/// extract community affiliation from F_uc
void TAGMFast::GetCmtyVV(TVec<TIntV>& CmtyVV, const double Thres, const int MinSz) {
CmtyVV.Gen(NumComs, 0);
TIntFltH CIDSumFH(NumComs);
for (int c = 0; c < SumFV.Len(); c++) {
CIDSumFH.AddDat(c, SumFV[c]);
}
CIDSumFH.SortByDat(false);
for (int c = 0; c < NumComs; c++) {
int CID = CIDSumFH.GetKey(c);
TIntFltH NIDFucH(F.Len() / 10);
TIntV CmtyV;
IAssert(SumFV[CID] == CIDSumFH.GetDat(CID));
if (SumFV[CID] < Thres) { continue; }
for (int u = 0; u < F.Len(); u++) {
int NID = u;
if (! NodesOk) { NID = NIDV[u]; }
if (GetCom(u, CID) >= Thres) { NIDFucH.AddDat(NID, GetCom(u, CID)); }
}
NIDFucH.SortByDat(false);
NIDFucH.GetKeyV(CmtyV);
if (CmtyV.Len() >= MinSz) { CmtyVV.Add(CmtyV); }
}
if ( NumComs != CmtyVV.Len()) {
printf("Community vector generated. %d communities are ommitted\n", NumComs.Val - CmtyVV.Len());
}
}
void TTokenizer::GetTokens(const TStrV& TextV, TVec<TStrV>& TokenVV) const {
IAssert(TextV.Len() == TokenVV.Len()); // shall we rather say Tokens.Gen(Texts.Len(), 0); ?
for (int TextN = 0; TextN < TextV.Len(); TextN++) {
TStrV& TokenV = TokenVV[TextN];
TokenVV.Gen(32,0); // assume there will be at least 32 tokens, to avoid small resizes
GetTokens(TextV[TextN], TokenV);
}
}
void TIndex::TQmGixSumWithoutFqMerger<TQmGixItem, TQmGixResItem>::Def(const TQmGixKey& Key,
TVec<TQmGixItem>& MainV, TVec<TQmGixResItem>& ResV) const {
ResV.Gen(MainV.Len(), 0);
for (TQmGixItem& Item : MainV) {
ResV.Add(TQmGixResItem(Item.Val, 1));
}
}
int main(int argc, char* argv[]) {
Env = TEnv(argc, argv, TNotify::StdNotify);
Env.PrepArgs(TStr::Fmt("agmgen. build: %s, %s. Time: %s", __TIME__, __DATE__, TExeTm::GetCurTm()));
TExeTm ExeTm;
Try
const TStr InFNm = Env.GetIfArgPrefixStr("-i:", "DEMO", "Community affiliation data");
const TStr OutFPrx = Env.GetIfArgPrefixStr("-o:", "agm", "out file name prefix");
const int RndSeed = Env.GetIfArgPrefixInt("-rs:",10,"Rnd Seed");
const double DensityCoef= Env.GetIfArgPrefixFlt("-a:",0.6,"Power-law Coefficient a of density (density ~ N^(-a)");
const double ScaleCoef= Env.GetIfArgPrefixFlt("-c:",1.3,"Scaling Coefficient c of density (density ~ c");
TRnd Rnd(RndSeed);
TVec<TIntV> CmtyVV;
if(InFNm=="DEMO") {
CmtyVV.Gen(2);
TIntV NIdV;
for(int i=0;i<25;i++) {
TIntV& CmtyV = CmtyVV[0];
CmtyV.Add(i+1);
}
for(int i=15;i<40;i++) {
TIntV& CmtyV = CmtyVV[1];
CmtyV.Add(i+1);
}
}
else {
TVec<TIntV> CmtyVV;
TSsParser Ss(InFNm, ssfWhiteSep);
while (Ss.Next()) {
if(Ss.GetFlds()>0) {
TIntV CmtyV;
for(int i=0;i<Ss.GetFlds();i++) {
if(Ss.IsInt(i)){CmtyV.Add(Ss.GetInt(i));}
}
CmtyVV.Add(CmtyV);
}
}
printf("community loading completed (%d communities)\n",CmtyVV.Len());
}
PUNGraph AG = TAGM::GenAGM(CmtyVV,DensityCoef,ScaleCoef,Rnd);
TSnap::SaveEdgeList(AG,OutFPrx + ".edgelist.txt");
if(AG->GetNodes()<50) {
TAGM::GVizComGraph(AG,CmtyVV,OutFPrx + ".graph.gif");
}
Catch
printf("\nrun time: %s (%s)\n", ExeTm.GetTmStr(), TSecTm::GetCurTm().GetTmStr().CStr());
return 0;
}
void TPartialGS::GetBasisV(TVec<TFltV>& q) {
const int l = R.Len();
q.Gen(l);
for (int i = 0; i < l; i++) {
q[i].Gen(l);
q[i].PutAll(0.0);
}
for (int n = 0; n < l; n++) {
double Rnn = R[n][0];
for (int j = 0; j < n; j++) {
TLinAlg::AddVec(-R[j][n-j]/Rnn, q[j], q[n], q[n]);
}
q[n][n] = 1/Rnn;
}
}
void TGnuPlot::LoadTs(const TStr& FNm, TStrV& ColNmV, TVec<TFltKdV>& ColV) {
PSs Ss = TSs::LoadTxt(ssfTabSep, FNm);
int row = 0;
ColNmV.Clr();
while (Ss->At(0, row)[0] == '#') { row++; }
for (int c = 1; c < Ss->GetXLen(row); c+=2) {
ColNmV.Add(Ss->At(c, row));
}
row++;
ColV.Gen(ColNmV.Len(), ColNmV.Len());
for (; row < Ss->GetYLen(); row++) {
for (int c = 0; c < Ss->GetXLen(row); c+=2) {
if (Ss->At(c,row).Empty()) break;
ColV[c/2].Add(TFltKd(Ss->At(c,row).GetFlt(), Ss->At(c+1,row).GetFlt()));
}
}
}
/// load bipartite community affiliation graph from text file (each row contains the member node IDs for each community)
void TAGMUtil::LoadCmtyVV(const TStr& InFNm, TVec<TIntV>& CmtyVV) {
CmtyVV.Gen(Kilo(100), 0);
TSsParser Ss(InFNm, ssfWhiteSep);
while (Ss.Next()) {
if(Ss.GetFlds() > 0) {
TIntV CmtyV;
for (int i = 0; i < Ss.GetFlds(); i++) {
if (Ss.IsInt(i)) {
CmtyV.Add(Ss.GetInt(i));
}
}
CmtyVV.Add(CmtyV);
}
}
CmtyVV.Pack();
printf("community loading completed (%d communities)\n",CmtyVV.Len());
}
void TMKCCASemSpace::CalcKCCA(const TVec<TBowMatrix>& BowMatrixV, const int& Dims,
const double& Kapa, const int& CGMxIter, const int& HorstMxIter,
TVec<TFltVV>& AlphaVV) {
// some basic constants
const int Langs = BowMatrixV.Len();
IAssert(Langs > 1);
const int Docs = BowMatrixV[0].GetCols();
// reservate space on the output
AlphaVV.Gen(Langs);
for (int LangN = 0; LangN < Langs; LangN++) {
IAssert(BowMatrixV[LangN].GetCols() == Docs);
AlphaVV[LangN].Gen(Docs, Dims);
AlphaVV[LangN].PutAll(0.0);
}
// par vektorjev
TFltV x(Docs); // vektor dolzine docs
TLAMisc::FillRnd(x, TRnd(1)); // napolnemo z random elementi
TLinAlg::Normalize(x); // normiramo vektor
TFltV y(BowMatrixV[1].GetRows()); // nek vektor
// spreminanje elemnta v matriki
AlphaVV[1](5,4) = 4.43;
// skalarni produkt 5 stolpca matrike alphaVV[1] z vektorjem x
TLinAlg::DotProduct(AlphaVV[1], 5, x);
// pristeje 0.4 * prvi stolpec drugemu stoplcu
TLinAlg::AddVec(0.4, AlphaVV[1], 0, AlphaVV[1], 1);
// pristeje 0.4 * prvi stoplec vektorju x
TLinAlg::AddVec(0.4, AlphaVV[1], 0, x);
// naracunamo normo petega stoplca
double Norm = TLinAlg::Norm(AlphaVV[1], 5);
// preberemo cetrto vrstico matrike
TFltV z; AlphaVV[1].GetRow(4, z);
// preberemo cetrti stoplec matrike
TFltV zz; AlphaVV[1].GetCol(4, zz);
// mnozimo vektor x z matriko BowMatrixV[1]
BowMatrixV[1].Multiply(x, y);
BowMatrixV[1].MultiplyT(y, x);
BowMatrixV[1].Multiply(AlphaVV[1], 0, y);
}
void TStrFeatureSpace::GetAddIdFreqs(const TStrV& Features, TVec<TKeyDat<TStrFSSize, TInt> >& IdFreqs) {
TIntH Freqs;
for (int i = 0; i < Features.Len(); i++) {
int Id = GetAddId(Features[i]);
TInt Freq = 0;
if (Freqs.IsKeyGetDat(Id, Freq)) {
Freqs.AddDat(Id, Freq + 1);
} else {
Freqs.AddDat(Id, 1);
}
}
IdFreqs.Gen(Freqs.Len());
for (int i = 0; i < Freqs.Len(); i++) {
TInt Key, Freq;
Freqs.GetKeyDat(i, Key, Freq);
IdFreqs[i].Key = Key;
IdFreqs[i].Dat = Freq;
}
IdFreqs.Sort();
}
void TStrFeatureSpace::GetIdFreqs(const TVec<const char *>& Features, TVec<TKeyDat<TStrFSSize, TInt> >& IdFreqs) const {
TIntH Freqs;
for (int i = 0; i < Features.Len(); i++) {
TInt Id;
if (GetIfExistsId(Features[i], Id)) {
TInt Freq = 0;
if (Freqs.IsKeyGetDat(Id, Freq)) {
Freqs.AddDat(Id, Freq + 1);
} else {
Freqs.AddDat(Id, 1);
}
}
}
IdFreqs.Gen(Freqs.Len());
for (int i = 0; i < Freqs.Len(); i++) {
TInt Key, Freq;
Freqs.GetKeyDat(i, Key, Freq);
IdFreqs[i].Key = Key;
IdFreqs[i].Dat = Freq;
}
IdFreqs.Sort();
}
void TGraphEnumUtils::GetIsoGraphs(const PNGraph &G, TVec<PNGraph> &isoG) {
int nodes = G->GetNodes();
//
TIntV v(nodes); for(int i=0; i<nodes; i++) v[i]=i;
TVec<TIntV> perms; GetPermutations(v, 0, perms);
isoG.Gen(perms.Len());
//
for(int i=0; i<perms.Len(); i++) {
isoG[i] = TNGraph::New();
//Add nodes
for(int j=0; j<nodes; j++) isoG[i]->AddNode(j);
//Add edges
for(TNGraph::TEdgeI eIt=G->BegEI(); eIt<G->EndEI(); eIt++) {
int srcId = eIt.GetSrcNId();
int dstId = eIt.GetDstNId();
//
int pSrcId = perms[i][srcId];
int pDstId = perms[i][dstId];
//
isoG[i]->AddEdge(pSrcId, pDstId);
}
}
}
void TAGMFit::GetCmtyVV(TVec<TIntV>& CmtyVV, TFltV& QV, const double QMax) {
CmtyVV.Gen(CIDNSetV.Len(), 0);
QV.Gen(CIDNSetV.Len(), 0);
TIntFltH CIDLambdaH(CIDNSetV.Len());
for (int c = 0; c < CIDNSetV.Len(); c++) {
CIDLambdaH.AddDat(c, LambdaV[c]);
}
CIDLambdaH.SortByDat(false);
for (int c = 0; c < CIDNSetV.Len(); c++) {
int CID = CIDLambdaH.GetKey(c);
IAssert(LambdaV[CID] >= MinLambda);
double Q = exp( - (double) LambdaV[CID]);
if (Q > QMax) { continue; }
TIntV CmtyV;
CIDNSetV[CID].GetKeyV(CmtyV);
if (CmtyV.Len() == 0) { continue; }
if (CID == BaseCID) { //if the community is the base community(epsilon community), discard
IAssert(CmtyV.Len() == G->GetNodes());
} else {
CmtyVV.Add(CmtyV);
QV.Add(Q);
}
}
}
void TCliqueOverlap::CalculateOverlapMtx(const TVec<TIntV>& MaxCliques, int MinNodeOverlap, TVec<TIntV>& OverlapMtx) {
OverlapMtx.Clr();
//
int n = MaxCliques.Len();
//Convert max cliques to HashSets
TVec<THashSet<TInt> > cliques;
for (int i=0; i<n; i++) {
const int len = MaxCliques[i].Len();
cliques.Add();
if (len < MinNodeOverlap) { continue; }
THashSet<TInt>& set = cliques.Last(); set.Gen(len);
for (int j=0; j<len; j++) { set.AddKey(MaxCliques[i][j]); }
}
//Init clique clique overlap matrix
n = cliques.Len();
OverlapMtx.Gen(n);
for (int i=0; i<n; i++) OverlapMtx[i].Gen(n);
//Calculate clique clique overlap matrix
for (int i=0; i<n; i++) {
OverlapMtx[i][i] = cliques[i].Len();
for (int j=i+1; j<n; j++) {
OverlapMtx[i][j] = Intersection(cliques[i], cliques[j]); }
}
}
void TGraphEnumUtils::GetIsoGraphs(uint64 graphId, int nodes, TVec<PNGraph> &isoG) {
TIntV v(nodes); for(int i=0; i<nodes; i++) v[i]=i;
TVec<TIntV> perms; GetPermutations(v, 0, perms);
isoG.Gen(perms.Len());
//
TVec<TPair<int,int> > edges;
GetEdges(graphId, nodes, edges);
//
for(int i=0; i<perms.Len(); i++) {
isoG[i] = TNGraph::New();
//Add nodes
for(int j=0; j<nodes; j++) isoG[i]->AddNode(j);
//Add edges
for(int j=0; j<edges.Len(); j++) {
int srcId = edges[j].Val1;
int dstId = edges[j].Val2;
//
int pSrcId = perms[i][srcId];
int pDstId = perms[i][dstId];
//
isoG[i]->AddEdge(pSrcId, pDstId);
}
}
}
