mirror of https://github.com/alibaba/MNN.git
618 lines
23 KiB
C++
618 lines
23 KiB
C++
//
|
|
// ConvolutionCommon.cpp
|
|
// MNN
|
|
//
|
|
// Created by MNN on 2020/03/02.
|
|
// Copyright © 2018, Alibaba Group Holding Limited
|
|
//
|
|
|
|
#include "ConvolutionCommon.hpp"
|
|
#include <math.h>
|
|
#include "half.hpp"
|
|
namespace MNN {
|
|
static inline void *MNNMemoryAllocAlignZeroAlign(size_t size) {
|
|
return MNNMemoryCallocAlign(size, MNN_MEMORY_ALIGN_DEFAULT);
|
|
}
|
|
static int ReadBlobDim(unsigned char *&myfile, unsigned short *shape, int shapeBufCnt) {
|
|
int uSize = myfile[0];
|
|
myfile++;
|
|
if (uSize > 4) {
|
|
printf("Read shape error!\n");
|
|
return 0;
|
|
}
|
|
int copyLength = uSize;
|
|
if (copyLength > shapeBufCnt) {
|
|
copyLength = shapeBufCnt;
|
|
}
|
|
::memcpy(shape, myfile, sizeof(unsigned short) * copyLength);
|
|
myfile += copyLength * sizeof(unsigned short);
|
|
return copyLength;
|
|
}
|
|
|
|
static double _log2(double x) {
|
|
return log(x) / log(2);
|
|
}
|
|
|
|
static uint32_t atLestBitsCnt(uint32_t n) {
|
|
for (uint32_t i = 0; i < 32; i++) {
|
|
int32_t t = n << i;
|
|
if (t < 0)
|
|
return 32 - i - (((t << 1) == 0) ? 1 : 0);
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
static void SplitBufToArray(uint8_t *buf, size_t bufLen, uint8_t *arr, size_t arrLen, size_t iNeedBits) {
|
|
unsigned char cMask = (1 << (iNeedBits)) - 1;
|
|
unsigned char *tmp = (unsigned char *)buf;
|
|
int iOffset = 0;
|
|
for (unsigned int i = 0; i < arrLen; i++) {
|
|
unsigned char idx = 0;
|
|
long uShift = 8 - iNeedBits - iOffset % 8;
|
|
if (uShift < 0) {
|
|
idx = (tmp[iOffset / 8] << (0 - uShift)) & cMask;
|
|
idx |= (tmp[(iOffset / 8) + 1] >> (8 + uShift)) & cMask;
|
|
} else {
|
|
idx = (tmp[iOffset / 8] >> uShift) & cMask;
|
|
}
|
|
iOffset += iNeedBits;
|
|
if (iOffset % 8 == 0) {
|
|
tmp += iOffset / 8;
|
|
iOffset = 0;
|
|
}
|
|
arr[i] = idx;
|
|
}
|
|
}
|
|
|
|
// fixme!!! not efficiency
|
|
typedef struct _SIMPLE_SET {
|
|
int8_t *UniSet;
|
|
uint32_t UniSetSize;
|
|
uint32_t CurUniCnt;
|
|
} SIMPLE_SET, *PSIMPLE_SET;
|
|
|
|
static PSIMPLE_SET CreateSimpleSet(uint32_t maxSize) {
|
|
PSIMPLE_SET set = (PSIMPLE_SET)calloc(1, sizeof(SIMPLE_SET));
|
|
if (set == nullptr)
|
|
return nullptr;
|
|
set->UniSet = (int8_t *)calloc(maxSize, sizeof(int8_t));
|
|
set->UniSetSize = maxSize;
|
|
set->CurUniCnt = 0;
|
|
return set;
|
|
}
|
|
|
|
static void SimpleRank(int8_t *data, uint32_t cnt, int up) {
|
|
if (up) {
|
|
for (uint32_t i = 0; i < cnt; i++) {
|
|
for (uint32_t j = i + 1; j < cnt; j++) {
|
|
if (data[i] > data[j]) {
|
|
int8_t tmp = data[i];
|
|
data[i] = data[j];
|
|
data[j] = tmp;
|
|
}
|
|
}
|
|
}
|
|
} else {
|
|
for (uint32_t i = 0; i < cnt; i++) {
|
|
for (uint32_t j = i + 1; j < cnt; j++) {
|
|
if (data[i] < data[j]) {
|
|
int8_t tmp = data[i];
|
|
data[i] = data[j];
|
|
data[j] = tmp;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
static void InsertSimpleSet(PSIMPLE_SET set, int8_t value) {
|
|
if (set->CurUniCnt >= set->UniSetSize)
|
|
return;
|
|
for (uint32_t i = 0; i < set->CurUniCnt; i++) {
|
|
if (set->UniSet[i] == value)
|
|
return;
|
|
}
|
|
set->UniSet[set->CurUniCnt++] = value;
|
|
// SimpleRank(set->UniSet, set->CurUniCnt, 1);
|
|
}
|
|
|
|
void DestorySimpleSet(PSIMPLE_SET set) {
|
|
if (set->UniSet != nullptr)
|
|
free(set->UniSet);
|
|
free(set);
|
|
}
|
|
|
|
typedef struct _SIMPLE_MAP {
|
|
int8_t *CharCharMap;
|
|
uint32_t CharMapSize;
|
|
uint32_t CurMapCnt;
|
|
} SIMPLE_MAP, *PSIMPLE_MAP;
|
|
|
|
static PSIMPLE_MAP CreateSimpleMap(uint32_t MaxCnt) {
|
|
PSIMPLE_MAP map = (PSIMPLE_MAP)calloc(1, sizeof(SIMPLE_MAP));
|
|
if (map == nullptr)
|
|
return nullptr;
|
|
map->CharMapSize = MaxCnt * sizeof(int8_t);
|
|
map->CurMapCnt = 0;
|
|
map->CharCharMap = (int8_t *)calloc(1, MaxCnt * 2);
|
|
return map;
|
|
}
|
|
|
|
static void DestroySimpleMap(PSIMPLE_MAP map) {
|
|
if (map->CharCharMap)
|
|
free(map->CharCharMap);
|
|
free(map);
|
|
}
|
|
|
|
static void InsertMap(PSIMPLE_MAP map, int8_t k, int8_t v) {
|
|
for (uint32_t i = 0; i < map->CurMapCnt; i++) {
|
|
if (map->CharCharMap[i * 2] == k) {
|
|
map->CharCharMap[i * 2 + 1] = v;
|
|
return;
|
|
}
|
|
}
|
|
if (map->CurMapCnt >= map->CharMapSize)
|
|
return;
|
|
map->CharCharMap[map->CurMapCnt * 2] = k;
|
|
map->CharCharMap[map->CurMapCnt * 2 + 1] = v;
|
|
map->CurMapCnt++;
|
|
}
|
|
|
|
static int8_t FindInMap(PSIMPLE_MAP map, int8_t k, int *found) {
|
|
for (uint32_t i = 0; i < map->CurMapCnt; i++) {
|
|
if (map->CharCharMap[i * 2] == k) {
|
|
if (found != nullptr)
|
|
*found = 1;
|
|
return map->CharCharMap[i * 2 + 1];
|
|
}
|
|
}
|
|
if (found != nullptr)
|
|
*found = 0;
|
|
return 0;
|
|
}
|
|
|
|
static void StreamSizeRead(void *dst, int unit, size_t count, unsigned char *&file) {
|
|
::memcpy(dst, file, unit * count);
|
|
file += (unit * count);
|
|
}
|
|
|
|
static int8_t *ReadQuanData_c(unsigned char *&s, uint32_t *len) {
|
|
int8_t *blob = nullptr;
|
|
int8_t *samples = nullptr;
|
|
uint8_t *idxBuf = nullptr;
|
|
uint8_t *idxBytes = nullptr;
|
|
uint32_t dataCnt = 1;
|
|
|
|
do {
|
|
// blob shape
|
|
unsigned short shape[64] = {0};
|
|
uint32_t shapeDim = (uint32_t)ReadBlobDim(s, shape, 64);
|
|
if (shapeDim == 0 || shapeDim > 64)
|
|
break;
|
|
for (uint32_t i = 0; i < shapeDim; i++)
|
|
dataCnt *= shape[i];
|
|
|
|
// sample
|
|
uint32_t sampleCnt = 0;
|
|
StreamSizeRead(&sampleCnt, 1, 1, s);
|
|
if (0 == sampleCnt) {
|
|
sampleCnt = 256;
|
|
}
|
|
samples = (int8_t *)MNNMemoryAllocAlignZeroAlign(sampleCnt);
|
|
if (samples == nullptr)
|
|
break;
|
|
StreamSizeRead(samples, 1, sampleCnt, s);
|
|
SimpleRank(samples, sampleCnt, 1);
|
|
// index
|
|
uint32_t idxBitsCnt = atLestBitsCnt(sampleCnt);
|
|
idxBitsCnt = idxBitsCnt < 1 ? 1 : idxBitsCnt;
|
|
size_t idxBufSize = ceil(idxBitsCnt * dataCnt * 0.125);
|
|
idxBuf = (uint8_t *)MNNMemoryAllocAlignZeroAlign(idxBufSize);
|
|
if (nullptr == idxBuf) {
|
|
MNN_ERROR("Not enought memory\n");
|
|
break;
|
|
}
|
|
StreamSizeRead(idxBuf, 1, idxBufSize, s);
|
|
// split index value into bytes
|
|
idxBytes = (uint8_t *)MNNMemoryAllocAlignZeroAlign(dataCnt * sizeof(uint8_t));
|
|
if (idxBitsCnt == 0 || nullptr == idxBytes) {
|
|
break;
|
|
}
|
|
SplitBufToArray(idxBuf, (uint32_t)idxBufSize, idxBytes, (uint32_t)dataCnt, (uint32_t)idxBitsCnt);
|
|
int i = 0;
|
|
blob = (int8_t *)MNNMemoryAllocAlignZeroAlign((size_t)dataCnt);
|
|
if (nullptr == blob) {
|
|
break;
|
|
}
|
|
for (i = 0; i < dataCnt; i++) {
|
|
if (idxBytes[i] >= sampleCnt) {
|
|
MNN_PRINT("iNeedBits is %u\nRead quan weights error with idx:%d\n", idxBitsCnt, (int)idxBytes[i]);
|
|
break;
|
|
}
|
|
blob[i] = samples[idxBytes[i]];
|
|
}
|
|
if (i < dataCnt) {
|
|
MNNMemoryFreeAlign(blob);
|
|
blob = nullptr;
|
|
break;
|
|
}
|
|
} while (0);
|
|
|
|
if (samples != nullptr)
|
|
MNNMemoryFreeAlign(samples);
|
|
if (idxBuf != nullptr)
|
|
MNNMemoryFreeAlign(idxBuf);
|
|
if (idxBytes != nullptr)
|
|
MNNMemoryFreeAlign(idxBytes);
|
|
if (len)
|
|
*len = blob ? dataCnt : 0;
|
|
return blob;
|
|
}
|
|
|
|
static int8_t *ReadSparseQuanData_c(unsigned char *&myfile, uint32_t *len, const flatbuffers::Vector<float> *alpha) {
|
|
// MNN_ERROR("sparse:%d\n", 1);
|
|
unsigned short shape[64] = {0};
|
|
uint32_t ucMapSize = 0;
|
|
PSIMPLE_SET setWeight = CreateSimpleSet(256);
|
|
if (setWeight == nullptr) {
|
|
return nullptr;
|
|
}
|
|
std::shared_ptr<unsigned int> __autoReleaseSetWeight(nullptr, [setWeight](void *) { DestorySimpleSet(setWeight); });
|
|
unsigned int nnz;
|
|
unsigned char iIdxNeedBits;
|
|
int8_t *blob = nullptr;
|
|
// 1. weights blob shape(unsigned int32)
|
|
int ShapeDim = ReadBlobDim(myfile, shape, 64);
|
|
int Size = sizeof(int8_t);
|
|
for (int i = 0; i < ShapeDim; i++)
|
|
Size *= shape[i];
|
|
blob = (int8_t *)MNNMemoryAllocAlignZeroAlign((size_t)Size);
|
|
if (blob == nullptr)
|
|
return nullptr;
|
|
// 2. nnz
|
|
StreamSizeRead(&nnz, 4, 1, myfile);
|
|
// 3. max_step use # bits () (unsigned char)
|
|
StreamSizeRead(&iIdxNeedBits, 1, 1, myfile);
|
|
// read idx array
|
|
// 4. buf for steps ceil(nnz*step need bits/8)
|
|
AutoStorage<unsigned char> arrIdxBuffer(nnz);
|
|
unsigned char *arrIdx = arrIdxBuffer.get();
|
|
if (nullptr == arrIdx) {
|
|
return nullptr;
|
|
}
|
|
{
|
|
size_t bufLen = (size_t)(ceil(0.125 * iIdxNeedBits * nnz));
|
|
char *buf = (char *)MNNMemoryAllocAlignZeroAlign(bufLen * sizeof(char));
|
|
if (nullptr == buf) {
|
|
return nullptr;
|
|
}
|
|
StreamSizeRead(buf, 1, bufLen, myfile);
|
|
SplitBufToArray((uint8_t *)buf, (uint32_t)bufLen, (uint8_t *)arrIdx, (uint32_t)nnz, (uint32_t)iIdxNeedBits);
|
|
MNNMemoryFreeAlign(buf);
|
|
}
|
|
// 5. Avalable values Count(unsigned char)
|
|
StreamSizeRead(&ucMapSize, 1, 1, myfile);
|
|
if (0 == ucMapSize) {
|
|
ucMapSize = 256;
|
|
}
|
|
// 6. valueset(signed char * valueset_size)
|
|
for (int i = 0; i < ucMapSize; i++) {
|
|
int8_t tmp;
|
|
StreamSizeRead(&tmp, 1, 1, myfile);
|
|
InsertSimpleSet(setWeight, tmp);
|
|
}
|
|
SimpleRank(setWeight->UniSet, setWeight->CurUniCnt, 1);
|
|
// map<unsigned char, signed char> mapWeight;
|
|
PSIMPLE_MAP mapWeight = CreateSimpleMap(256);
|
|
if (mapWeight == nullptr) {
|
|
return nullptr;
|
|
}
|
|
std::shared_ptr<unsigned int> __autoReleaseMapWeight(nullptr, [mapWeight](void *) { DestroySimpleMap(mapWeight); });
|
|
|
|
for (int i = 0; i < setWeight->CurUniCnt; i++) {
|
|
InsertMap(mapWeight, i, setWeight->UniSet[i]);
|
|
}
|
|
// unsigned char iIdx = 0;
|
|
// 7. none zero weights indexes(nnz*ceil(log2(Avalable_values_Count))/8)
|
|
AutoStorage<unsigned char> arrWeightIdxBuffer(nnz);
|
|
unsigned char *arrWeightIdx = arrWeightIdxBuffer.get();
|
|
if (nullptr == arrWeightIdx) {
|
|
return nullptr;
|
|
}
|
|
{
|
|
int iDataNeedBits = (int)ceil(_log2(ucMapSize));
|
|
iDataNeedBits = iDataNeedBits < 1 ? 1 : iDataNeedBits;
|
|
size_t bufLen = (size_t)(ceil(0.125 * iDataNeedBits * nnz));
|
|
char *buf = (char *)MNNMemoryAllocAlignZeroAlign(bufLen * sizeof(char));
|
|
if (nullptr == buf) {
|
|
return nullptr;
|
|
}
|
|
StreamSizeRead(buf, 1, bufLen, myfile);
|
|
SplitBufToArray((uint8_t *)buf, (uint32_t)bufLen, (uint8_t *)arrWeightIdx, (uint32_t)nnz,
|
|
(uint32_t)iDataNeedBits);
|
|
MNNMemoryFreeAlign(buf);
|
|
}
|
|
// set blob data with idx and weight idx
|
|
{
|
|
if (alpha->size() == 2 * shape[0]) {
|
|
auto alphaPtr = alpha->data();
|
|
int area = Size / shape[0];
|
|
for (int i = 0; i < shape[0]; i++) {
|
|
float min = alphaPtr[2*i];
|
|
float scale = alphaPtr[2*i+1];
|
|
int zeroQuant = -128;
|
|
if (scale > 1e-6) {
|
|
zeroQuant = round((0.0f - min) / scale) + (-128);
|
|
}
|
|
memset(blob+area*i, zeroQuant, area * sizeof(signed char));
|
|
}
|
|
} else {
|
|
memset(blob, 0, Size * sizeof(signed char)); //backward compability with previous symmetric weight quant
|
|
}
|
|
int iPreIdx = 0;
|
|
for (int i = 0; i < nnz; i++) {
|
|
iPreIdx += arrIdx[i];
|
|
int found = 0;
|
|
int8_t value = FindInMap(mapWeight, arrWeightIdx[i], &found);
|
|
if (!found) {
|
|
MNN_ERROR("Read quan weights error with idx:%d\n", arrWeightIdx[i]);
|
|
MNNMemoryFreeAlign(blob);
|
|
return nullptr;
|
|
}
|
|
blob[iPreIdx] = value;
|
|
}
|
|
}
|
|
*len = Size;
|
|
return blob;
|
|
}
|
|
std::shared_ptr<ConvolutionCommon::Int8Common> ConvolutionCommon::load(const IDSTQuan *quan, bool forceFloat, bool forceInt8) {
|
|
auto result = std::make_shared<Int8Common>();
|
|
uint32_t weightLength = 0;
|
|
int8_t *buffer = nullptr;
|
|
auto originBuffer = (unsigned char *)quan->buffer()->data();
|
|
if (1 == quan->type()) {
|
|
buffer = ReadQuanData_c(originBuffer, &weightLength);
|
|
}
|
|
if (2 == quan->type()) {
|
|
buffer = ReadSparseQuanData_c(originBuffer, &weightLength, quan->alpha());
|
|
}
|
|
// read fp16 data
|
|
if (3 == quan->type()) {
|
|
weightLength = quan->buffer()->size() / sizeof(half_float::half);
|
|
std::vector<int8_t> tempHalfWeight(quan->buffer()->size());
|
|
::memcpy(tempHalfWeight.data(), quan->buffer()->data(), quan->buffer()->size());
|
|
auto halfWeight = reinterpret_cast<half_float::half *>(tempHalfWeight.data());
|
|
result->weightFloat.reset(weightLength);
|
|
if (nullptr == result->weightFloat.get()) {
|
|
MNN_PRINT("Alloc memory error for extract fp16 back to float\n");
|
|
return nullptr;
|
|
}
|
|
std::transform(halfWeight, halfWeight + weightLength, result->weightFloat.get(),
|
|
[](half_float::half h) { return float(h); });
|
|
return result;
|
|
}
|
|
|
|
// weight int8 only
|
|
if (4 == quan->type()) {
|
|
weightLength = quan->buffer()->size();
|
|
result->weight.reset(weightLength);
|
|
::memcpy(result->weight.get(), quan->buffer()->data(), weightLength);
|
|
}
|
|
|
|
if (result->weight.get() == nullptr) {
|
|
if (nullptr == buffer) {
|
|
MNN_PRINT("Alloc memory error for extract idst int8\n");
|
|
return nullptr;
|
|
}
|
|
result->weight.set(buffer, weightLength);
|
|
}
|
|
result->quan = quan;
|
|
result->alpha.reset(quan->alpha()->size());
|
|
if (nullptr == result->alpha.get()) {
|
|
MNN_PRINT("Alloc memory error for extract idst int8\n");
|
|
return nullptr;
|
|
}
|
|
::memcpy(result->alpha.get(), quan->alpha()->data(), quan->alpha()->size() * sizeof(float));
|
|
if (forceInt8) {
|
|
return result;
|
|
}
|
|
if (!quan->has_scaleInt() || forceFloat) {
|
|
// Back to float
|
|
result->weightFloat.reset(weightLength);
|
|
if (nullptr == result->weightFloat.get()) {
|
|
MNN_PRINT("Alloc memory error for extract idst int8/ Back to float\n");
|
|
return nullptr;
|
|
}
|
|
int outputCount = 0;
|
|
bool oldType4 = (quan->type() == 4 && quan->aMin() == 0 && std::abs(quan->quantScale()) < 1e-6);
|
|
if (quan->readType() != 0 || oldType4) {
|
|
outputCount = result->alpha.size() / 2;
|
|
} else {
|
|
outputCount = result->alpha.size(); // backward compability with previous symmetric quantization
|
|
}
|
|
int partWeightSize = weightLength / outputCount;
|
|
for (int o = 0; o < outputCount; ++o) {
|
|
auto dstW = result->weightFloat.get() + o * partWeightSize;
|
|
auto srcW = result->weight.get() + o * partWeightSize;
|
|
float extraFactor = quan->quantScale();
|
|
// for old type 4 models, their quan->quantScale is 0. which will introduce a bug here
|
|
if (oldType4) {
|
|
extraFactor = 1.0f;
|
|
}
|
|
if (result->alpha.size() == 2 * outputCount) {
|
|
float min = result->alpha.get()[2*o];
|
|
float alpha = result->alpha.get()[2*o+1];
|
|
// clampMin is minVal in asymmetric quant, clampMin = -(2^(bit))
|
|
// and old version clampMin is -128
|
|
float clampMin = quan->aMin() == 0 ? -128 : quan->aMin();
|
|
for (int j = 0; j < partWeightSize; ++j) {
|
|
dstW[j] = (( (float)srcW[j] - clampMin ) * alpha + min) * extraFactor;
|
|
}
|
|
} else {
|
|
float alpha = result->alpha.get()[o];
|
|
for (int j = 0; j < partWeightSize; ++j) {
|
|
dstW[j] = ((float)srcW[j]) * alpha * extraFactor;
|
|
}
|
|
}
|
|
}
|
|
|
|
result->weight.release();
|
|
result->alpha.release();
|
|
}
|
|
|
|
return result;
|
|
}
|
|
|
|
void ConvolutionCommon::getConvParameters(std::shared_ptr<Int8Common> *quanCommon, const MNN::Convolution2D *conv2d, const float** originWeight, int* originWeightSize) {
|
|
*originWeight = nullptr;
|
|
*originWeightSize = 0;
|
|
if (nullptr != conv2d->quanParameter()) {
|
|
*quanCommon = load(conv2d->quanParameter(), false);
|
|
*originWeight = (*quanCommon)->weightFloat.get();
|
|
*originWeightSize = (*quanCommon)->weightFloat.size();
|
|
}
|
|
if (*originWeight == nullptr) {
|
|
*originWeight = conv2d->weight()->data();
|
|
*originWeightSize = conv2d->weight()->size();
|
|
}
|
|
}
|
|
|
|
bool ConvolutionCommon::getConvInt8Parameters(const MNN::Convolution2D* conv2d, std::shared_ptr<Int8Common>& quanCommon,
|
|
const int8_t*& weight, float*& scale, int32_t*& bias,
|
|
float inputScale, float outputScale, int inputZeroPoint, int outputZeroPoint) {
|
|
int outputCount = conv2d->common()->outputCount();
|
|
// fix xcode UndefinedBehaviorSanitizer
|
|
if (conv2d->symmetricQuan()->weight() != nullptr) {
|
|
weight = conv2d->symmetricQuan()->weight()->data();
|
|
}
|
|
if (conv2d->quanParameter() && conv2d->quanParameter()->buffer()) {
|
|
quanCommon = ConvolutionCommon::load(conv2d->quanParameter(), false, true);
|
|
weight = quanCommon->weight.get();
|
|
}
|
|
if (weight == nullptr) {
|
|
MNN_ERROR("ConvolutionCommon::getConvInt8Parameters: No weight data!");
|
|
return false;
|
|
}
|
|
if (conv2d->symmetricQuan()->bias() && conv2d->symmetricQuan()->scale()) {
|
|
MNN_ASSERT(conv2d->symmetricQuan()->bias()->size() == outputCount && conv2d->symmetricQuan()->scale()->size() == outputCount);
|
|
::memcpy(bias, conv2d->symmetricQuan()->bias()->data(), outputCount * sizeof(int32_t));
|
|
::memcpy(scale, conv2d->symmetricQuan()->scale()->data(), outputCount * sizeof(float));
|
|
return true;
|
|
}
|
|
if (conv2d->bias() && conv2d->quanParameter()->alpha()) {
|
|
const int kernelNum = conv2d->common()->outputCount();
|
|
int kernelChannel = conv2d->common()->inputCount();
|
|
int group = conv2d->common()->group();
|
|
if ((kernelChannel == kernelNum) && (group == kernelChannel)) {
|
|
kernelChannel = 1; // depthwise
|
|
}
|
|
const int kernelSize = kernelChannel * conv2d->common()->kernelX() * conv2d->common()->kernelY();
|
|
|
|
// // reference for how to get quantized bias
|
|
// auto remains = _ReduceSum(_Cast<int32_t>(mInputZeroPoint) * _Cast<int32_t>(quanWeight), {1, 2, 3}, true);
|
|
// MNN_ASSERT((mOutputZeroPoint->getInfo()->dim.size() == 0) && (mOutputZeroPoint->getInfo()->size == 1)); // only support per-tensor, per-channel is removed.
|
|
// auto outputZeroPointFused = _Cast<int32_t>(_Cast<float>(mOutputZeroPoint) * _Reciprocal(convScale));
|
|
// auto quanBias = _Cast<int32_t>(fusedBias * _Reciprocal(weightScale * mInputScale)) - remains + outputZeroPointFused;
|
|
|
|
|
|
// compute remains used in asymmetric quant
|
|
std::vector<int> remains;
|
|
for (int i = 0; i < kernelNum; i++) {
|
|
int temp = 0;
|
|
int offset = i * kernelSize;
|
|
for (int j = 0; j < kernelSize; j++) {
|
|
temp += inputZeroPoint * weight[offset + j];
|
|
}
|
|
remains.emplace_back(temp);
|
|
}
|
|
|
|
inputScale = inputScale == 0.f ? conv2d->quanParameter()->scaleIn() : inputScale;
|
|
outputScale = outputScale == 0.f ? conv2d->quanParameter()->scaleOut() : outputScale;
|
|
auto biasData = conv2d->bias()->data();
|
|
auto alphaData = conv2d->quanParameter()->alpha()->data();
|
|
auto alphaScale = inputScale / outputScale;
|
|
for (int i = 0; i < outputCount; i++) {
|
|
auto alphaValue = alphaData[i];
|
|
if (fabs(alphaValue) < 1e-6) {
|
|
alphaValue = 1e-6;
|
|
}
|
|
scale[i] = alphaValue * alphaScale;
|
|
// compute outputZeroPointFused in asymmetric quant
|
|
int outputZeroPointFused = static_cast<int32_t>(outputZeroPoint / scale[i]);
|
|
bias[i] = static_cast<int32_t>(biasData[i] / (inputScale * alphaValue)) - remains[i] + outputZeroPointFused;
|
|
}
|
|
return true;
|
|
}
|
|
MNN_ERROR("ConvolutionCommon::getConvInt8Parameters: No bias & scale data!");
|
|
return false;
|
|
}
|
|
|
|
std::pair<int, int> ConvolutionCommon::convolutionPad(const Tensor *input, const Tensor *output,
|
|
const Convolution2DCommon *mCommon) {
|
|
if (mCommon->padMode() == PadMode_SAME) {
|
|
int kernelWidthSize = (mCommon->kernelX() - 1) * mCommon->dilateX() + 1;
|
|
int kernelHeightSize = (mCommon->kernelY() - 1) * mCommon->dilateY() + 1;
|
|
|
|
int padNeededWidth = (output->width() - 1) * mCommon->strideX() + kernelWidthSize - input->width();
|
|
int padNeededHeight = (output->height() - 1) * mCommon->strideY() + kernelHeightSize - input->height();
|
|
auto mPadX = padNeededWidth / 2;
|
|
auto mPadY = padNeededHeight / 2;
|
|
return std::make_pair(mPadX, mPadY);
|
|
}
|
|
auto mPadX = mCommon->padX();
|
|
auto mPadY = mCommon->padY();
|
|
if (nullptr != mCommon->pads() && mCommon->pads()->size() >= 2) {
|
|
mPadX = mCommon->pads()->data()[1];
|
|
mPadY = mCommon->pads()->data()[0];
|
|
}
|
|
return std::make_pair(mPadX, mPadY);
|
|
}
|
|
|
|
std::tuple<int, int, int, int> ConvolutionCommon::convolutionPadFull(const Tensor* input, const Tensor* output,
|
|
const Convolution2DCommon* common) {
|
|
auto pad = convolutionPad(input, output, common);
|
|
int iw = input->width();
|
|
int ih = input->height();
|
|
int ow = output->width();
|
|
int oh = output->height();
|
|
|
|
int right = (ow - 1) * common->strideX() + (common->kernelX() - 1) * common->dilateX() - pad.first;
|
|
int padRight = 0;
|
|
if (right >= iw) {
|
|
padRight = right - iw + 1;
|
|
}
|
|
int bottom = (oh - 1) * common->strideY() + (common->kernelY() - 1) * common->dilateY() - pad.second;
|
|
int padBottom = 0;
|
|
if (bottom >= ih) {
|
|
padBottom = bottom - ih + 1;
|
|
}
|
|
return std::make_tuple(pad.first, pad.second, padRight, padBottom);
|
|
}
|
|
|
|
std::pair<int, int> ConvolutionCommon::convolutionTransposePad(const Tensor *input, const Tensor *output,
|
|
const Convolution2DCommon *mCommon) {
|
|
if (mCommon->padMode() == PadMode_SAME) {
|
|
const int outputWidth = output->width();
|
|
const int outputHeight = output->height();
|
|
|
|
const int outputWidthPadded = (input->width() - 1) * mCommon->strideX() + mCommon->kernelX();
|
|
const int outputHeightPadded = (input->height() - 1) * mCommon->strideY() + mCommon->kernelY();
|
|
|
|
const int padNeededWidth = outputWidthPadded - outputWidth;
|
|
const int padNeededHeight = outputHeightPadded - outputHeight;
|
|
|
|
auto mPadX = padNeededWidth / 2;
|
|
auto mPadY = padNeededHeight / 2;
|
|
return std::make_pair(mPadX, mPadY);
|
|
}
|
|
auto mPadX = mCommon->padX();
|
|
auto mPadY = mCommon->padY();
|
|
if (nullptr != mCommon->pads() && mCommon->pads()->size() >= 2) {
|
|
mPadY = mCommon->pads()->data()[0];
|
|
mPadX = mCommon->pads()->data()[1];
|
|
}
|
|
return std::make_pair(mPadX, mPadY);
|
|
}
|
|
|
|
} // namespace MNN
|