void GetEigenVectorCentr(const PUNGraph& Graph, TIntFltH& EigenH, const double& Eps, const int& MaxIter) { const int NNodes = Graph->GetNodes(); EigenH.Gen(NNodes); for (TUNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++) { EigenH.AddDat(NI.GetId(), 1.0/NNodes); IAssert(NI.GetId() == EigenH.GetKey(EigenH.Len()-1)); } TFltV TmpV(NNodes); double diff = TFlt::Mx; for (int iter = 0; iter < MaxIter; iter++) { int j = 0; for (TUNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++, j++) { TmpV[j] = 0; for (int e = 0; e < NI.GetOutDeg(); e++) { TmpV[j] += EigenH.GetDat(NI.GetOutNId(e)); } } double sum = 0; for (int i = 0; i < TmpV.Len(); i++) { EigenH[i] = TmpV[i]; sum += EigenH[i]; } for (int i = 0; i < EigenH.Len(); i++) { EigenH[i] /= sum; } if (fabs(diff-sum) < Eps) { break; } //printf("\tdiff:%f\tsum:%f\n", fabs(diff-sum), sum); diff = sum; } }
int GetWeightedPageRankMP1(const PNEANet Graph, TIntFltH& PRankH, const TStr& Attr, const double& C, const double& Eps, const int& MaxIter) { if (!Graph->IsFltAttrE(Attr)) return -1; TFltV Weights = Graph->GetFltAttrVecE(Attr); int mxid = Graph->GetMxNId(); TFltV OutWeights(mxid); Graph->GetWeightOutEdgesV(OutWeights, Weights); /*for (TNEANet::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++) { OutWeights[NI.GetId()] = Graph->GetWeightOutEdges(NI, Attr); }*/ /*TIntFltH Weights; for (TNEANet::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++) { Weights.AddDat(NI.GetId(), Graph->GetWeightOutEdges(NI, Attr)); }*/ const int NNodes = Graph->GetNodes(); TVec<TNEANet::TNodeI> NV; //const double OneOver = 1.0/double(NNodes); PRankH.Gen(NNodes); for (TNEANet::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++) { NV.Add(NI); PRankH.AddDat(NI.GetId(), 1.0/NNodes); //IAssert(NI.GetId() == PRankH.GetKey(PRankH.Len()-1)); } TFltV TmpV(NNodes); for (int iter = 0; iter < MaxIter; iter++) { #pragma omp parallel for schedule(dynamic,10000) for (int j = 0; j < NNodes; j++) { TNEANet::TNodeI NI = NV[j]; TmpV[j] = 0; for (int e = 0; e < NI.GetInDeg(); e++) { const int InNId = NI.GetInNId(e); const TFlt OutWeight = OutWeights[InNId]; int EId = Graph->GetEId(InNId, NI.GetId()); const TFlt Weight = Weights[Graph->GetFltKeyIdE(EId)]; if (OutWeight > 0) { TmpV[j] += PRankH.GetDat(InNId) * Weight / OutWeight; } } TmpV[j] = C*TmpV[j]; // Berkhin (the correct way of doing it) //TmpV[j] = C*TmpV[j] + (1.0-C)*OneOver; // iGraph } double diff=0, sum=0, NewVal; #pragma omp parallel for reduction(+:sum) schedule(dynamic,10000) for (int i = 0; i < TmpV.Len(); i++) { sum += TmpV[i]; } const double Leaked = (1.0-sum) / double(NNodes); #pragma omp parallel for reduction(+:diff) schedule(dynamic,10000) for (int i = 0; i < PRankH.Len(); i++) { // re-instert leaked PageRank NewVal = TmpV[i] + Leaked; // Berkhin //NewVal = TmpV[i] / sum; // iGraph diff += fabs(NewVal-PRankH[i]); PRankH[i] = NewVal; } if (diff < Eps) { break; } } return 0; }
double TAGMFast::GetStepSizeByLineSearch(const int UID, const TIntFltH& DeltaV, const TIntFltH& GradV, const double& Alpha, const double& Beta, const int MaxIter) { double StepSize = 1.0; double InitLikelihood = LikelihoodForRow(UID); TIntFltH NewVarV(DeltaV.Len()); for(int iter = 0; iter < MaxIter; iter++) { for (int i = 0; i < DeltaV.Len(); i++){ int CID = DeltaV.GetKey(i); double NewVal = GetCom(UID, CID) + StepSize * DeltaV.GetDat(CID); if (NewVal < MinVal) { NewVal = MinVal; } if (NewVal > MaxVal) { NewVal = MaxVal; } NewVarV.AddDat(CID, NewVal); } if (LikelihoodForRow(UID, NewVarV) < InitLikelihood + Alpha * StepSize * DotProduct(GradV, DeltaV)) { StepSize *= Beta; } else { break; } if (iter == MaxIter - 1) { StepSize = 0.0; break; } } return StepSize; }
void GetEigenVectorCentr(const PUNGraph& Graph, TIntFltH& NIdEigenH, const double& Eps, const int& MaxIter) { const int NNodes = Graph->GetNodes(); NIdEigenH.Gen(NNodes); // initialize vector values for (TUNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++) { NIdEigenH.AddDat(NI.GetId(), 1.0 / NNodes); IAssert(NI.GetId() == NIdEigenH.GetKey(NIdEigenH.Len() - 1)); } TFltV TmpV(NNodes); for (int iter = 0; iter < MaxIter; iter++) { int j = 0; // add neighbor values for (TUNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++, j++) { TmpV[j] = 0; for (int e = 0; e < NI.GetOutDeg(); e++) { TmpV[j] += NIdEigenH.GetDat(NI.GetOutNId(e)); } } // normalize double sum = 0; for (int i = 0; i < TmpV.Len(); i++) { sum += (TmpV[i] * TmpV[i]); } sum = sqrt(sum); for (int i = 0; i < TmpV.Len(); i++) { TmpV[i] /= sum; } // compute difference double diff = 0.0; j = 0; for (TUNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++, j++) { diff += fabs(NIdEigenH.GetDat(NI.GetId()) - TmpV[j]); } // set new values j = 0; for (TUNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++, j++) { NIdEigenH.AddDat(NI.GetId(), TmpV[j]); } if (diff < Eps) { break; } } }
int main(int argc, char* argv[]) { Env = TEnv(argc, argv, TNotify::StdNotify); Env.PrepArgs(TStr::Fmt("Inverse PageRank. Build: %s, %s. Time: %s", __TIME__, __DATE__, TExeTm::GetCurTm())); TExeTm ExeTm; Try const TStr Iput = Env.GetIfArgPrefixStr("-i:", "Input.txt", "Input File" ); const TStr Oput = Env.GetIfArgPrefixStr("-o:", "Output.txt", "Output File"); FILE* fpI = fopen(Iput.CStr(), "r"); FILE* fpO = fopen(Oput.CStr(), "w"); const double C = 0.85; const int MaxIter = 50; const double Eps = 1e-9; PNGraph Graph = TSnap::LoadEdgeList< PNGraph > (Iput); fprintf(fpO, "\nNodes: %d, Edges: %d\n\n", Graph->GetNodes(), Graph->GetEdges()); const int NNodes = Graph->GetNodes(); const double OneOver = (double) 1.0 / (double) NNodes; TIntFltH PRankH; PRankH.Gen(NNodes); for (TNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++) PRankH.AddDat(NI.GetId(), OneOver); TFltV TmpV(NNodes); for (int iter = 0; iter < MaxIter; iter++) { int j = 0; for (TNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++, j++) { TmpV[j] = 0; for (int e = 0; e < NI.GetOutDeg(); e++) { const int OutNId = NI.GetOutNId(e); const int InDeg = Graph->GetNI(OutNId).GetInDeg(); if (InDeg > 0) TmpV[j] += PRankH.GetDat(OutNId) / InDeg; } TmpV[j] = C * TmpV[j]; } for (int i = 0; i < PRankH.Len(); i++) PRankH[i] = TmpV[i]; /* double diff = 0, sum = 0, NewVal; for (int i = 0; i < TmpV.Len(); i++) sum += TmpV[i]; const double Leaked = (double) (1.0 - sum) / (double) NNodes; for (int i = 0; i < PRankH.Len(); i++) { NewVal = TmpV[i] + Leaked; diff += fabs(NewVal - PRankH[i]); PRankH[i] = NewVal; } if (diff < Eps) break; */ } fprintf(fpO, "Node ID\t\tInverse PageRank\n"); for (TNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++){ int Id = NI.GetId(); double ipr = PRankH.GetDat(Id); fprintf(fpO, "%d\t\t\t%.5lf\n", Id, ipr); } Catch printf("\nRun Time: %s (%s)\n", ExeTm.GetTmStr(), TSecTm::GetCurTm().GetTmStr().CStr()); return 0; }
int main(int argc, char* argv[]) { Env = TEnv(argc, argv, TNotify::StdNotify); Env.PrepArgs(TStr::Fmt("Trust Rank. Build: %s, %s. Time: %s", __TIME__, __DATE__, TExeTm::GetCurTm())); TExeTm ExeTm; Try const TStr Gnod = Env.GetIfArgPrefixStr("-g:", "Gnode.txt", "Good Nodes"); const TStr Bnod = Env.GetIfArgPrefixStr("-b:", "Bnode.txt", "Bad Nodes" ); const TStr Iput = Env.GetIfArgPrefixStr("-i:", "Input.txt", "Input File"); const TStr Oput = Env.GetIfArgPrefixStr("-o:", "Output.txt", "Output File"); const double C = 0.85; const int MaxIter = 50; const double Eps = 1e-9; FILE* fpO = fopen(Oput.CStr(), "w"); PNGraph Graph = TSnap::LoadEdgeList< PNGraph > (Iput); fprintf(fpO, "\nNodes: %d, Edges: %d\n\n", Graph->GetNodes(), Graph->GetEdges()); const int NNodes = Graph->GetNodes(); TIntFltH TRankH; TRankH.Gen(NNodes); int maxNId = 0, NId = 0, ret = 0; for (TNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++) maxNId = max(maxNId, NI.GetId()); TFltV initialTrustScore(maxNId + 1); for (int i = 0; i < initialTrustScore.Len(); i++) initialTrustScore[i] = 0.5; FILE* fpI = fopen(Gnod.CStr(), "r"); while (true) { ret = fscanf(fpI, "%d", &NId); if (ret == EOF) break; if (Graph->IsNode(NId)) initialTrustScore[NId] = 1.0; } fclose(fpI); fpI = fopen(Bnod.CStr(), "r"); while (true) { ret = fscanf(fpI, "%d", &NId); if (ret == EOF) break; if (Graph->IsNode(NId)) initialTrustScore[NId] = 0.0; } fclose(fpI); double Tot = 0.0; for(int i = 0; i < initialTrustScore.Len(); i++) Tot += initialTrustScore[i]; for(int i = 0; i < initialTrustScore.Len(); i++) initialTrustScore[i] /= Tot; for (TNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++) TRankH.AddDat( NI.GetId(), initialTrustScore[NI.GetId()] ); TFltV TmpV(NNodes); for (int iter = 0; iter < MaxIter; iter++) { int j = 0; for (TNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++, j++) { TmpV[j] = 0; for (int e = 0; e < NI.GetOutDeg(); e++) { const int OutNId = NI.GetOutNId(e); const int InDeg = Graph->GetNI(InNId).GetInDeg(); if (InDeg > 0) TmpV[j] += (double) TRankH.GetDat(OutNId) / (double) InDeg; } TmpV[j] = C * TmpV[j] + (1.0 - C) * initialTrustScore[NI.GetId()]; } for (int i = 0; i < TRankH.Len(); i++) TRankH[i] = TmpV[i]; } fprintf(fpO, "Node ID\t\tTrustRank\n"); for (TNGraph::TNodeI NI = Graph->BegNI(); NI < Graph->EndNI(); NI++){ int Id = NI.GetId(); double tr = TRankH.GetDat(Id); fprintf(fpO, "%d\t\t\t%.5lf\n", Id, tr); } fclose(fpO); Catch printf("\nRun Time: %s (%s)\n", ExeTm.GetTmStr(), TSecTm::GetCurTm().GetTmStr().CStr()); return 0; }
int TAGMFast::MLEGradAscentParallel(const double& Thres, const int& MaxIter, const int ChunkNum, const int ChunkSize, const TStr PlotNm, const double StepAlpha, const double StepBeta) { //parallel time_t InitTime = time(NULL); uint64 StartTm = TSecTm::GetCurTm().GetAbsSecs(); TExeTm ExeTm, CheckTm; double PrevL = Likelihood(true); TIntFltPrV IterLV; int PrevIter = 0; int iter = 0; TIntV NIdxV(F.Len(), 0); for (int i = 0; i < F.Len(); i++) { NIdxV.Add(i); } TIntV NIDOPTV(F.Len()); //check if a node needs optimization or not 1: does not require optimization NIDOPTV.PutAll(0); TVec<TIntFltH> NewF(ChunkNum * ChunkSize); TIntV NewNIDV(ChunkNum * ChunkSize); for (iter = 0; iter < MaxIter; iter++) { NIdxV.Clr(false); for (int i = 0; i < F.Len(); i++) { if (NIDOPTV[i] == 0) { NIdxV.Add(i); } } IAssert (NIdxV.Len() <= F.Len()); NIdxV.Shuffle(Rnd); // compute gradient for chunk of nodes #pragma omp parallel for schedule(static, 1) for (int TIdx = 0; TIdx < ChunkNum; TIdx++) { TIntFltH GradV; for (int ui = TIdx * ChunkSize; ui < (TIdx + 1) * ChunkSize; ui++) { NewNIDV[ui] = -1; if (ui > NIdxV.Len()) { continue; } int u = NIdxV[ui]; // //find set of candidate c (we only need to consider c to which a neighbor of u belongs to) TUNGraph::TNodeI UI = G->GetNI(u); TIntSet CIDSet(5 * UI.GetDeg()); TIntFltH CurFU = F[u]; for (int e = 0; e < UI.GetDeg(); e++) { if (HOVIDSV[u].IsKey(UI.GetNbrNId(e))) { continue; } TIntFltH& NbhCIDH = F[UI.GetNbrNId(e)]; for (TIntFltH::TIter CI = NbhCIDH.BegI(); CI < NbhCIDH.EndI(); CI++) { CIDSet.AddKey(CI.GetKey()); } } if (CIDSet.Empty()) { CurFU.Clr(); } else { for (TIntFltH::TIter CI = CurFU.BegI(); CI < CurFU.EndI(); CI++) { //remove the community membership which U does not share with its neighbors if (! CIDSet.IsKey(CI.GetKey())) { CurFU.DelIfKey(CI.GetKey()); } } GradientForRow(u, GradV, CIDSet); if (Norm2(GradV) < 1e-4) { NIDOPTV[u] = 1; continue; } double LearnRate = GetStepSizeByLineSearch(u, GradV, GradV, StepAlpha, StepBeta, 5); if (LearnRate <= 1e-5) { NewNIDV[ui] = -2; continue; } for (int ci = 0; ci < GradV.Len(); ci++) { int CID = GradV.GetKey(ci); double Change = LearnRate * GradV.GetDat(CID); double NewFuc = CurFU.IsKey(CID)? CurFU.GetDat(CID) + Change : Change; if (NewFuc <= 0.0) { CurFU.DelIfKey(CID); } else { CurFU.AddDat(CID) = NewFuc; } } CurFU.Defrag(); } //store changes NewF[ui] = CurFU; NewNIDV[ui] = u; } } int NumNoChangeGrad = 0; int NumNoChangeStepSize = 0; for (int ui = 0; ui < NewNIDV.Len(); ui++) { int NewNID = NewNIDV[ui]; if (NewNID == -1) { NumNoChangeGrad++; continue; } if (NewNID == -2) { NumNoChangeStepSize++; continue; } for (TIntFltH::TIter CI = F[NewNID].BegI(); CI < F[NewNID].EndI(); CI++) { SumFV[CI.GetKey()] -= CI.GetDat(); } } #pragma omp parallel for for (int ui = 0; ui < NewNIDV.Len(); ui++) { int NewNID = NewNIDV[ui]; if (NewNID < 0) { continue; } F[NewNID] = NewF[ui]; } for (int ui = 0; ui < NewNIDV.Len(); ui++) { int NewNID = NewNIDV[ui]; if (NewNID < 0) { continue; } for (TIntFltH::TIter CI = F[NewNID].BegI(); CI < F[NewNID].EndI(); CI++) { SumFV[CI.GetKey()] += CI.GetDat(); } } // update the nodes who are optimal for (int ui = 0; ui < NewNIDV.Len(); ui++) { int NewNID = NewNIDV[ui]; if (NewNID < 0) { continue; } TUNGraph::TNodeI UI = G->GetNI(NewNID); NIDOPTV[NewNID] = 0; for (int e = 0; e < UI.GetDeg(); e++) { NIDOPTV[UI.GetNbrNId(e)] = 0; } } int OPTCnt = 0; for (int i = 0; i < NIDOPTV.Len(); i++) { if (NIDOPTV[i] == 1) { OPTCnt++; } } if (! PlotNm.Empty()) { printf("\r%d iterations [%s] %d secs", iter * ChunkSize * ChunkNum, ExeTm.GetTmStr(), int(TSecTm::GetCurTm().GetAbsSecs() - StartTm)); if (PrevL > TFlt::Mn) { printf(" (%f) %d g %d s %d OPT", PrevL, NumNoChangeGrad, NumNoChangeStepSize, OPTCnt); } fflush(stdout); } if ((iter - PrevIter) * ChunkSize * ChunkNum >= G->GetNodes()) { PrevIter = iter; double CurL = Likelihood(true); IterLV.Add(TIntFltPr(iter * ChunkSize * ChunkNum, CurL)); printf("\r%d iterations, Likelihood: %f, Diff: %f [%d secs]", iter, CurL, CurL - PrevL, int(time(NULL) - InitTime)); fflush(stdout); if (CurL - PrevL <= Thres * fabs(PrevL)) { break; } else { PrevL = CurL; } } } if (! PlotNm.Empty()) { printf("\nMLE completed with %d iterations(%s secs)\n", iter, int(TSecTm::GetCurTm().GetAbsSecs() - StartTm)); TGnuPlot::PlotValV(IterLV, PlotNm + ".likelihood_Q");[] } else {
int TAGMFast::MLEGradAscent(const double& Thres, const int& MaxIter, const TStr PlotNm, const double StepAlpha, const double StepBeta) { time_t InitTime = time(NULL); TExeTm ExeTm, CheckTm; int iter = 0, PrevIter = 0; TIntFltPrV IterLV; TUNGraph::TNodeI UI; double PrevL = TFlt::Mn, CurL = 0.0; TIntV NIdxV(F.Len(), 0); for (int i = 0; i < F.Len(); i++) { NIdxV.Add(i); } IAssert(NIdxV.Len() == F.Len()); TIntFltH GradV; while(iter < MaxIter) { NIdxV.Shuffle(Rnd); for (int ui = 0; ui < F.Len(); ui++, iter++) { int u = NIdxV[ui]; // //find set of candidate c (we only need to consider c to which a neighbor of u belongs to) UI = G->GetNI(u); TIntSet CIDSet(5 * UI.GetDeg()); for (int e = 0; e < UI.GetDeg(); e++) { if (HOVIDSV[u].IsKey(UI.GetNbrNId(e))) { continue; } TIntFltH& NbhCIDH = F[UI.GetNbrNId(e)]; for (TIntFltH::TIter CI = NbhCIDH.BegI(); CI < NbhCIDH.EndI(); CI++) { CIDSet.AddKey(CI.GetKey()); } } for (TIntFltH::TIter CI = F[u].BegI(); CI < F[u].EndI(); CI++) { //remove the community membership which U does not share with its neighbors if (! CIDSet.IsKey(CI.GetKey())) { DelCom(u, CI.GetKey()); } } if (CIDSet.Empty()) { continue; } GradientForRow(u, GradV, CIDSet); if (Norm2(GradV) < 1e-4) { continue; } double LearnRate = GetStepSizeByLineSearch(u, GradV, GradV, StepAlpha, StepBeta); if (LearnRate == 0.0) { continue; } for (int ci = 0; ci < GradV.Len(); ci++) { int CID = GradV.GetKey(ci); double Change = LearnRate * GradV.GetDat(CID); double NewFuc = GetCom(u, CID) + Change; if (NewFuc <= 0.0) { DelCom(u, CID); } else { AddCom(u, CID, NewFuc); } } if (! PlotNm.Empty() && (iter + 1) % G->GetNodes() == 0) { IterLV.Add(TIntFltPr(iter, Likelihood(false))); } } printf("\r%d iterations (%f) [%lu sec]", iter, CurL, time(NULL) - InitTime); fflush(stdout); if (iter - PrevIter >= 2 * G->GetNodes() && iter > 10000) { PrevIter = iter; CurL = Likelihood(); if (PrevL > TFlt::Mn && ! PlotNm.Empty()) { printf("\r%d iterations, Likelihood: %f, Diff: %f", iter, CurL, CurL - PrevL); } fflush(stdout); if (CurL - PrevL <= Thres * fabs(PrevL)) { break; } else { PrevL = CurL; } } } printf("\n"); printf("MLE for Lambda completed with %d iterations(%s)\n", iter, ExeTm.GetTmStr()); if (! PlotNm.Empty()) { TGnuPlot::PlotValV(IterLV, PlotNm + ".likelihood_Q"); } return iter; }
void TAGMFast::GradientForRow(const int UID, TIntFltH& GradU, const TIntSet& CIDSet) { GradU.Gen(CIDSet.Len()); TFltV HOSumFV; //adjust for Fv of v hold out if (HOVIDSV[UID].Len() > 0) { HOSumFV.Gen(SumFV.Len()); for (int e = 0; e < HOVIDSV[UID].Len(); e++) { for (int c = 0; c < SumFV.Len(); c++) { HOSumFV[c] += GetCom(HOVIDSV[UID][e], c); } } } TUNGraph::TNodeI NI = G->GetNI(UID); int Deg = NI.GetDeg(); TFltV PredV(Deg), GradV(CIDSet.Len()); TIntV CIDV(CIDSet.Len()); if (DoParallel && Deg + CIDSet.Len() > 10) { #pragma omp parallel for schedule(static, 1) for (int e = 0; e < Deg; e++) { if (NI.GetNbrNId(e) == UID) { continue; } if (HOVIDSV[UID].IsKey(NI.GetNbrNId(e))) { continue; } PredV[e] = Prediction(UID, NI.GetNbrNId(e)); } #pragma omp parallel for schedule(static, 1) for (int c = 0; c < CIDSet.Len(); c++) { int CID = CIDSet.GetKey(c); double Val = 0.0; for (int e = 0; e < Deg; e++) { int VID = NI.GetNbrNId(e); if (VID == UID) { continue; } if (HOVIDSV[UID].IsKey(VID)) { continue; } Val += PredV[e] * GetCom(VID, CID) / (1.0 - PredV[e]) + NegWgt * GetCom(VID, CID); } double HOSum = HOVIDSV[UID].Len() > 0? HOSumFV[CID].Val: 0.0;//subtract Hold out pairs only if hold out pairs exist Val -= NegWgt * (SumFV[CID] - HOSum - GetCom(UID, CID)); CIDV[c] = CID; GradV[c] = Val; } } else { for (int e = 0; e < Deg; e++) { if (NI.GetNbrNId(e) == UID) { continue; } if (HOVIDSV[UID].IsKey(NI.GetNbrNId(e))) { continue; } PredV[e] = Prediction(UID, NI.GetNbrNId(e)); } for (int c = 0; c < CIDSet.Len(); c++) { int CID = CIDSet.GetKey(c); double Val = 0.0; for (int e = 0; e < Deg; e++) { int VID = NI.GetNbrNId(e); if (VID == UID) { continue; } if (HOVIDSV[UID].IsKey(VID)) { continue; } Val += PredV[e] * GetCom(VID, CID) / (1.0 - PredV[e]) + NegWgt * GetCom(VID, CID); } double HOSum = HOVIDSV[UID].Len() > 0? HOSumFV[CID].Val: 0.0;//subtract Hold out pairs only if hold out pairs exist Val -= NegWgt * (SumFV[CID] - HOSum - GetCom(UID, CID)); CIDV[c] = CID; GradV[c] = Val; } } //add regularization if (RegCoef > 0.0) { //L1 for (int c = 0; c < GradV.Len(); c++) { GradV[c] -= RegCoef; } } if (RegCoef < 0.0) { //L2 for (int c = 0; c < GradV.Len(); c++) { GradV[c] += 2 * RegCoef * GetCom(UID, CIDV[c]); } } for (int c = 0; c < GradV.Len(); c++) { if (GetCom(UID, CIDV[c]) == 0.0 && GradV[c] < 0.0) { continue; } if (fabs(GradV[c]) < 0.0001) { continue; } GradU.AddDat(CIDV[c], GradV[c]); } for (int c = 0; c < GradU.Len(); c++) { if (GradU[c] >= 10) { GradU[c] = 10; } if (GradU[c] <= -10) { GradU[c] = -10; } IAssert(GradU[c] >= -10); } }