantArg pooling_parameter_quant_in = {0.118615470826625824, 31};
static QuantArg pooling_parameter_quant_out = {0.07136065512895584106, 0}; static QuantArg pooling_parameter_quant_out = {0.118615470826625824, 31};
static QuantArg *pooling_parameter_quant[2] = { &pooling_parameter_quant_in, &pooling_parameter_quant_out}; static QuantArg *pooling_parameter_quant[2] = { &pooling_parameter_quant_in, &pooling_parameter_quant_out};
const PoolingParameter pooling_parameter = {{ "", 92, g_thread_num}, (PoolMode)(1), (RoundMode)(2), (PadMode)(2), (ActType)(0), 0, false, 2, 2, 2, 2, 11, 11, 1, 12, 5, 5, 1, 12, 0, 0, 0, 0, 0, pooling_parameter_quant, false}; const PoolingParameter pooling_parameter = {{ "", true, 92, g_thread_num, 0}, (PoolMode)(1), (RoundMode)(2), (PadMode)(2), (ActType)(0), 0, false, 2, 2, 2, 2, 11, 11, 1, 12, 5, 5, 1, 12, 0, 0, 0, 0, 0, pooling_parameter_quant, false};
MaxPoolingInt8((int8_t *)(g_Buffer+0), (int8_t *)(g_Buffer+1456), (PoolingParameter *)&pooling_parameter, 0); MaxPoolingInt8((int8_t *)(g_Buffer + 0), (int8_t *)(g_Buffer + 1456), (PoolingParameter *)&pooling_parameter, 0);
} }
{ {
const ReshapeQuantArg reshape_quant_arg = {{0.07136065512895584106, 0}, {0.07136065512895584106, 0}, -128, 127}; const ReshapeQuantArg reshape_quant_arg = {{0.118615470826625824, 31}, {0.118615470826625824, 31}, -128, 127};
Int8Reshape((int8_t *)(g_Buffer+1456), (int8_t *)(g_Buffer+0), 300, reshape_quant_arg); Int8Reshape((int8_t *)(g_Buffer + 1456), (int8_t *)(g_Buffer + 0), 300, reshape_quant_arg);
} }
{ {
int32_t tmp_weight_zp = 1; int32_t tmp_weight_zp = 0;
RowMajor2Row16x4MajorInt8((int8_t *)(g_Buffer+0)+0, (int8_t *)(g_Buffer+10928), 1, 300); RowMajor2Row16x4MajorInt8((int8_t *)(g_Buffer + 0)+0, (int8_t *)(g_Buffer + 10144), 1, 300);
CalcInputSums((int8_t *)(g_Buffer+0)+0, 1, 300, tmp_weight_zp, (int *)(g_Buffer+12144), RowMajor); CalcInputSums((int8_t *)(g_Buffer + 0)+0, 1, 300, tmp_weight_zp, (int *)(g_Buffer + 11360), RowMajor);
float filter_scale[20] = {0.003479549195617437363, 0.004490676335990428925, 0.004529818892478942871, 0.002983231563121080399, 0.003455155529081821442, 0.003223794745281338692, 0.003272445406764745712, 0.003801185870543122292, 0.003679843153804540634, 0.003040234791114926338, 0.003704284550622105598, 0.003355232765898108482, 0.002904496388509869576, 0.003024494973942637444, 0.002794801956042647362, 0.004355110693722963333, 0.003499472280964255333, 0.004184196703135967255, 0.003057289868593215942, 0.003264668164774775505}; float filter_scale[1] = {0.007667620200663805008};
int filter_zp[20] = {1, 12, 3, 2, -10, -5, -11, 5, 12, 22, 16, 1, -5, 15, 13, 5, -10, -5, -6, 0}; int filter_zp[1] = {0};
int left_shift[20] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}; int left_shift[1] = {0};
int right_shift[20] = {-10, -9, -9, -10, -10, -10, -10, -9, -9, -10, -9, -10, -10, -10, -10, -9, -10, -9, -10, -10}; int right_shift[1] = {-8};
int multiplier[20] = {2108215049, 1360422072, 1372280070, 1807502393, 2093435146, 1953256619, 1982733521, 1151545365, 1114785262, 1842040025, 1122189669, 2032893316, 1759797843, 1832503464, 1693335354, 1319353429, 2120286176, 1267576078, 1852373503, 1978021333}; int multiplier[1] = {1379728867};
const MatmulQuantParameter matmul_quant_parameter = {{0.07136065512895584106, 0}, {0, 0}, {0.258998185396194458, 0}, -128, 127, filter_scale, filter_zp, left_shift, right_shift, multiplier}; const MatmulQuantParameter matmul_quant_parameter = {{0.118615470826625824, 31}, {0, 0}, {0.3623915016651153564, 11}, -128, 127, filter_scale, filter_zp, left_shift, right_shift, multiplier};
int32_t *cur_left = matmul_quant_parameter.left_shift_ + 0; int32_t *cur_left = matmul_quant_parameter.left_shift_;
int32_t *cur_right = matmul_quant_parameter.right_shift_ + 0; int32_t *cur_right = matmul_quant_parameter.right_shift_;
int32_t *cur_mul = matmul_quant_parameter.quant_multiplier_ + 0; int32_t *cur_mul = matmul_quant_parameter.quant_multiplier_ ;
int32_t *cur_zp = matmul_quant_parameter.filter_zp_ + 0; int32_t *cur_zp = matmul_quant_parameter.filter_zp_ ;
MatmulInt8Opt((int8_t *)(g_Buffer+10928), g_Weight15+0 + 0, (int8_t *)(g_Buffer+304)+0+0, 1, 20, 304, (int *)(g_Buffer+12144), g_Weight16+0, -128, 127, 0, cur_mul, cur_left, cur_right, 20, true, cur_zp); MatmulInt8Opt((int8_t *)(g_Buffer + 10144), g_Weight15+0 + 0, (int8_t *)(g_Buffer + 304)+0+0, 1, 20, 304, (int *)(g_Buffer + 11360), g_Weight16+0, -128, 127, 11, cur_mul, cur_left, cur_right, 20, false, cur_zp);
} }
{ {
int32_t tmp_weight_zp = 1; int32_t tmp_weight_zp = 0;
RowMajor2Row16x4MajorInt8((int8_t *)(g_Buffer+304)+0, (int8_t *)(g_Buffer+10928), 1, 20); RowMajor2Row16x4MajorInt8((int8_t *)(g_Buffer + 304)+0, (int8_t *)(g_Buffer + 10144), 1, 20);
CalcInputSums((int8_t *)(g_Buffer+304)+0, 1, 20, tmp_weight_zp, (int *)(g_Buffer+11056), RowMajor); CalcInputSums((int8_t *)(g_Buffer + 304)+0, 1, 20, tmp_weight_zp, (int *)(g_Buffer + 10272), RowMajor);
float filter_scale[10] = {0.004678330849856138229, 0.005127115640789270401, 0.00471437256783246994, 0.004531511571258306503, 0.005476122256368398666, 0.004348111804574728012, 0.004803542047739028931, 0.006081215571612119675, 0.004532597027719020844, 0.004762654658406972885}; float filter_scale[1] = {0.006908571347594261169};
int filter_zp[10] = {7, -2, 9, 2, -6, 21, 16, 10, -19, 8}; int filter_zp[1] = {0};
int left_shift[10] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0}; int left_shift[1] = {0};
int right_shift[10] = {-8, -8, -8, -8, -8, -8, -8, -8, -8, -8}; int right_shift[1] = {-8};
int multiplier[10] = {1242805482, 1362025788, 1252380041, 1203802750, 1454739904, 1155082292, 1276068015, 1615483838, 1204091115, 1265206260}; int multiplier[1] = {1282256865};
const MatmulQuantParameter matmul_quant_parameter = {{0.258998185396194458, 0}, {0, 0}, {0.5359870791435241699, 0}, -128, 127, filter_scale, filter_zp, left_shift, right_shift, multiplier}; const MatmulQuantParameter matmul_quant_parameter = {{0.3623915016651153564, 11}, {0, 0}, {1.073398709297180176, -20}, -128, 127, filter_scale, filter_zp, left_shift, right_shift, multiplier};
int32_t *cur_left = matmul_quant_parameter.left_shift_ + 0; int32_t *cur_left = matmul_quant_parameter.left_shift_;
int32_t *cur_right = matmul_quant_parameter.right_shift_ + 0; int32_t *cur_right = matmul_quant_parameter.right_shift_;
int32_t *cur_mul = matmul_quant_parameter.quant_multiplier_ + 0; int32_t *cur_mul = matmul_quant_parameter.quant_multiplier_ ;
int32_t *cur_zp = matmul_quant_parameter.filter_zp_ + 0; int32_t *cur_zp = matmul_quant_parameter.filter_zp_ ;
MatmulInt8Opt((int8_t *)(g_Buffer+10928), g_Weight18+0 + 0, (int8_t *)(g_Buffer+0)+0+0, 1, 10, 32, (int *)(g_Buffer+11056), g_Weight19+0, -128, 127, 0, cur_mul, cur_left, cur_right, 10, true, cur_zp); MatmulInt8Opt((int8_t *)(g_Buffer + 10144), g_Weight18+0 + 0, (int8_t *)(g_Buffer + 0)+0+0, 1, 10, 32, (int *)(g_Buffer + 10272), g_Weight19+0, -128, 127, -20, cur_mul, cur_left, cur_right, 10, false, cur_zp);
} }
{ {
DoDequantizeInt8ToFp32((int8_t *)(g_Buffer+0), (float *)(g_Buffer+16), 0.5359870791435241699, 0, 10); const SoftmaxQuantArg quant_args = {{1.073398709297180176, 20}, {0.00390625, -128}, -128, 127, 1152553088, 27, 27};
const SoftmaxParameter softmax_parameter = {{ "", true, 138, g_thread_num, 0}, 1, {1, 10}, 10, 2};
memset((int *)(g_Buffer + 10144), 0, 40);
memset((int *)(g_Buffer + 10184), 0, 40);
SoftmaxInt8((int8_t *)(g_Buffer + 0), (int8_t *)(g_Buffer + 16), 1, (int *)(g_Buffer + 10144), (int *)(g_Buffer + 10184), quant_args, (SoftmaxParameter *)&softmax_parameter);
} }
{ {
const SoftmaxParameter softmax_parameter = {{ "", 138, g_thread_num}, 1, {1, 10}, 10, 2}; DoDequantizeInt8ToFp32((int8_t *)(g_Buffer + 16), (float *)(g_Buffer + 32), 0.00390625, -128, 10);
memset((float *)(g_Buffer+10928), 0, 4);
Softmax((float *)(g_Buffer+16), (float *)(g_Buffer+56), (float *)(g_Buffer+10928), &softmax_parameter);
} }
} }

3
mindspore/lite/micro/example/mnist_x86/src/net.cmake
View File

@ -4,7 +4,6 @@ set(OP_SRC
common_func_int8.c.o common_func_int8.c.o
conv3x3_int8.c.o conv3x3_int8.c.o
conv_int8.c.o conv_int8.c.o
exp_fp32.c.o
fixed_point.c.o fixed_point.c.o
matmul_int8.c.o matmul_int8.c.o
matmul_int8_wrapper.c.o matmul_int8_wrapper.c.o
@ -12,7 +11,7 @@ set(OP_SRC
pooling_int8.c.o pooling_int8.c.o
quant_dtype_cast_int8.c.o quant_dtype_cast_int8.c.o
reshape_int8.c.o reshape_int8.c.o
softmax_fp32.c.o softmax_int8.c.o
weight.c.o weight.c.o
net.c.o net.c.o
session.cc.o session.cc.o

5
mindspore/lite/micro/example/mnist_x86/src/session.cc
View File

@ -30,14 +30,14 @@ int LiteSession::CompileGraph(lite::Model *model) {
in_shape_0[1] = 28; in_shape_0[1] = 28;
in_shape_0[2] = 28; in_shape_0[2] = 28;
in_shape_0[3] = 1; in_shape_0[3] = 1;
inputs_[0] = new (std::nothrow) MTensor(String("graph_input-0"), kNumberTypeFloat32, in_shape_0); inputs_[0] = new (std::nothrow) MTensor(String("graph_input-0"), kNumberTypeInt8, in_shape_0);
MS_ERROR_IF_NULL(inputs_[0]); MS_ERROR_IF_NULL(inputs_[0]);
outputs_.resize(1); outputs_.resize(1);
Vector<int32_t> out_shape_0; Vector<int32_t> out_shape_0;
out_shape_0.resize(2); out_shape_0.resize(2);
out_shape_0[0] = 1; out_shape_0[0] = 1;
out_shape_0[1] = 10; out_shape_0[1] = 10;
outputs_[0] = new (std::nothrow) MTensor(String("Softmax-7"), kNumberTypeFloat32, out_shape_0); outputs_[0] = new (std::nothrow) MTensor(String("int8toft32_Softmax-7_post0/output-0"), kNumberTypeFloat32, out_shape_0);
MS_ERROR_IF_NULL(outputs_[0]); MS_ERROR_IF_NULL(outputs_[0]);
int ret = Init(model->buf, static_cast<MModel *>(model)->buf_size()); int ret = Init(model->buf, static_cast<MModel *>(model)->buf_size());
return ret; return ret;
@ -126,7 +126,6 @@ mindspore::tensor::MSTensor *LiteSession::GetOutputByTensorName(const String &te
} }
return nullptr; return nullptr;
} }
} // namespace lite } // namespace lite
session::LiteSession *session::LiteSession::CreateSession(const lite::Context *context) { session::LiteSession *session::LiteSession::CreateSession(const lite::Context *context) {
auto *session = new (std::nothrow) lite::LiteSession(); auto *session = new (std::nothrow) lite::LiteSession();

1
mindspore/lite/micro/example/mnist_x86/src/session.h
View File

@ -78,6 +78,7 @@ class LiteSession : public session::LiteSession {
Vector<MTensor *> outputs_; Vector<MTensor *> outputs_;
void *runtime_buffer_; void *runtime_buffer_;
}; };
} // namespace lite } // namespace lite
} // namespace mindspore } // namespace mindspore

1
mindspore/lite/micro/example/mnist_x86/src/tensor.h
View File

@ -59,7 +59,6 @@ class MTensor : public mindspore::tensor::MSTensor {
void *data_ = nullptr; void *data_ = nullptr;
Vector<QuantArg> quant_params_; Vector<QuantArg> quant_params_;
}; };
} // namespace lite } // namespace lite
} // namespace mindspore } // namespace mindspore

22
mindspore/lite/micro/example/mnist_x86/src/weight.c
View File

@ -29,6 +29,10 @@ int32_t *g_Weight16 = NULL;
int32_t *g_Weight17 = NULL; int32_t *g_Weight17 = NULL;
int8_t *g_Weight18 = NULL; int8_t *g_Weight18 = NULL;
int32_t *g_Weight19 = NULL; int32_t *g_Weight19 = NULL;
int8_t g_Weight6[6000];
int32_t g_Weight7[20];
int8_t g_Weight8[200];
int32_t g_Weight9[10];
int Init(void *weight_buffer, int weight_size) { int Init(void *weight_buffer, int weight_size) {
if (weight_buffer == NULL) { if (weight_buffer == NULL) {
@ -39,19 +43,19 @@ int Init(void *weight_buffer, int weight_size) {
size_t size; size_t size;
size_t offset; size_t offset;
}; };
int8_t *g_Weight6 = (weight_buffer + 9312);
int32_t *g_Weight7 = (weight_buffer + 15312);
int8_t *g_Weight8 = (weight_buffer + 15392);
int32_t *g_Weight9 = (weight_buffer + 15592);
struct ModelParameter model_params[] = { struct ModelParameter model_params[] = {
{g_Weight10, 3072, 0}, {g_Weight10, 3072, 0},
{g_Weight11, 48, 3072}, {g_Weight11, 48, 3072},
{g_Weight12, 6144, 3120}, {g_Weight12, 6144, 3120},
{g_Weight13, 48, 9264}, {g_Weight13, 48, 9264},
{g_Weight6, 6000, 9312},
{g_Weight7, 80, 15312},
{g_Weight8, 200, 15392},
{g_Weight9, 40, 15592},
}; };
for(int i = 0; i < 4; ++i) { for(int i = 0; i < 8; ++i) {
if (model_params[i].offset + model_params[i].size > weight_size) { if (model_params[i].offset + model_params[i].size > weight_size) {
return RET_ERROR; return RET_ERROR;
} }
@ -74,8 +78,8 @@ if (g_Weight15 == NULL) {
return RET_ERROR; return RET_ERROR;
} }
memset(g_Weight15, 0, 6080); memset(g_Weight15, 0, 6080);
int init_filter_zp[20] = {1, 12, 3, 2, -10, -5, -11, 5, 12, 22, 16, 1, -5, 15, 13, 5, -10, -5, -6, 0}; int init_filter_zp[1] = {0};
InitInt8MatrixB(g_Weight6, g_Weight16, g_Weight15, 1, 300, 20, 20, 304, 0, init_filter_zp, g_Weight14, true, true); InitInt8MatrixB(g_Weight6, g_Weight16, g_Weight15, 1, 300, 20, 20, 304, 31, init_filter_zp, g_Weight14, true, false);
} }
{ {
g_Weight17 = malloc(48); g_Weight17 = malloc(48);
@ -94,8 +98,8 @@ if (g_Weight18 == NULL) {
return RET_ERROR; return RET_ERROR;
} }
memset(g_Weight18, 0, 384); memset(g_Weight18, 0, 384);
int init_filter_zp[10] = {7, -2, 9, 2, -6, 21, 16, 10, -19, 8}; int init_filter_zp[1] = {0};
InitInt8MatrixB(g_Weight8, g_Weight19, g_Weight18, 1, 20, 10, 12, 32, 0, init_filter_zp, g_Weight17, true, true); InitInt8MatrixB(g_Weight8, g_Weight19, g_Weight18, 1, 20, 10, 12, 32, 11, init_filter_zp, g_Weight17, true, false);
} }
return RET_OK; return RET_OK;
} }

6
mindspore/lite/micro/example/mnist_x86/src/weight.h
View File

@ -17,7 +17,6 @@
#include "nnacl/common_func.h" #include "nnacl/common_func.h"
#include "nnacl/errorcode.h" #include "nnacl/errorcode.h"
#include "nnacl/fp32/softmax_fp32.h"
#include "nnacl/int8/common_func_int8.h" #include "nnacl/int8/common_func_int8.h"
#include "nnacl/int8/conv3x3_int8.h" #include "nnacl/int8/conv3x3_int8.h"
#include "nnacl/int8/conv_int8.h" #include "nnacl/int8/conv_int8.h"
@ -25,6 +24,7 @@
#include "nnacl/int8/pooling_int8.h" #include "nnacl/int8/pooling_int8.h"
#include "nnacl/int8/quant_dtype_cast_int8.h" #include "nnacl/int8/quant_dtype_cast_int8.h"
#include "nnacl/int8/reshape_int8.h" #include "nnacl/int8/reshape_int8.h"
#include "nnacl/int8/softmax_int8.h"
#include "wrapper/int8/matmul_int8_wrapper.h" #include "wrapper/int8/matmul_int8_wrapper.h"
#include <stdlib.h> #include <stdlib.h>
#include <string.h> #include <string.h>
@ -45,3 +45,7 @@ extern int32_t *g_Weight16;
extern int32_t *g_Weight17; extern int32_t *g_Weight17;
extern int8_t *g_Weight18; extern int8_t *g_Weight18;
extern int32_t *g_Weight19; extern int32_t *g_Weight19;
extern int8_t g_Weight6[];
extern int32_t g_Weight7[];
extern int8_t g_Weight8[];
extern int32_t g_Weight9[];

31
mindspore/lite/micro/example/mobilenetv2/README.md
View File

@ -55,12 +55,26 @@ codegen编译[MobileNetv2模型](https://download.mindspore.cn/model_zoo/officia
在编译此工程之前需要预先获取安卓平台对应的[Release包](https://www.mindspore.cn/tutorial/lite/zh-CN/master/use/downloads.html)。 在编译此工程之前需要预先获取安卓平台对应的[Release包](https://www.mindspore.cn/tutorial/lite/zh-CN/master/use/downloads.html)。
算子静态库的目录如下: 安卓平台对应的Release包的目录如下:
```text
```bash mindspore-lite-{version}-inference-android-{arch}
├── operator_library # 对应平台算子库目录 ├── inference
├── include # 对应平台算子库头文件目录 │ ├── include # 推理框架头文件
└── lib # 对应平台算子库静态库目录 │ ├── lib # 推理框架库
│ │ ├── libmindspore-lite.a # MindSpore Lite推理框架的静态库
│ │ └── libmindspore-lite.so # MindSpore Lite推理框架的动态库
│ ├── minddata # 图像处理库
│ │ ├── include
│ │ └── lib
│ │ └── libminddata-lite.so # 图像处理动态库文件
│ └── third_party # NPU库
│ └── hiai_ddk
└── tools
├── benchmark # 基准测试工具
│ └── benchmark
└── codegen # 代码生成工具
├── include # 算子头文件
└── lib # 算子静态库
``` ```
生成代码工程目录如下： 生成代码工程目录如下：
@ -68,7 +82,6 @@ codegen编译[MobileNetv2模型](https://download.mindspore.cn/model_zoo/officia
```bash ```bash
├── mobilenetv2 # 生成代码的根目录 ├── mobilenetv2 # 生成代码的根目录
├── benchmark # 生成代码的benchmark目录 ├── benchmark # 生成代码的benchmark目录
├── include # 模型推理代码对外暴露头文件目录
└── src # 模型推理代码目录 └── src # 模型推理代码目录
``` ```
@ -91,7 +104,7 @@ cmake -DCMAKE_BUILD_TYPE=Release \
-DANDROID_TOOLCHAIN_NAME="aarch64-linux-android-clang" \ -DANDROID_TOOLCHAIN_NAME="aarch64-linux-android-clang" \
-DANDROID_NATIVE_API_LEVEL="19" \ -DANDROID_NATIVE_API_LEVEL="19" \
-DPLATFORM_ARM64=ON \ -DPLATFORM_ARM64=ON \
-DPKG_PATH={path to}/mindspore-lite-{version}-inference-android .. -DPKG_PATH={path to}/mindspore-lite-{version}-inference-android-{arch} ..
make make
``` ```
@ -104,7 +117,7 @@ cmake -DCMAKE_BUILD_TYPE=Release \
-DANDROID_TOOLCHAIN_NAME="clang" \ -DANDROID_TOOLCHAIN_NAME="clang" \
-DANDROID_NATIVE_API_LEVEL="19" \ -DANDROID_NATIVE_API_LEVEL="19" \
-DMICRO_BUILD_ARM32=ON \ -DMICRO_BUILD_ARM32=ON \
-DPKG_PATH={path to}/mindspore-lite-{version}-inference-android .. -DPKG_PATH={path to}/mindspore-lite-{version}-inference-android-{arch} ..
make make
``` ```