Skip to content

N
NiuTrans.Tensor
Commit 100f4611 by liyinqiao
Options

Bug fixed: 1. MatrixMul, Select, Sort, TopK, Loss; 2. Add other tests.

parent a3a7145f
正在显示 包含 57 行增加 和 40 行删除
source/core/MatrixMul.cpp
......@@ -58,12 +58,12 @@ void MatrixMul(XTensor * a, MATRIX_TRANS_TYPE transposedA,
CheckNTErrors((a->order >= 2 && b->order >= 2 && c->order >= 2),
"Input tensors must have a order > 2!");
int an = transposedA == X_TRANS ? a->dimSize[1] : a->dimSize[0];
int am = transposedA == X_TRANS ? a->dimSize[0] : a->dimSize[1];
int bn = transposedB == X_TRANS ? b->dimSize[1] : b->dimSize[0];
int bm = transposedB == X_TRANS ? b->dimSize[0] : b->dimSize[1];
int cn = c->dimSize[0];
int cm = c->dimSize[1];
int an = transposedA == X_TRANS ? a->dimSizeRDI[0] : a->dimSizeRDI[1];
int am = transposedA == X_TRANS ? a->dimSizeRDI[1] : a->dimSizeRDI[0];
int bn = transposedB == X_TRANS ? b->dimSizeRDI[0] : b->dimSizeRDI[1];
int bm = transposedB == X_TRANS ? b->dimSizeRDI[1] : b->dimSizeRDI[0];
int cn = c->dimSizeRDI[1];
int cm = c->dimSizeRDI[0];
CheckNTErrors((am == bn && an == cn && bm == cm),
"Unmatched tensors in multiplication!");
......@@ -79,13 +79,13 @@ void MatrixMul(XTensor * a, MATRIX_TRANS_TYPE transposedA,
int cBlockNum = 1;
for (int i = 2; i < a->order; i++) {
CheckNTErrors((a->dimSizeRDI[i] == c->dimSizeRDI[i]), "Incorrect tensor sizes!");
CheckNTErrors((a->dimSizeRDI[i] == c->dimSizeRDI[i - 2 + b->order]), "Incorrect tensor sizes!");
aBlockNum *= a->dimSizeRDI[i];
cBlockNum *= a->dimSizeRDI[i];
}
for (int i = 2; i < b->order; i++) {
CheckNTErrors((b->dimSizeRDI[i] == c->dimSizeRDI[i - 2 + a->order]), "Incorrect tensor sizes!");
CheckNTErrors((b->dimSizeRDI[i] == c->dimSizeRDI[i]), "Incorrect tensor sizes!");
bBlockNum *= b->dimSizeRDI[i];
cBlockNum *= b->dimSizeRDI[i];
}
......@@ -93,9 +93,9 @@ void MatrixMul(XTensor * a, MATRIX_TRANS_TYPE transposedA,
XList * aList = new XList(10);
XList * bList = new XList(10);
XList * cList = new XList(10);
int aDimSize[2] = { -a->dimSize[0], a->dimSize[1] };
int bDimSize[2] = { -b->dimSize[0], b->dimSize[1] };
int cDimSize[2] = { -c->dimSize[0], c->dimSize[1] };
int aDimSize[2] = { a->dimSizeRDI[1], a->dimSizeRDI[0] };
int bDimSize[2] = { b->dimSizeRDI[1], b->dimSizeRDI[0] };
int cDimSize[2] = { c->dimSizeRDI[1], c->dimSizeRDI[0] };
bool isSparseMul = false;
......
source/core/MatrixMulBatched.cpp
......@@ -52,12 +52,12 @@ void MatrixMulBatched(XTensor * a, MATRIX_TRANS_TYPE transposedA,
CheckNTErrors((a->order >= 2 && b->order >= 2 && c->order >= 2),
"Input tensors must have a order > 2!");
int an = transposedA == X_TRANS ? a->dimSize[1] : a->dimSize[0];
int am = transposedA == X_TRANS ? a->dimSize[0] : a->dimSize[1];
int bn = transposedB == X_TRANS ? b->dimSize[1] : b->dimSize[0];
int bm = transposedB == X_TRANS ? b->dimSize[0] : b->dimSize[1];
int cn = c->dimSize[0];
int cm = c->dimSize[1];
int an = transposedA == X_TRANS ? a->dimSizeRDI[0] : a->dimSizeRDI[1];
int am = transposedA == X_TRANS ? a->dimSizeRDI[1] : a->dimSizeRDI[0];
int bn = transposedB == X_TRANS ? b->dimSizeRDI[0] : b->dimSizeRDI[1];
int bm = transposedB == X_TRANS ? b->dimSizeRDI[1] : b->dimSizeRDI[0];
int cn = c->dimSizeRDI[1];
int cm = c->dimSizeRDI[0];
CheckNTErrors((am == bn && an == cn && bm == cm),
"Unmatched tensors in multiplication!");
......@@ -79,9 +79,9 @@ void MatrixMulBatched(XTensor * a, MATRIX_TRANS_TYPE transposedA,
XList * aList = new XList(10);
XList * bList = new XList(10);
XList * cList = new XList(10);
int aDimSize[2] = { -a->dimSizeRDI[0], a->dimSizeRDI[1] };
int bDimSize[2] = { -b->dimSizeRDI[0], b->dimSizeRDI[1] };
int cDimSize[2] = { -c->dimSizeRDI[0], c->dimSizeRDI[1] };
int aDimSize[2] = { -a->dimSizeRDI[1], a->dimSizeRDI[0] };
int bDimSize[2] = { -b->dimSizeRDI[1], b->dimSizeRDI[0] };
int cDimSize[2] = { -c->dimSizeRDI[1], c->dimSizeRDI[0] };
for (int p = 0; p < blockNum; p++) {
void * ap = (char*)a->data + aRealBlockSize * p;
......@@ -106,8 +106,9 @@ void MatrixMulBatched(XTensor * a, MATRIX_TRANS_TYPE transposedA,
int devIDBackup;
ProtectCudaDev(a->devID, devIDBackup);
CudaBLASMatrixMULList(a->mem != NULL ? a->mem->GetCublasHandle() : GDevs.GetCudaHandle(a->devID),
aList, transposedA,
cublasHandle_t * handle = a->mem != NULL ? a->mem->GetCublasHandle() : GDevs.GetCudaHandle(a->devID);
CudaBLASMatrixMULList(handle,
aList, transposedA,
bList, transposedB,
cList, aList->count,
alpha, beta);
......
source/core/Select.cpp
......@@ -47,23 +47,28 @@ void SelectRange(XTensor * a, int dim, int low, int high, XTensor * c)
for(int i = 0; i < a->order; i++){
if(i == dim){
CheckNTErrors(low > 0 && low < a->dimSize[dim], "Illegal range specified!");
CheckNTErrors(high > 0 && high < a->dimSize[dim], "Illegal range specified!");
CheckNTErrors(high > 0 && high <= a->dimSize[dim], "Illegal range specified!");
}
else{
CheckNTErrors(a->dimSize[i] == c->dimSize[i], "The size of the dimensions should be same!");
}
}
int dimRDI = a->order - dim - 1;
int stride = 1;
for(int i = 0; i < dim; i++)
for(int i = 0; i < dimRDI; i++)
stride *= a->dimSizeRDI[i];
int copyTimes = 1;
for (int i = dimRDI + 1; i < a->order; i++)
copyTimes *= a->dimSizeRDI[i];
int blockSize = stride * (high - low) * a->unitSize;
int stepSizeS = stride * a->dimSize[dim] * a->unitSize;
int stepSizeT = stride * c->dimSize[dim] * a->unitSize;
char * s = (char*)a->data + stride * low * a->unitSize;
char * t = (char*)c->data;
for(int i = 0; i < high - low; i++){
for(int i = 0; i < copyTimes; i++){
XMemCopy(t, c->devID, s, a->devID, blockSize);
s += stepSizeS;
t += stepSizeT;
......
source/core/Sort.cu
......@@ -235,10 +235,6 @@ void CudaSortBig(XTensor * a, XTensor * b, XTensor * indexA, XTensor * indexB, i
int m = GetNextPower2(strideNum);
int n = stride * blockNum;
/* recheck */
/*void * buf = mem->AllocBuf(mem->devID, n * m * a->unitSize);
void * bufIndex = (indexA != NULL && indexB != NULL) ? mem->AllocBuf(mem->devID, n * m * sizeof(int)) : NULL;*/
/* change by liyinqiao */
void * buf = mem != NULL ? mem->AllocBuf(a->devID, n * m * a->unitSize) : XMemAlloc(a->devID, n * m * a->unitSize);
void * bufIndex = NULL;
if (indexA != NULL && indexB != NULL) {
......@@ -294,11 +290,6 @@ void CudaSortBig(XTensor * a, XTensor * b, XTensor * indexA, XTensor * indexB, i
KernelReorganizeBack<int> << <dim3(cudaGrids[1], cudaGrids[0]), dim3(cudaBlocks[1], cudaBlocks[0]) >> >
(bufIndex, indexB->data, m, n, stride, k, blockNum);
/* recheck */
/*mem->ReleaseBuf(mem->devID, n * m * a->unitSize);
if (indexA != NULL && indexB != NULL)
mem->ReleaseBuf(mem->devID, n * m * sizeof(int));*/
/* change by liyinqiao */
if (mem != NULL)
mem->ReleaseBuf(a->devID, n * m * a->unitSize);
else
......
source/core/TopK.cu
......@@ -20,6 +20,7 @@
*/
#include "../XDevice.h"
#include "../XUtility.h"
#include "../XTensor.h"
#include "TopK.h"
#include "TopK.cuh"
......@@ -393,7 +394,7 @@ void CudaTopK(XTensor * a, XTensor * b, XTensor * index, int dim, int k)
int cudaGrids[3];
int cudaBlocks[3];
GDevs.GetCudaThread2D(a->mem->devID,
GDevs.GetCudaThread2D(a->devID,
workerNum, stride * blockNum, MAX_INT,
cudaGrids, cudaBlocks);
......@@ -434,14 +435,17 @@ void CudaTopK(XTensor * a, XTensor * b, XTensor * index, int dim, int k)
memcpy(dimSize, a->dimSize, sizeof(int) * a->order);
dimSize[0] = -dimSize[0];
XTensor * indexA = new XTensor(a->order, dimSize, X_INT, 1.0F, a->mem);
indexA->data = a->mem->AllocBuf(a->devID, a->unitNum * sizeof(int));
indexA->data = a->mem != NULL ? a->mem->AllocBuf(a->devID, a->unitNum * sizeof(int)) : XMemAlloc(a->devID, a->unitNum * sizeof(int));
/* make the index tensor */
indexA->SetAscendingOrder(dim);
CudaSortBig(a, b, indexA, index, dim, k);
a->mem->ReleaseBuf(a->devID, a->unitNum * sizeof(int));
if (a->mem != NULL)
a->mem->ReleaseBuf(a->devID, a->unitNum * sizeof(int));
else
XMemFree(a->devID, indexA->data);
delete indexA;
}
......
source/function/Loss.cpp
......@@ -374,15 +374,15 @@ void LossBackward(XTensor * dedy, XTensor * t, XTensor * y,
LOSS_FUNCTION_NAME LFName,
int leadDim, int tBeg, int tLen, int yBeg)
{
CheckNTErrors((tLen >= 0 && tLen < y->unitNum), "Illegal input length!");
CheckNTErrors((tLen < y->unitNum), "Illegal input length!");
CheckNTErrors((XTensor::IsIdentical(t, y)&& XTensor::IsIdentical(dedy, y)),
"The input tensors must be of the same size!");
CheckNTErrors((t->dimSizeRDI[0] == 1 && y->dimSizeRDI[0] == 1 && dedy->dimSizeRDI[1] == 1), "TODO!");
//CheckNTErrors((t->dimSizeRDI[0] == 1 && y->dimSizeRDI[0] == 1 && dedy->dimSizeRDI[0] == 1), "TODO!");
CheckNTErrors((t->order > leadDim && leadDim >= 0), "Illegal leading dimension!");
CheckNTErrors((t->dataType == DEFAULT_DTYPE && y->dataType == DEFAULT_DTYPE),
"TODO!");
int leadDimRDI = y->order - leadDim - 1;
int leadDimRDI = leadDim >= 0 ? y->order - leadDim - 1 : -1;
if(leadDimRDI < 0){
leadDimRDI = y->dimSizeRDI[y->order - 1];
tBeg = 0;
......
source/test/Test.cpp
......@@ -31,6 +31,7 @@ bool Test()
wrong = !TestConcatenate() || wrong;
wrong = !TestConcatenateSolely() || wrong;
wrong = !TestCopyValues() || wrong;
wrong = !TestMatrixMul() || wrong;
wrong = !TestMatrixMul2D() || wrong;
wrong = !TestMatrixMulBatchedCPU() || wrong;
......@@ -42,12 +43,19 @@ bool Test()
wrong = !TestReduceMax() || wrong;
wrong = !TestReduceMean() || wrong;
wrong = !TestReduceSum() || wrong;
wrong = !TestReduceSumSquared() || wrong;
wrong = !TestReduceVariance() || wrong;
wrong = !TestScaleAndShift() || wrong;
wrong = !TestSelect() || wrong;
wrong = !TestSort() || wrong;
wrong = !TestSplit() || wrong;
wrong = !TestSum() || wrong;
wrong = !TestTopK() || wrong;
wrong = !TestUnsqueeze() || wrong;
wrong = !TestXMem() || wrong;
//wrong = !TestHardTanH() || wrong;
wrong = !TestIdentity || wrong;
//wrong = !TestLoss() || wrong;
//wrong = !TestRectify() || wrong;
wrong = !TestSigmoid() || wrong;
......
source/test/Test.h
......@@ -24,6 +24,7 @@
#include "TConcatenate.h"
#include "TConcatenateSolely.h"
#include "TCopyValues.h"
#include "TMatrixMul.h"
#include "TMatrixMul2D.h"
#include "TMatrixMULBatchedCPU.h"
......@@ -35,12 +36,19 @@
#include "TReduceMax.h"
#include "TReduceMean.h"
#include "TReduceSum.h"
#include "TReduceSumSquared.h"
#include "TReduceVariance.h"
#include "TScaleAndShift.h"
#include "TSelect.h"
#include "TSort.h"
#include "TSplit.h"
#include "TSum.h"
#include "TTopK.h"
#include "TUnsqueeze.h"
#include "TXMem.h"
#include "THardTanH.h"
#include "TIdentity.h"
#include "TLoss.h"
#include "TRectify.h"
#include "TSigmoid.h"
......
Markdown 格式
0% 或
您添加了 0 人 到此讨论。请谨慎行事。
请先完成此评论的编辑！
请 注册 或者 后发表评论