int main(int argc, char* argv[]) {
LogisticRegression lr;
LogisticRegressionScoreTest lrst;
LogisticRegressionPermutationTest lrpt;
Vector y;
Matrix x;
Matrix cov;
LoadVector("input.y", y);
LoadMatrix("input.x", x);
LoadMatrix("input.cov", cov);
Matrix xall;
xall = x;
xall.StackRight(cov); // 1 + x + cov
if (lr.FitLogisticModel(xall, y, 100) == false) {
fprintf(stderr, "Fitting failed!\n");
return -1;
}
Vector& beta = lr.GetCovEst();
Matrix& v = lr.GetCovB();
Vector& pWald = lr.GetAsyPvalue();
fprintf(stdout, "wald_beta\t");
Print(beta);
fputc('\n', stdout);
fprintf(stdout, "wald_vcov\t");
Print(v);
fputc('\n', stdout);
fprintf(stdout, "wald_p\t");
Print(pWald[1]);
fputc('\n', stdout);
if (lrpt.FitLogisticModelCov(xall, 1, y, 2000, -1) == false) {
fprintf(stderr, "Fitting failed!\n");
return -1;
}
fprintf(stdout, "permutation_p\t");
double permu_p = lrpt.getPvalue();
Print(permu_p);
fputc('\n', stdout);
if (lrst.FitLogisticModel(xall, y, 1, 100) == false) {
fprintf(stderr, "Fitting failed!\n");
return -1;
}
fprintf(stdout, "score_p\t");
double score_p = lrst.GetPvalue();
Print(score_p);
fputc('\n', stdout);
return 0;
};
int main(int argc, char *argv[]) {
Vector Y;
Matrix X;
Matrix Cov;
LoadVector("input.linear.mvt.y", Y);
LoadMatrix("input.linear.mvt.x", X);
LoadMatrix("input.linear.mvt.cov", Cov);
{
Matrix x;
Vector y;
x = X;
y = Y;
LinearRegressionVT linear;
if (!linear.FitNullModel(Cov, Y)) {
fprintf(stderr, "Fitting failed! - step 1!\n");
return -1;
}
if (!linear.TestCovariate(Cov, Y, X)) {
fprintf(stderr, "Fitting failed - step 2!\n");
return -1;
}
dumpToFile(linear.GetU(), stdout);
dumpToFile(linear.GetV(), stdout);
dumpToFile(linear.GetT(), stdout);
dumpToFile(linear.GetCov(), stdout);
fprintf(stdout, "%g\t0\n", linear.GetPvalue(), 0);
}
return 0;
};
/*
* Draws the plant.
*/
void drawPlant(int i,float scale,float radscale,int string)
{
char *ptr = lsystem[string];
char ch[] = { '\0','\0' };
GLfloat c1[] = { 0.6549f,0.4901f,0.2392f }; /* light brown */
GLfloat c2[] = { 0.3607f,0.2510f,0.2000f }; /* dark brown */
GLfloat c3[] = { 0.1373f,0.5568f,0.1373f }; /* forest green */
if(i==0)
return;
PushMatrix();
while(*ptr != '\0')
{
switch(*ptr)
{
case 'F':
Rotate(tilt/10000,0.0,0.0,1.0); /* tilt very very slightly */
LoadMatrix();
if(do_growth && floor(growth)==i)
{
draw_branch(LENGTH*scale*(growth-1),RADIUS*radscale,BASE_TRI,c1,c2);
Translate(0.0,LENGTH*scale*(growth-1),0.0);
}
else
{
draw_branch(LENGTH*scale,RADIUS*radscale,BASE_TRI,c1,c2);
Translate(0.0,LENGTH*scale,0.0);
}
break;
case '[':
PushMatrix();
break;
case ']':
PopMatrix();
break;
case 'L':
if(do_growth && floor(growth)==i)
draw_leaf(4*(growth-1)*scale,1*(growth-1)*scale,25,c3);
else
draw_leaf(4*scale,1*scale,25,c3);
break;
case 'R':
Rotate(yrotate,0.0,1.0,0.0);
break;
case 'T':
Rotate(tilt,0.0,0.0,1.0);
break;
default:
if(isdigit(*ptr))
{
ch[0] = *ptr;
drawPlant(i-1,scale*SCALE,radscale*RADSCALE,atoi(ch));
}
break;
}
ptr++;
}
PopMatrix();
}
AutoEncoder::AutoEncoder(std::string add,
const int rows, const int cols,UINT hiddenUnits,
double sparsity):
IO_dim(cols), hidden_units(hiddenUnits), samples(rows), st_sparsity(sparsity)
{
/*
* Description:
* A constructed function to initialize the Sparse AutoEncoder
*
* @param rows: number of samples
* @param cols: dimension of samples
* @param hiddenUnits: number of hidden units
* @param sparsity: standard sparsity we want to approximate
*
*/
LoadMatrix(*OriginalData, add, cols, rows);
HiddenMatrix = MatrixXf::Zero(samples, hidden_units);
Sparsity_average = VectorXf::Zero(hidden_units);
OutputMatrix = MatrixXf::Zero(samples, IO_dim);
/* Initialize the en/decode weight matrix */
double w_init_interval = sqrt(6./(IO_dim + hidden_units + 1));
Weight_encode = w_init_interval * MatrixXf::Random(hidden_units, IO_dim);
Weight_decode = w_init_interval * MatrixXf::Random(IO_dim, hidden_units);
Bias_encode = VectorXf::Ones(hidden_units);
Bias_decode = VectorXf::Ones(IO_dim);
}
Animation* AnimationLoader::LoadAnimation(BinaryReader &br)
{
Animation *anim = new Animation();
bool hasOwnAnim = false;
float angleScale;
anim ->id = br.Read<int>();
anim ->worldTMInv = LoadMatrix(br);
anim ->pos = LoadVec3Anim(br, anim ->localPos);
anim ->rot = LoadQuatAnim(br, anim ->localRot, hasOwnAnim, angleScale);
anim ->scale = LoadVec3Anim(br, anim ->localScale);
if (hasOwnAnim)
{
anim->hasOwnRotate = hasOwnAnim;
anim->angleScale = angleScale;
}
int subAnimsCount = br.Read<int>();
for (int i = 0; i < subAnimsCount; i++)
anim ->subAnims.push_back(LoadAnimation(br));
return anim;
}
void DrawNode(R3Scene *scene, R3Node *node)
{
// Push transformation onto stack
glPushMatrix();
LoadMatrix(&node->transformation);
// Load material
if (node->material) LoadMaterial(node->material);
// Draw shape
if (node->shape) DrawShape(node->shape);
// Draw children nodes
for (int i = 0; i < (int) node->children.size(); i++)
DrawNode(scene, node->children[i]);
// Restore previous transformation
glPopMatrix();
// Show bounding box
if (show_bboxes) {
GLboolean lighting = glIsEnabled(GL_LIGHTING);
glDisable(GL_LIGHTING);
node->bbox.Outline();
if (lighting) glEnable(GL_LIGHTING);
}
}
void nuiGLPainter::MultMatrix(const nuiMatrix& rMatrix)
{
NUI_RETURN_IF_RENDERING_DISABLED;
nuiPainter::MultMatrix(rMatrix);
LoadMatrix(mMatrixStack.top());
nuiCheckForGLErrors();
}
void nuiGLPainter::PopMatrix()
{
NUI_RETURN_IF_RENDERING_DISABLED;
nuiPainter::PopMatrix();
LoadMatrix(mMatrixStack.top());
nuiCheckForGLErrors();
}
double EvalDeterminant( char *FileName )
{
int i;
LoadMatrix(FileName);
sum = 0;
for (i = 0; i < N; i++)
P[i] = i;
Go(0);
return sum;
}
int main(int argc, char* argv[]) {
LinearRegression lr;
LinearRegressionScoreTest lrst;
LinearRegressionPermutationTest lrpt;
Vector y;
Matrix x;
LoadVector("input.y", y);
LoadMatrix("input.x", x);
if (lr.FitLinearModel(x, y) == false) {
fprintf(stderr, "Fitting failed!\n");
return -1;
}
Vector& beta = lr.GetCovEst();
Matrix& v = lr.GetCovB();
Vector& pWald = lr.GetAsyPvalue();
fprintf(stdout, "wald_beta\t");
Print(beta);
fputc('\n', stdout);
fprintf(stdout, "wald_vcov\t");
Print(v);
fputc('\n', stdout);
fprintf(stdout, "wald_p\t");
Print(pWald[1]);
fputc('\n', stdout);
if (lrpt.FitLinearModel(x, 1, y, 200, 0.05) == false) {
fprintf(stderr, "Fitting failed!\n");
return -1;
}
fprintf(stdout, "permutation_p\t");
double permu_p = lrpt.getPvalue();
Print(permu_p);
fputc('\n', stdout);
if (lrst.FitLinearModel(x, y, 1) == false) {
fprintf(stderr, "Fitting failed!\n");
return -1;
}
fprintf(stdout, "score_p\t");
double score_p = lrst.GetPvalue();
Print(score_p);
fputc('\n', stdout);
return 0;
};
void gatherAEff(const char *inpFile=NULL) {
std::string line;
std::vector<std::string> lines;
if (!inpFile) {
std::cout << "\ninpFile=NULL\n";
std::cout << "Creating sample input file\n";
const char *sampleFName="aeff_sample.inp";
std::ofstream fout(sampleFName);
fout << "../root_files/constants/DY_j22_19789pb/acceptance_constants1D.root\n";
fout << "../root_files/constants/DY_j22_19789pb/event_efficiency_constants1D.root\n";
fout << "../root_files/constants/DY_j22_19789pb/scale_factors_1D_Full2012_hltEffOld_PU.root\n";
fout.close();
std::cout << "check the file <" << sampleFName << ">\n";
std::cout << "1D/2D in the name does not matter\n";
return;
}
else { // load input
std::ifstream finput(inpFile);
if (!finput) {
std::cout << "failed to open file <" << inpFile << ">\n";
return;
}
getline(finput,line); lines.push_back(line);
std::cout << "line1=<" << line << ">\n";
getline(finput,line); lines.push_back(line);
std::cout << "line2=<" << line << ">\n";
getline(finput,line); lines.push_back(line);
std::cout << "line3=<" << line << ">\n";
finput.close();
}
std::cout << dashline;
std::string fnameAcc,fnameEff,fnameRho;
int count=0;
for (unsigned int i=0; i<lines.size(); ++i) {
line=lines[i];
AdjustDim(line);
if (PosOk(line,"acceptance")) { fnameAcc=line; count++; }
else if (PosOk(line,"efficiency")) { fnameEff=line; count++; }
else if (PosOk(line,"scale_factors")) { fnameRho=line; count++; }
else {
std::cout << "could not identify the file <" << line << ">\n";
return;
}
}
if (count!=3) {
std::cout << "not all files were identified\n";
return;
}
std::cout << "files were identified ok\n";
std::cout << dashline;
TMatrixD effM(DYTools::nMassBins,DYTools::nYBinsMax);
TMatrixD effErrM(effM);
TMatrixD accM(effM), accErrM(effM);
TMatrixD rhoM(effM), rhoErrM(effM);
if (!LoadMatrix(fnameAcc,accM,accErrM,"acceptanceMatrix","acceptanceErrMatrix") ||
!LoadMatrix(fnameEff,effM,effErrM,"efficiencyArray","efficiencyErrArray") ||
!LoadMatrix(fnameRho,rhoM,rhoErrM,"scaleFactor","scaleFactorErr") ) {
std::cout << "failed to load field\n";
return;
}
TString outFName;
if (DYTools::study2D) {
outFName="dyee_aeff_2D.root";
TFile fout(outFName,"recreate");
accM.Write("acceptance");
accErrM.Write("acceptanceErr");
effM.Write("efficiency");
effErrM.Write("efficiencyErr");
rhoM.Write("scaleFactor");
rhoErrM.Write("scaleFactorErr");
unfolding::writeBinningArrays(fout);
for (int i=0; i<DYTools::nMassBins; ++i) {
TString massRange=Form("_%1.0lf_%2.0lf",DYTools::massBinLimits[i],DYTools::massBinLimits[i+1]);
TString hAccName=TString("hAcc") + massRange;
Histo_t *hAcc=extractRapidityDependence(hAccName,"",accM,accErrM,i,0);
TString hEffName=TString("hEff") + massRange;
Histo_t *hEff=extractRapidityDependence(hEffName,"",effM,effErrM,i,0);
TString hRhoName=TString("hRho") + massRange;
Histo_t *hRho=extractRapidityDependence(hRhoName,"",rhoM,rhoErrM,i,0);
if (!hAcc || !hEff || !hRho) {
std::cout << "got unexpected null histo\n";
break;
}
hAcc->Write();
hEff->Write();
hRho->Write();
}
fout.Close();
}
else { // 1D case
outFName="dyee_aeff_1D.root";
std::cout << "accM: rows " << accM.GetNrows() << ", cols " << accM.GetNcols() << "\n";
TVectorD eff(DYTools::nMassBins), effErr(eff);
TVectorD acc(eff), accErr(eff);
TVectorD rho(eff), rhoErr(eff);
GetMassProfile1D(accM,accErrM, acc,accErr);
GetMassProfile1D(effM,effErrM, eff,effErr);
GetMassProfile1D(rhoM,rhoErrM, rho,rhoErr);
Histo_t *hAcc=extractMassDependence("hAcc","",accM,accErrM,0,0,0);
Histo_t *hEff=extractMassDependence("hEff","",effM,effErrM,0,0,0);
Histo_t *hRho=extractMassDependence("hRho","",rhoM,rhoErrM,0,0,0);
if (!hAcc || !hEff || !hRho) {
std::cout << "got unexpected null histo\n";
}
else {
TFile fout(outFName,"recreate");
acc.Write("acceptance");
accErr.Write("acceptanceErr");
eff.Write("efficiency");
effErr.Write("efficiencyErr");
rho.Write("scaleFactor");
rhoErr.Write("scaleFactorErr");
unfolding::writeBinningArrays(fout);
hAcc->Write();
hEff->Write();
hRho->Write();
fout.Close();
}
}
std::cout << "file <" << outFName << "> created\n";
return;
}
void nuiGLPainter::ApplyTexture(const nuiRenderState& rState, bool ForceApply)
{
// if ((rState.mTexturing && !rState.mpTexture) || (!rState.mTexturing && rState.mpTexture))
// {
// printf("bleh!\n");
// char* bleh = NULL;
// bleh[0] = 0;
// }
// 2D Textures:
std::map<nuiTexture*, TextureInfo>::const_iterator it = mTextures.find(rState.mpTexture);
bool uptodate = (it == mTextures.end()) ? false : ( !it->second.mReload && it->second.mTexture >= 0 );
if (ForceApply || (mState.mpTexture != rState.mpTexture) || (mState.mpTexture && !uptodate))
{
GLenum intarget = 0;
GLenum outtarget = 0;
if (mState.mpTexture)
{
outtarget = GetTextureTarget(mState.mpTexture->IsPowerOfTwo());
//mState.mpTexture->UnapplyGL(this); #TODO Un apply the texture
nuiCheckForGLErrors();
mState.mpTexture->Release();
nuiCheckForGLErrors();
}
//NGL_OUT(_T("Change texture to 0x%x (%ls)\n"), rState.mpTexture, rState.mpTexture?rState.mpTexture->GetSource().GetChars() : nglString::Empty.GetChars());
mState.mpTexture = rState.mpTexture ;
if (mState.mpTexture)
{
intarget = GetTextureTarget(mState.mpTexture->IsPowerOfTwo());
mState.mpTexture->Acquire();
nuiSurface* pSurface = mState.mpTexture->GetSurface();
if (pSurface)
{
std::map<nuiSurface*, FramebufferInfo>::const_iterator it = mFramebuffers.find(pSurface);
bool create = (it == mFramebuffers.end()) ? true : false;
if (create || pSurface->IsDirty())
{
PushClipping();
nuiRenderState s(mState);// PushState();
PushProjectionMatrix();
PushMatrix();
#ifdef _OPENGL_ES_
if (mpSurfaceStack.empty())
{
// mDefaultFramebuffer = 0;
// mDefaultRenderbuffer = 0;
glGetIntegerv(GL_FRAMEBUFFER_BINDING_NUI, &mDefaultFramebuffer);
glGetIntegerv(GL_RENDERBUFFER_BINDING_NUI, (GLint *) &mDefaultRenderbuffer);
}
#endif
PushSurface();
SetState(nuiRenderState());
ResetClipRect();
mClip.Set(0, 0, pSurface->GetWidth(), pSurface->GetHeight());
LoadMatrix(nglMatrixf());
NGL_ASSERT(pSurface);
SetSurface(pSurface);
//Set2DProjectionMatrix(nuiRect(0.0f, 0.0f, pSurface->GetWidth(), pSurface->GetHeight()));
nuiMatrix m;
m.Translate(-1.0f, 1.0f, 0.0f);
m.Scale(2.0f / pSurface->GetWidth(), -2.0f / pSurface->GetHeight(), 1.0f);
LoadProjectionMatrix(nuiRect(pSurface->GetWidth(), pSurface->GetHeight()), m);
// clear the surface with transparent black:
// nuiRenderState s2(mState);// PushState();
// mState.mClearColor = nuiColor(0.0f, 0.0f, 0.0f, 0.0f);
SetState(mState);
// ClearColor();
// SetState(s2);
//////////////////////////////
nuiDrawContext Ctx(nuiRect(pSurface->GetWidth(), pSurface->GetHeight()));
Ctx.SetPainter(this);
pSurface->Realize(&Ctx);
Ctx.SetPainter(NULL);
//////////////////////////////
PopSurface();
PopMatrix();
PopProjectionMatrix();
//PopState();
SetState(s);
PopClipping();
}
}
UploadTexture(mState.mpTexture);
nuiCheckForGLErrors();
}
//NGL_OUT(_T("Change texture type from 0x%x to 0x%x\n"), outtarget, intarget);
mTextureTarget = intarget;
if (intarget != outtarget)
{
// Texture Target has changed
if (outtarget)
{
glDisable(outtarget);
nuiCheckForGLErrors();
}
//NGL_OUT(_T("disable outtarget\n"));
if (intarget && mState.mTexturing && mState.mpTexture)
{
mState.mTexturing = rState.mTexturing;
//NGL_OUT(_T("enable intarget\n"));
glEnable(intarget);
nuiCheckForGLErrors();
}
}
else
{
// Texture Target have not changed
if (mState.mTexturing != rState.mTexturing) // Have texture on/off changed?
{
// Should enable or disable texturing
mState.mTexturing = rState.mTexturing;
if (mState.mTexturing)
{
glEnable(mTextureTarget);
nuiCheckForGLErrors();
}
else
{
glDisable(mTextureTarget);
nuiCheckForGLErrors();
}
}
}
}
if (ForceApply || (mState.mTexturing != rState.mTexturing))
{
// Texture have not changed, but texturing may have been enabled / disabled
mState.mTexturing = rState.mTexturing;
if (mState.mpTexture)
{
if (mTextureTarget && mState.mTexturing)
{
//NGL_OUT(_T("Enable 0x%x\n"), mTextureTarget);
glEnable(mTextureTarget);
nuiCheckForGLErrors();
}
else
{
//NGL_OUT(_T("Disable 0x%x\n"), mTextureTarget);
glDisable(mTextureTarget);
nuiCheckForGLErrors();
}
}
else
{
if (mTextureTarget)
{
//NGL_OUT(_T("Disable 0x%x\n"), mTextureTarget);
glDisable(mTextureTarget);
}
nuiCheckForGLErrors();
}
}
}
int main(int argc,char **argv){
int i,j,k = 0;
double MSE = 0,TrainingAccuracy;
float **input,**weight,*biase,**tranpI,**tempH,**H;
float **tranpw;
float **train_set;
float **T,**Y;
float **out;
train_set = (float **)calloc(DATASET,sizeof(float *));
tranpI = (float **)calloc(INPUT_NEURONS,sizeof(float *));
input = (float **)calloc(DATASET,sizeof(float *)); /*datasize * INPUT_NEURONS*/
weight = (float **)calloc(HIDDEN_NEURONS,sizeof(float *)); /*HIDDEN_NEURONS * INPUT_NEURONS*/
biase = (float *)calloc(HIDDEN_NEURONS,sizeof(float)); /*HIDDEN_NEURONS*/
tempH = (float **)calloc(HIDDEN_NEURONS,sizeof(float *)); /*HIDDEN_NEURONS * datasize*/
tranpw = (float **)calloc(INPUT_NEURONS,sizeof(float *));
H = (float **)calloc(DATASET,sizeof(float *));
T = (float **)calloc(DATASET,sizeof(float *));
Y = (float **)calloc(DATASET,sizeof(float *));
out = (float **)calloc(HIDDEN_NEURONS,sizeof(float *));
for(i=0;i<DATASET;i++){
train_set[i] = (float *)calloc(NUMROWS,sizeof(float));
input[i] = (float *)calloc(INPUT_NEURONS,sizeof(float));
H[i] = (float *)calloc(HIDDEN_NEURONS,sizeof(float));
}
for(i=0;i<DATASET;i++)
{
T[i] = (float *)calloc(OUTPUT_NEURONS,sizeof(float));
Y[i] = (float *)calloc(OUTPUT_NEURONS,sizeof(float));
}
for(i=0;i<INPUT_NEURONS;i++){
tranpI[i] = (float *)calloc(DATASET,sizeof(float));
tranpw[i] = (float *)calloc(HIDDEN_NEURONS,sizeof(float));
}
for(i=0;i<HIDDEN_NEURONS;i++)
{
weight[i] = (float *)calloc(INPUT_NEURONS,sizeof(float));
tempH[i] = (float *)calloc(DATASET,sizeof(float));
out[i] = (float *)calloc(OUTPUT_NEURONS,sizeof(float));
}
printf("begin to regression test...\n");
printf("begin to random weight and biase...\n");
/*得到随机的偏置和权重*/
//RandomWeight(weight,HIDDEN_NEURONS,INPUT_NEURONS);
//RandomBiase(biase,HIDDEN_NEURONS);
if(LoadMatrix(tranpw,"../result/weight",INPUT_NEURONS,HIDDEN_NEURONS) == 0){
printf("load input file error!!!\n");
return 0;
}
TranspositionMatrix(tranpw,weight,INPUT_NEURONS,HIDDEN_NEURONS);
if(LoadMatrix_s(biase,"../result/bias",1,HIDDEN_NEURONS) == 0){
printf("load input file error!!!\n");
return 0;
}
/*加载数据集到内存*/
printf("begin to load input from the file...\n");
if(LoadMatrix(train_set,"../sample/ppp",DATASET,NUMROWS) == 0){
printf("load input file error!!!\n");
return 0;
}
/*将数据集划分成输入和输出*/
for(i = 0;i < DATASET ;i++)
{
T[k++][0] = train_set[i][0];
for(j = 1;j <= INPUT_NEURONS;j++)
{
input[i][j-1] = train_set[i][j];
}
}
/*ELM*/
printf("begin to compute...\n");
TranspositionMatrix(input,tranpI,DATASET,INPUT_NEURONS);
printf("begin to compute step 1...\n");
MultiplyMatrix(weight,HIDDEN_NEURONS,INPUT_NEURONS, tranpI,INPUT_NEURONS,DATASET,tempH);
printf("begin to compute setp 2...\n");
AddMatrix_bais(tempH,biase,HIDDEN_NEURONS,DATASET);
printf("begin to compute step 3...\n");
SigmoidHandle(tempH,HIDDEN_NEURONS,DATASET);
printf("begin to compute step 4...\n");
TranspositionMatrix(tempH,H,HIDDEN_NEURONS,DATASET);
printf("begin to load input from the file...\n");
if(LoadMatrix(out,"../result/result",HIDDEN_NEURONS,OUTPUT_NEURONS) == 0){
printf("load input file error!!!\n");
return 0;
}
//检测准确率
MultiplyMatrix(H,DATASET,HIDDEN_NEURONS,out,HIDDEN_NEURONS,OUTPUT_NEURONS,Y);
for(i = 0;i< DATASET;i++)
{
MSE += (Y[i][0] - T[i][0])*(Y[i][0] - T[i][0]);
}
TrainingAccuracy = sqrt(MSE/DATASET);
printf("trainning accuracy :%f\n",TrainingAccuracy);
printf("test complete...\n");
//print(PIMatrix,DATASET,HIDDEN_NEURONS);
return 0;
}
/**
* Construct an ELM
* @param
* elm_type - 0 for regression; 1 for (both binary and multi-classes) classification
*/
void ELMTrain(int elm_type)
{
double starttime,endtime;
double TrainingAccuracy;
int i,j,k = 0;
float **input,**weight,*biase,**tranpI,**tempH,**H;
float **PIMatrix;
float **train_set;
float **T,**Y;
float **out;
train_set = (float **)calloc(DATASET,sizeof(float *));
tranpI = (float **)calloc(INPUT_NEURONS,sizeof(float *));
input = (float **)calloc(DATASET,sizeof(float *)); /*datasize * INPUT_NEURONS*/
weight = (float **)calloc(HIDDEN_NEURONS,sizeof(float *)); /*HIDDEN_NEURONS * INPUT_NEURONS*/
biase = (float *)calloc(HIDDEN_NEURONS,sizeof(float)); /*HIDDEN_NEURONS*/
tempH = (float **)calloc(HIDDEN_NEURONS,sizeof(float *)); /*HIDDEN_NEURONS * datasize*/
PIMatrix = (float **)calloc(HIDDEN_NEURONS,sizeof(float *));
H = (float **)calloc(DATASET,sizeof(float *));
T = (float **)calloc(DATASET,sizeof(float *));
Y = (float **)calloc(DATASET,sizeof(float *));
out = (float **)calloc(HIDDEN_NEURONS,sizeof(float *));
for(i=0;i<DATASET;i++){
train_set[i] = (float *)calloc(NUMROWS,sizeof(float));
input[i] = (float *)calloc(INPUT_NEURONS,sizeof(float));
H[i] = (float *)calloc(HIDDEN_NEURONS,sizeof(float));
}
for(i=0;i<DATASET;i++)
{
T[i] = (float *)calloc(OUTPUT_NEURONS,sizeof(float));
Y[i] = (float *)calloc(OUTPUT_NEURONS,sizeof(float));
}
for(i=0;i<INPUT_NEURONS;i++)
tranpI[i] = (float *)calloc(DATASET,sizeof(float));
for(i=0;i<HIDDEN_NEURONS;i++)
{
weight[i] = (float *)calloc(INPUT_NEURONS,sizeof(float));
tempH[i] = (float *)calloc(DATASET,sizeof(float));
out[i] = (float *)calloc(OUTPUT_NEURONS,sizeof(float));
PIMatrix[i] = (float *)calloc(DATASET,sizeof(float));
}
printf("begin to random weight and biase...\n");
/*得到随机的偏置和权重*/
RandomWeight(weight,HIDDEN_NEURONS,INPUT_NEURONS);
RandomBiase(biase,HIDDEN_NEURONS);
SaveMatrix(weight,"./result/weight",HIDDEN_NEURONS,INPUT_NEURONS);
SaveMatrix_s(biase,"./result/biase",1,HIDDEN_NEURONS);
/*加载数据集到内存*/
printf("begin to load input from the file...\n");
if(LoadMatrix(train_set,"./sample/frieman",DATASET,NUMROWS) == 0){
printf("load input file error!!!\n");
return;
}
/*将数据集划分成输入和输出*/
if(elm_type == 0){ //regression
/*
the first column is the output of the dataset
*/
for(i = 0;i < DATASET ;i++)
{
T[k++][0] = train_set[i][0];
for(j = 1;j <= INPUT_NEURONS;j++)
{
input[i][j-1] = train_set[i][j];
}
}
}else{ //classification
/*
the last column is the lable of class of the dataset
*/
InitMatrix(T,DATASET,OUTPUT_NEURONS,-1);
for(i = 0;i < DATASET ;i++)
{
//get the last column
T[k++][0] = train_set[i][0] - 1;//class label starts from 0,so minus one
for(j = 1;j <= INPUT_NEURONS;j++)
{
input[i][j-1] = train_set[i][j];
}
}
for(i = 0;i < DATASET ;i++)
{
for(j = 0;j < OUTPUT_NEURONS;j++)
{
k = T[i][0];
if(k < OUTPUT_NEURONS && k >= 0){
T[i][k] = 1;
}
if(k != 0){
T[i][0] = -1;
}
}
}
}
/*ELM*/
printf("begin to compute...\n");
starttime = omp_get_wtime();
TranspositionMatrix(input,tranpI,DATASET,INPUT_NEURONS);
printf("begin to compute step 1...\n");
MultiplyMatrix(weight,HIDDEN_NEURONS,INPUT_NEURONS, tranpI,INPUT_NEURONS,DATASET,tempH);
printf("begin to compute setp 2...\n");
AddMatrix_bais(tempH,biase,HIDDEN_NEURONS,DATASET);
printf("begin to compute step 3...\n");
SigmoidHandle(tempH,HIDDEN_NEURONS,DATASET);
printf("begin to compute step 4...\n");
TranspositionMatrix(tempH,H,HIDDEN_NEURONS,DATASET);
PseudoInverseMatrix(H,DATASET,HIDDEN_NEURONS,PIMatrix);
MultiplyMatrix(PIMatrix,HIDDEN_NEURONS,DATASET,T,DATASET,OUTPUT_NEURONS,out);
//SaveMatrix(H,"./result/H",DATASET,HIDDEN_NEURONS);
//SaveMatrix(PIMatrix,"./result/PIMatrix",HIDDEN_NEURONS,DATASET);
printf("begin to compute step 5...\n");
endtime = omp_get_wtime();
//保存输出权值
SaveMatrix(out,"./result/result",HIDDEN_NEURONS,OUTPUT_NEURONS);
MultiplyMatrix(H,DATASET,HIDDEN_NEURONS,out,HIDDEN_NEURONS,OUTPUT_NEURONS,Y);
printf("use time :%f\n",endtime - starttime);
printf("train complete...\n");
if(elm_type == 0){
//检测准确率
double MSE = 0;
for(i = 0;i< DATASET;i++)
{
MSE += (Y[i][0] - T[i][0])*(Y[i][0] - T[i][0]);
}
TrainingAccuracy = sqrt(MSE/DATASET);
printf("Regression/trainning accuracy :%f\n",TrainingAccuracy);
}else{
float MissClassificationRate_Training=0;
double maxtag1,maxtag2;
int tag1 = 0,tag2 = 0;
for (i = 0; i < DATASET; i++) {
maxtag1 = Y[i][0];
tag1 = 0;
maxtag2 = T[i][0];
tag2 = 0;
for (j = 1; j < OUTPUT_NEURONS; j++) {
if(Y[i][j] > maxtag1){
maxtag1 = Y[i][j];
tag1 = j;
}
if(T[i][j] > maxtag2){
maxtag2 = T[i][j];
tag2 = j;
}
}
if(tag1 != tag2)
MissClassificationRate_Training ++;
}
TrainingAccuracy = 1 - MissClassificationRate_Training*1.0f/DATASET;
printf("Classification/training accuracy :%f\n",TrainingAccuracy);
}
FreeMatrix(train_set,DATASET);
FreeMatrix(tranpI,INPUT_NEURONS);
FreeMatrix(input,DATASET);
FreeMatrix(weight,HIDDEN_NEURONS);
free(biase);
FreeMatrix(tempH,HIDDEN_NEURONS);
FreeMatrix(PIMatrix,HIDDEN_NEURONS);
FreeMatrix(H,DATASET);
FreeMatrix(T,DATASET);
FreeMatrix(Y,DATASET);
FreeMatrix(out,HIDDEN_NEURONS);
}
int main(int argc,char **argv){
int i,j,k = 0;
double TrainingAccuracy;
float **input,**weight,*biase,**tranpI,**tempH,**H;
float **train_set;
float **T,**Y;
float **out;
train_set = (float **)calloc(TESTSET,sizeof(float *));
tranpI = (float **)calloc(INPUT_NEURONS,sizeof(float *));
input = (float **)calloc(TESTSET,sizeof(float *)); /*datasize * INPUT_NEURONS*/
weight = (float **)calloc(HIDDEN_NEURONS,sizeof(float *)); /*HIDDEN_NEURONS * INPUT_NEURONS*/
biase = (float *)calloc(HIDDEN_NEURONS,sizeof(float)); /*HIDDEN_NEURONS*/
tempH = (float **)calloc(HIDDEN_NEURONS,sizeof(float *)); /*HIDDEN_NEURONS * datasize*/
H = (float **)calloc(TESTSET,sizeof(float *));
T = (float **)calloc(TESTSET,sizeof(float *));
Y = (float **)calloc(TESTSET,sizeof(float *));
out = (float **)calloc(HIDDEN_NEURONS,sizeof(float *));
for(i=0;i<TESTSET;i++){
train_set[i] = (float *)calloc(NUMROWS,sizeof(float));
input[i] = (float *)calloc(INPUT_NEURONS,sizeof(float));
H[i] = (float *)calloc(HIDDEN_NEURONS,sizeof(float));
}
for(i=0;i<TESTSET;i++)
{
T[i] = (float *)calloc(OUTPUT_NEURONS,sizeof(float));
Y[i] = (float *)calloc(OUTPUT_NEURONS,sizeof(float));
}
for(i=0;i<INPUT_NEURONS;i++){
tranpI[i] = (float *)calloc(TESTSET,sizeof(float));
}
for(i=0;i<HIDDEN_NEURONS;i++)
{
weight[i] = (float *)calloc(INPUT_NEURONS,sizeof(float));
tempH[i] = (float *)calloc(TESTSET,sizeof(float));
out[i] = (float *)calloc(OUTPUT_NEURONS,sizeof(float));
}
printf("begin to classification test...\n");
/*得到随机的偏置和权重*/
//RandomWeight(weight,HIDDEN_NEURONS,INPUT_NEURONS);
//RandomBiase(biase,HIDDEN_NEURONS);
printf("begin to load weight...\n");
if(LoadMatrix(weight,"../result/weight",HIDDEN_NEURONS,INPUT_NEURONS) == 0){
printf("load weight file error!!!\n");
return 0;
}
printf("begin to load bias...\n");
if(LoadMatrix_s(biase,"../result/biase",1,HIDDEN_NEURONS) == 0){
printf("load bias file error!!!\n");
return 0;
}
/*加载数据集到内存*/
printf("begin to load test file...\n");
if(LoadMatrix(train_set,"../sample/covtype_test",TESTSET,NUMROWS) == 0){
printf("load input file error!!!\n");
return 0;
}
InitMatrix(T,TESTSET,OUTPUT_NEURONS,-1);
/*将数据集划分成输入和输出*/
for(i = 0;i < TESTSET ;i++)
{
//get the last column
T[k++][0] = train_set[i][0] - 1;//class label starts from 0,so minus one
for(j = 1;j <= INPUT_NEURONS;j++)
{
input[i][j-1] = train_set[i][j];
}
}
for(i = 0;i < TESTSET ;i++)
{
for(j = 0;j < OUTPUT_NEURONS;j++)
{
k = T[i][0];
if(k < OUTPUT_NEURONS && k >= 0){
T[i][k] = 1;
}
if(k != 0){
T[i][0] = -1;
}
}
}
/*ELM*/
printf("begin to compute...\n");
TranspositionMatrix(input,tranpI,TESTSET,INPUT_NEURONS);
printf("begin to compute step 1...\n");
MultiplyMatrix(weight,HIDDEN_NEURONS,INPUT_NEURONS, tranpI,INPUT_NEURONS,TESTSET,tempH);
printf("begin to compute setp 2...\n");
AddMatrix_bais(tempH,biase,HIDDEN_NEURONS,TESTSET);
printf("begin to compute step 3...\n");
SigmoidHandle(tempH,HIDDEN_NEURONS,TESTSET);
printf("begin to compute step 4...\n");
TranspositionMatrix(tempH,H,HIDDEN_NEURONS,TESTSET);
printf("begin to load hidden output matrix from the file...\n");
if(LoadMatrix(out,"../result/result",HIDDEN_NEURONS,OUTPUT_NEURONS) == 0){
printf("load input file error!!!\n");
return 0;
}
MultiplyMatrix(H,TESTSET,HIDDEN_NEURONS,out,HIDDEN_NEURONS,OUTPUT_NEURONS,Y);
float MissClassificationRate_Training=0;
double maxtag1,maxtag2;
int tag1 = 0,tag2 = 0;
for (i = 0; i < TESTSET; i++) {
maxtag1 = Y[i][0];
tag1 = 0;
maxtag2 = T[i][0];
tag2 = 0;
for (j = 1; j < OUTPUT_NEURONS; j++) {
if(Y[i][j] > maxtag1){
maxtag1 = Y[i][j];
tag1 = j;
}
if(T[i][j] > maxtag2){
maxtag2 = T[i][j];
tag2 = j;
}
}
if(tag1 != tag2)
MissClassificationRate_Training ++;
}
TrainingAccuracy = 1 - MissClassificationRate_Training*1.0f/TESTSET;
printf("trainning accuracy :%f\n",TrainingAccuracy);
printf("test complete...\n");
//print(PIMatrix,TESTSET,HIDDEN_NEURONS);
return 0;
}
int main(int argc, char* argv[]) {
Matrix G;
Matrix Y;
Matrix Cov;
LoadMatrix("input.mt.g", G);
LoadMatrix("input.mt.y", Y);
LoadMatrix("input.mt.cov", Cov);
Cov.SetColumnLabel(0, "c1");
Cov.SetColumnLabel(1, "c2");
Y.SetColumnLabel(0, "y1");
Y.SetColumnLabel(1, "y2");
Y.SetColumnLabel(2, "y3");
FormulaVector tests;
{
const char* tp1[] = {"y1"};
const char* tc1[] = {"c1"};
std::vector<std::string> p1(tp1, tp1 + 1);
std::vector<std::string> c1(tc1, tc1 + 1);
tests.add(p1, c1);
}
{
const char* tp1[] = {"y2"};
const char* tc1[] = {"c2"};
std::vector<std::string> p1(tp1, tp1 + 1);
std::vector<std::string> c1(tc1, tc1 + 1);
tests.add(p1, c1);
}
{
const char* tp1[] = {"y2"};
const char* tc1[] = {"c1", "c2"};
std::vector<std::string> p1(tp1, tp1 + 1);
std::vector<std::string> c1(tc1, tc1 + 2);
tests.add(p1, c1);
}
{
const char* tp1[] = {"y1"};
const char* tc1[] = {"1"};
std::vector<std::string> p1(tp1, tp1 + 1);
std::vector<std::string> c1(tc1, tc1 + 1);
tests.add(p1, c1);
}
AccurateTimer t;
{
FastMultipleTraitLinearRegressionScoreTest mt(1024);
bool ret = mt.FitNullModel(Cov, Y, tests);
if (ret == false) {
printf("Fit null model failed!\n");
exit(1);
}
ret = mt.AddGenotype(G);
if (ret == false) {
printf("Add covariate failed!\n");
exit(1);
}
ret = mt.TestCovariateBlock();
if (ret == false) {
printf("Test covariate block failed!\n");
exit(1);
}
const Vector& u = mt.GetU(0);
printf("u\t");
Print(u);
printf("\n");
const Vector& v = mt.GetV(0);
printf("v\t");
Print(v);
printf("\n");
const Vector& pval = mt.GetPvalue(0);
printf("pval\t");
Print(pval);
printf("\n");
}
return 0;
}
void ShaderProgram::LoadProjectionMatrix(GLuint location)
{
LoadMatrix(location, Camera::GetProjMatrix());
}
void Direct3D::D3DMatrixStack::LoadMatrixClass(Matrix &mMatrix) { LoadMatrix((D3DXMATRIX*)(&mMatrix)); }
void nuiDrawContext::LoadIdentity()
{
nuiMatrix m;
LoadMatrix(m);
}
// the state is only guaranteed after a SetModelMatrix call
void SetModelMatrix ( const matrix2x3& m )
{
modelMatrix = m;
mvpMatrix = modelMatrix * viewMatrix * projectionMatrix;
LoadMatrix(mvpMatrix);
}
//回归
void ELMTrain()
{
double starttime,endtime;
int i,j,k = 0;
double MSE = 0,TrainingAccuracy;
float **input,**weight,*biase,**tranpI,**tempH,**H;
float **PIMatrix;
float **train_set;
float **T,**Y;
float **out;
train_set = (float **)calloc(DATASET,sizeof(float *));
tranpI = (float **)calloc(INPUT_NEURONS,sizeof(float *));
input = (float **)calloc(DATASET,sizeof(float *)); /*datasize * INPUT_NEURONS*/
weight = (float **)calloc(HIDDEN_NEURONS,sizeof(float *)); /*HIDDEN_NEURONS * INPUT_NEURONS*/
biase = (float *)calloc(HIDDEN_NEURONS,sizeof(float)); /*HIDDEN_NEURONS*/
tempH = (float **)calloc(HIDDEN_NEURONS,sizeof(float *)); /*HIDDEN_NEURONS * datasize*/
PIMatrix = (float **)calloc(HIDDEN_NEURONS,sizeof(float *));
H = (float **)calloc(DATASET,sizeof(float *));
T = (float **)calloc(DATASET,sizeof(float *));
Y = (float **)calloc(DATASET,sizeof(float *));
out = (float **)calloc(HIDDEN_NEURONS,sizeof(float *));
for(i=0;i<DATASET;i++){
train_set[i] = (float *)calloc(NUMROWS,sizeof(float));
input[i] = (float *)calloc(INPUT_NEURONS,sizeof(float));
H[i] = (float *)calloc(HIDDEN_NEURONS,sizeof(float));
}
for(i=0;i<DATASET;i++)
{
T[i] = (float *)calloc(OUTPUT_NEURONS,sizeof(float));
Y[i] = (float *)calloc(OUTPUT_NEURONS,sizeof(float));
}
for(i=0;i<INPUT_NEURONS;i++)
tranpI[i] = (float *)calloc(DATASET,sizeof(float));
for(i=0;i<HIDDEN_NEURONS;i++)
{
weight[i] = (float *)calloc(INPUT_NEURONS,sizeof(float));
tempH[i] = (float *)calloc(DATASET,sizeof(float));
out[i] = (float *)calloc(OUTPUT_NEURONS,sizeof(float));
PIMatrix[i] = (float *)calloc(DATASET,sizeof(float));
}
printf("begin to random weight and biase...\n");
/*得到随机的偏置和权重*/
RandomWeight(weight,HIDDEN_NEURONS,INPUT_NEURONS);
RandomBiase(biase,HIDDEN_NEURONS);
/*加载数据集到内存*/
printf("begin to load input from the file...\n");
if(LoadMatrix(train_set,"../TrainingDataSet/covtype",DATASET,NUMROWS,1) == 0){
printf("load input file error!!!\n");
return;
}
/*将数据集划分成输入和输出*/
for(i = 0;i < DATASET ;i++)
{
T[k++][0] = train_set[i][0];
for(j = 1;j <= INPUT_NEURONS;j++)
{
input[i][j-1] = train_set[i][j];
}
}
SaveMatrix(input,"./result/input",DATASET,INPUT_NEURONS);
/*ELM*/
printf("begin to compute...\n");
starttime = omp_get_wtime();
TranspositionMatrix(input,tranpI,DATASET,INPUT_NEURONS);
printf("begin to compute step 1...\n");
MultiplyMatrix(weight,HIDDEN_NEURONS,INPUT_NEURONS, tranpI,INPUT_NEURONS,DATASET,tempH);
printf("begin to compute setp 2...\n");
AddMatrix_bais(tempH,biase,HIDDEN_NEURONS,DATASET);
printf("begin to compute step 3...\n");
SigmoidHandle(tempH,HIDDEN_NEURONS,DATASET);
printf("begin to compute step 4...\n");
TranspositionMatrix(tempH,H,HIDDEN_NEURONS,DATASET);
PseudoInverseMatrix(H,DATASET,HIDDEN_NEURONS,PIMatrix);
MultiplyMatrix(PIMatrix,HIDDEN_NEURONS,DATASET,T,DATASET,OUTPUT_NEURONS,out);
//SaveMatrix(H,"./result/H",DATASET,HIDDEN_NEURONS);
//SaveMatrix(PIMatrix,"./result/PIMatrix",HIDDEN_NEURONS,DATASET);
printf("begin to compute step 5...\n");
endtime = omp_get_wtime();
//保存输出权值
SaveMatrix(out,"./result/result",HIDDEN_NEURONS,OUTPUT_NEURONS);
//检测准确率
MultiplyMatrix(H,DATASET,HIDDEN_NEURONS,out,HIDDEN_NEURONS,OUTPUT_NEURONS,Y);
for(i = 0;i< DATASET;i++)
{
MSE += (Y[i][0] - T[i][0])*(Y[i][0] - T[i][0]);
}
SaveMatrix(T,"./result/t",DATASET,OUTPUT_NEURONS);
SaveMatrix(Y,"./result/y",DATASET,OUTPUT_NEURONS);
TrainingAccuracy = sqrt(MSE/DATASET);
printf("use time :%f\n",endtime - starttime);
printf("trainning accuracy :%f\n",TrainingAccuracy);
printf("train complete...\n");
//print(PIMatrix,DATASET,HIDDEN_NEURONS);
}