void FactorAnalysisStat::substractChannelStats(){
if (verbose) cout <<"(FactorAnalysisStat) Compute and Substract Channel FA Stats... "<<endl;
RealVector <double> UX;UX.setSize(_supervsize); UX.setAllValues(0.0);double* ux=UX.getArray();
// Compute Occupations and Statistics
unsigned long loc=0;
unsigned long sent=0;
XLine *pline; String *pFile;
fileList.rewind();
double *super_mean=_super_mean.getArray();
double *N_h=_matN_h.getArray();
double *S_X=_matS_X.getArray();
while((pline=fileList.getLine())!=NULL) {
while((pFile=pline->getElement())!=NULL) {
this->getUX(UX,*pFile);
for(unsigned long k=0;k<_supervsize;k++)
ux[k]+=super_mean[k];
for(unsigned long k=0;k<_mixsize;k++)
for (unsigned long i=0;i<_vsize;i++)
S_X[loc*_supervsize+(k*_vsize+i)]-= N_h[sent*_mixsize+k]*ux[i+k*_vsize];
sent++;
}
loc++;
}
};
void FactorAnalysisStat::getXEstimate(){
if (verbose) cout << "(FactorAnalysisStat) Compute X Estimate "<<endl;
RealVector <double> AUX;
AUX.setSize(_rang);
XLine *pline;
String *pFile;
_matX.setAllValues(0.0);
double *X=_matX.getArray();
double *U=_matU.getArray();
double *S_X_h=_matS_X_h.getArray();
double *aux=AUX.getArray();
double *super_invvar=_super_invvar.getArray();
fileList.rewind();
while((pline=fileList.getLine())!=NULL) {
while((pFile=pline->getElement())!=NULL) {
unsigned long sent=_ndxTable.sessionNb(*pFile);
AUX.setAllValues(0.0);
for(unsigned long i=0;i<_rang;i++)
for(unsigned long k=0;k<_supervsize;k++)
aux[i]+= U[k*_rang+i]*super_invvar[k]*S_X_h[sent*_supervsize+k];
double *l_h_inv=_l_h_inv[sent].getArray();
for(unsigned long i=0;i<_rang;i++)
for(unsigned long k=0;k<_rang;k++)
X[sent*_rang+i]+=l_h_inv[i*_rang+k]*aux[k];
sent++;
}
}
};
//-------------------------------------------------------------------------
// Load the NGRAM table, selecting the nbSelected first
void NGram::load(const String filename,Config &config){
XList input(filename,config);
XLine *linep; // Pointor on the current test line
input.getLine(0);
unsigned long idx=0;
while (((linep=input.getLine()) != NULL)&&(idx<getSize())){
for (unsigned long i=0;i<getOrder();i++){
short int a=linep->getElement(i).toLong();
setSymbol(idx,i,a);
}
if (linep->getElementCount()==(getOrder()+1)){
unsigned long count=linep->getElement(getOrder()).toLong();
setCount(idx,count);_totalCount+=count;}
else setCount(idx,0);
idx++;
}
if (idx!=getSize()){
cout << "WARNING ! Number of ngram in the file["<<idx<<"] < to the number requested ["<<getSize()<<"]"<<endl;
setSize(idx);
}
if (verboseLevel>1){
cout <<"load symbol table from ["<<filename <<"]"<<endl;
showTable();
}
}
// Information on the quantity of data available for each client
// Outputs a list with the selected files for a defined quantity of data
int ExtractTargetDataInfo(Config& config)
{
String inputClientListFileName = config.getParam("targetIdList");
bool fixedLabelSelectedFrame;
String labelSelectedFrames;
if (config.existsParam("useIdForSelectedFrame")) // the ID of each speaker is used as labelSelectedFrame
fixedLabelSelectedFrame=false;
else{ // the label is decided by the command line and is unique for the run
labelSelectedFrames=config.getParam("labelSelectedFrames");
if (verbose) cout << "Computing on" << labelSelectedFrames << " label" << endl;
fixedLabelSelectedFrame=true;
}
unsigned long maxFrame=config.getParam("maxFrame").toLong();
String outputFilename=config.getParam("outputFilename");
ofstream outputFile(outputFilename.c_str(),ios::out| ios::trunc);
try{
XList inputClientList(inputClientListFileName,config); // read the Id + filenames for each client
XLine * linep;
if (verbose) cout << "InfoTarget" << endl;
// *********** Target loop *****************
while ((linep=inputClientList.getLine()) != NULL){ // linep gives the XLine with the Id of a given client and the list of files
String *id=linep->getElement(); // Get the Client ID (id)
outputFile<<*id;
String currentFile="";
XLine featureFileListp=linep->getElements(); // Get the list of feature file for the client (end of the line)
if (verbose) cout << "Info model ["<<*id<<"]"<<endl;
if (!fixedLabelSelectedFrame){ // the ID is used as label for selecting the frame
labelSelectedFrames=*id;
if (debug) cout <<*id<<" is used for label selected frames"<<endl;
}
// label files reading - It creates, for each file and each label, a cluster of segments - will be integrated witth the featre s - asap
SegServer segmentsServer; // Reading the segmentation files for each feature input file
LabelServer labelServer;
initializeClusters(featureFileListp,segmentsServer,labelServer,config); // Reading the segmentation files for each feature input file
unsigned long codeSelectedFrame=labelServer.getLabelIndexByString(labelSelectedFrames); // Get the index of the cluster with in interest audio segments
SegCluster& selectedSegments=segmentsServer.getCluster(codeSelectedFrame); // Gives the cluster of the selected/used segments
Seg *seg; // Will give the current segment
unsigned long frameCount=0;
selectedSegments.rewind(); // at the begin of the selected segments list
while(((seg=selectedSegments.getSeg())!=NULL) && (frameCount<maxFrame)){ // For each of the selected segments until the amount of data is get
frameCount+=seg->length();
cout << seg->sourceName()<<" "<<seg->begin()<<" "<<seg->length()<<" Total time="<<frameCount<<endl;
if (seg->sourceName()!=currentFile){
outputFile<<" "<<seg->sourceName();
currentFile=seg->sourceName();
}
} // end of the initial Train Iteration loop
outputFile<<endl;
if (verbose) cout << "Save info client ["<<*id<<"]" << endl;
} // end of the the target loop
} // fin try
catch (Exception& e)
{
cout << e.toString().c_str() << endl;
}
return 0;
}
FactorAnalysisStat::FactorAnalysisStat(String & featFilename,FeatureServer & fs,Config & config):_ms(config),_ss(config){ // constructor for a single file
XList faNdx;
XLine featLine;
featLine.addElement(featFilename);
faNdx.addLine()=featLine;
_init(faNdx,fs,config);
}
//-------------------------------------------------------------------------
// Get a tab with indexes of speakers with maximum likelihood (mdtm and etf only)
void getTarClientIdx(Config & config, XList & inputList, unsigned long nbLoc, unsigned long * tarTab)
{
XLine *linep;
unsigned long cpt=0;
unsigned long cpttab=0;
double minLLK=config.getParam("minLLK").toDouble();
double maxScore=minLLK;
unsigned long idxTar=0;
bool verbose=config.existsParam("verbose");
while((linep=inputList.getLine())!=NULL)
{
double score=linep->getElement(6).toDouble();
if (score>=maxScore)
{
maxScore=score; // this is the maximum score
idxTar=cpt; // index is just the line
if (verbose) {cout << "giving highest score to " << linep->getElement(1) << " "<<maxScore << endl;}
}
if (cpt%nbLoc==(nbLoc-1)) // when changing segment
{
tarTab[cpttab]=idxTar; // idx of the target goes in the tab
if (verbose) {cout << linep->getElement(1) << " max score: "<<maxScore <<"idx: "<<idxTar<<"cpt: "<<cpt<< endl;}
cpttab++;
maxScore=minLLK; //reset maxScore
}
cpt++;
}
}
/*
* Checks what users match a given vector of ELines and sets their ban exempt flag accordingly.
*/
void XLineManager::CheckELines()
{
ContainerIter n = lookup_lines.find("E");
if (n == lookup_lines.end())
return;
XLineLookup& ELines = n->second;
if (ELines.empty())
return;
for (LocalUserList::const_iterator u2 = ServerInstance->Users->local_users.begin(); u2 != ServerInstance->Users->local_users.end(); u2++)
{
LocalUser* u = *u2;
/* This uses safe iteration to ensure that if a line expires here, it doenst trash the iterator */
LookupIter safei;
for (LookupIter i = ELines.begin(); i != ELines.end(); )
{
safei = i;
safei++;
XLine *e = i->second;
u->exempt = e->Matches(u);
i = safei;
}
}
}
EventReturn OnCheckKick(User *u, Channel *c, Anope::string &mask, Anope::string &reason) override
{
XLine *x = this->sqlines.CheckChannel(c);
if (x)
{
this->sqlines.OnMatch(u, x);
reason = x->GetReason();
return EVENT_STOP;
}
return EVENT_CONTINUE;
}
ModResult OnUserRegister(LocalUser* user) override
{
// Apply lines on user connect
XLine *rl = ServerInstance->XLines->MatchesLine("R", user);
if (rl)
{
// Bang. :P
rl->Apply(user);
return MOD_RES_DENY;
}
return MOD_RES_PASSTHRU;
}
CmdResult CommandAddLine::Handle(User* usr, std::vector<std::string>& params)
{
XLineFactory* xlf = ServerInstance->XLines->GetFactory(params[0]);
const std::string& setter = usr->nick;
if (!xlf)
{
ServerInstance->SNO->WriteToSnoMask('d',"%s sent me an unknown ADDLINE type (%s).",setter.c_str(),params[0].c_str());
return CMD_FAILURE;
}
XLine* xl = NULL;
try
{
xl = xlf->Generate(ServerInstance->Time(), ConvToInt(params[4]), params[2], params[5], params[1]);
}
catch (ModuleException &e)
{
ServerInstance->SNO->WriteToSnoMask('d',"Unable to ADDLINE type %s from %s: %s", params[0].c_str(), setter.c_str(), e.GetReason().c_str());
return CMD_FAILURE;
}
xl->SetCreateTime(ConvToInt(params[3]));
if (ServerInstance->XLines->AddLine(xl, NULL))
{
if (xl->duration)
{
std::string timestr = InspIRCd::TimeString(xl->expiry);
ServerInstance->SNO->WriteToSnoMask('X',"%s added %s%s on %s to expire on %s: %s",setter.c_str(),params[0].c_str(),params[0].length() == 1 ? "-line" : "",
params[1].c_str(), timestr.c_str(), params[5].c_str());
}
else
{
ServerInstance->SNO->WriteToSnoMask('X',"%s added permanent %s%s on %s: %s",setter.c_str(),params[0].c_str(),params[0].length() == 1 ? "-line" : "",
params[1].c_str(),params[5].c_str());
}
TreeServer* remoteserver = TreeServer::Get(usr);
if (!remoteserver->IsBursting())
{
ServerInstance->XLines->ApplyLines();
}
return CMD_SUCCESS;
}
else
{
delete xl;
return CMD_FAILURE;
}
}
void OnUserPostNick(User *user, const std::string &oldnick) override
{
if (!IS_LOCAL(user))
return;
if (!MatchOnNickChange)
return;
XLine *rl = ServerInstance->XLines->MatchesLine("R", user);
if (rl)
{
// Bang! :D
rl->Apply(user);
}
}
bool LocalUser::CheckLines(bool doZline)
{
const char* check[] = { "G" , "K", (doZline) ? "Z" : NULL, NULL };
if (!this->exempt)
{
for (int n = 0; check[n]; ++n)
{
XLine *r = ServerInstance->XLines->MatchesLine(check[n], this);
if (r)
{
r->Apply(this);
return true;
}
}
}
return false;
}
// applies lines, removing clients and changing nicks etc as applicable
void XLineManager::ApplyLines()
{
for (std::vector<User*>::const_iterator u2 = ServerInstance->Users->local_users.begin(); u2 != ServerInstance->Users->local_users.end(); u2++)
{
User* u = (User*)(*u2);
// Don't ban people who are exempt.
if (u->exempt)
continue;
for (std::vector<XLine *>::iterator i = pending_lines.begin(); i != pending_lines.end(); i++)
{
XLine *x = *i;
if (x->Matches(u))
x->Apply(u);
}
}
pending_lines.clear();
}
// applies lines, removing clients and changing nicks etc as applicable
void XLineManager::ApplyLines()
{
LocalUserList::reverse_iterator u2 = ServerInstance->Users->local_users.rbegin();
while (u2 != ServerInstance->Users->local_users.rend())
{
LocalUser* u = *u2++;
// Don't ban people who are exempt.
if (u->exempt)
continue;
for (std::vector<XLine *>::iterator i = pending_lines.begin(); i != pending_lines.end(); i++)
{
XLine *x = *i;
if (x->Matches(u))
x->Apply(u);
}
}
pending_lines.clear();
}
static void ReadXLine(ServerConfig* conf, const std::string& tag, const std::string& key, XLineFactory* make)
{
insp::flat_set<std::string> configlines;
ConfigTagList tags = conf->ConfTags(tag);
for(ConfigIter i = tags.first; i != tags.second; ++i)
{
ConfigTag* ctag = i->second;
std::string mask;
if (!ctag->readString(key, mask))
throw CoreException("<"+tag+":"+key+"> missing at " + ctag->getTagLocation());
std::string reason = ctag->getString("reason", "<Config>");
XLine* xl = make->Generate(ServerInstance->Time(), 0, "<Config>", reason, mask);
xl->from_config = true;
configlines.insert(xl->Displayable());
if (!ServerInstance->XLines->AddLine(xl, NULL))
delete xl;
}
ServerInstance->XLines->ExpireRemovedConfigLines(make->GetType(), configlines);
}
//-------------------------------------------------------------------------
// Produce a tab with mean and cov by segment without the maximum score(mdtm and etf only)
void getSegmentalMeanCovWithoutMax(Config & config, XList & inputList, unsigned long nbLoc, unsigned long * tarTab, double* meanTab, double * covTab)
{
XLine *linep;
unsigned long cpt=0;
unsigned long cpttab=0;
double minLLK=config.getParam("minLLK").toDouble();
double maxScore=minLLK;
double meanAcc=0.0;
double covAcc=0.0;
unsigned long idxTar=0;
bool verbose=config.existsParam("verbose");
while((linep=inputList.getLine())!=NULL)
{
double score=linep->getElement(6).toDouble();
if (score>=maxScore)
{
maxScore=score; // this is the maximum score
idxTar=cpt; // index is just the line
if (verbose) {cout << "giving highest score to " << linep->getElement(1) << " "<<maxScore << endl;}
}
meanAcc+=score; // Accumulate mean and cov
covAcc+=(score*score);
if (cpt%nbLoc==(nbLoc-1)) // when changing segment
{
tarTab[cpttab]=idxTar;
meanAcc-=maxScore; //remove max from Stats
covAcc-=(maxScore*maxScore);
meanTab[cpttab]=meanAcc;
covTab[cpttab]=covAcc;
if (verbose) {cout << linep->getElement(1) << " max score: "<<maxScore <<"idx: "<<idxTar<<"cpt: "<<cpt<< " meanA: "<<meanAcc<<" covA: "<<covAcc<<endl;}
cpttab++;
maxScore=minLLK;
meanAcc=0.0;
covAcc=0.0;
}
cpt++;
}
}
void SendAkill(User *u, XLine *x) override
{
if (x->IsRegex() || x->HasNickOrReal())
{
if (!u)
{
/*
* No user (this akill was just added), and contains nick and/or realname.
* Find users that match and ban them.
*/
for (user_map::const_iterator it = UserListByNick.begin(); it != UserListByNick.end(); ++it)
if (x->manager->Check(it->second, x))
this->SendAkill(it->second, x);
return;
}
XLine *old = x;
if (old->manager->HasEntry("*@" + u->host))
return;
/* We can't akill x as it has a nick and/or realname included, so create a new akill for *@host */
XLine *xl = new XLine("*@" + u->host, old->GetBy(), old->GetExpires(), old->GetReason(), old->id);
old->manager->AddXLine(xl);
x = xl;
Log(Config->GetClient("OperServ"), "akill") << "AKILL: Added an akill for " << x->GetMask() << " because " << u->GetMask() << "#"
<< u->realname << " matches " << old->GetMask();
}
/* Calculate the time left before this would expire, capping it at 2 days */
time_t timeleft = x->GetExpires() - Anope::CurTime;
if (timeleft > 172800 || !x->GetExpires())
timeleft = 172800;
UplinkSocket::Message(Config->GetClient("OperServ")) << "KLINE * " << timeleft << " " << x->GetUser() << " " << x->GetHost() << " :" << x->GetReason();
}
void FactorAnalysisStat::estimateAndInverseLUnThreaded(){
if (verbose) cout << "(FactorAnalysisStat) Inverse L Matrix ... "<<endl;
unsigned long mk;
DoubleSquareMatrix L(_rang);
L.setAllValues(0.0);
RealVector <double> AUX;
AUX.setSize(_rang);
unsigned long sent=0;
XLine *pline;
String *pFile;
fileList.rewind();
double *N_h=_matN_h.getArray();
double *U=_matU.getArray();
double *LV=L.getArray();
double *super_invvar=_super_invvar.getArray();
while((pline=fileList.getLine())!=NULL) {
while((pFile=pline->getElement())!=NULL) {
L.setAllValues(0.0);
AUX.setAllValues(0.0);
for(unsigned long i=0;i<_rang;i++){
for(unsigned long j=0;j<=i;j++){
for(unsigned long k=0;k<_supervsize;k++){
mk=k/_vsize;
LV[i*_rang+j]+=N_h[sent*_mixsize+mk]*super_invvar[k]*U[k*_rang+i]*U[k*_rang+j];
}
}
}
for(unsigned long i=0;i<_rang;i++)
for(unsigned long j=i+1;j<_rang;j++)
LV[i*_rang+j]=LV[j*_rang+i];
for(unsigned long i=0;i<_rang;i++)
LV[i*_rang+i]+=1.0;
L.invert(_l_h_inv[sent]);
sent++;
}
}
};
void OnLookupComplete(const Query *record) override
{
if (!user || user->Quitting())
return;
const ResourceRecord &ans_record = record->answers[0];
// Replies should be in 127.0.0.0/8
if (ans_record.rdata.find("127.") != 0)
return;
sockaddrs sresult;
sresult.pton(AF_INET, ans_record.rdata);
int result = sresult.sa4.sin_addr.s_addr >> 24;
Blacklist::Reply *reply = blacklist.Find(result);
if (!blacklist.replies.empty() && !reply)
return;
if (reply && reply->allow_account && user->Account())
return;
Anope::string reason = this->blacklist.reason, addr = user->ip.addr();
reason = reason.replace_all_cs("%n", user->nick);
reason = reason.replace_all_cs("%u", user->GetIdent());
reason = reason.replace_all_cs("%g", user->realname);
reason = reason.replace_all_cs("%h", user->host);
reason = reason.replace_all_cs("%i", addr);
reason = reason.replace_all_cs("%r", reply ? reply->reason : "");
reason = reason.replace_all_cs("%N", Config->GetBlock("networkinfo")->Get<Anope::string>("networkname"));
ServiceBot *OperServ = Config->GetClient("OperServ");
creator->logger.Category("dnsbl").Bot(OperServ).Log(_("{0} ({1}) appears in {2}"), user->GetMask(), addr, this->blacklist.name);
XLine *x = Serialize::New<XLine *>();
x->SetMask("*@" + addr);
x->SetBy(OperServ ? OperServ->nick : "m_dnsbl");
x->SetCreated(Anope::CurTime);
x->SetExpires(Anope::CurTime + this->blacklist.bantime);
x->SetReason(reason);
x->SetID(XLineManager::GenerateUID());
if (this->add_to_akill && akills)
{
akills->AddXLine(x);
akills->Send(NULL, x);
}
else
{
IRCD->Send<messages::Akill>(nullptr, x);
delete x;
}
}
//-------------------------------------------------------------------------
// Produce a tab with mean and cov by segment (mdtm and etf only)
void getSegmentalMeanCov(XList & inputList, unsigned long nbLoc, double* meanTab, double * covTab)
{
XLine *linep;
unsigned long cpt=0;
unsigned long cpttab=0;
double meanAcc=0.0;
double covAcc=0.0;
while((linep=inputList.getLine())!=NULL)
{
double score=linep->getElement(6).toDouble();
meanAcc+=score; // Accumulate mean and cov
covAcc+=(score*score);
if (cpt%nbLoc==(nbLoc-1)) // when changing segment
{
meanTab[cpttab]=meanAcc;
covTab[cpttab]=covAcc;
cpttab++;
meanAcc=0.0;
covAcc=0.0;
}
cpt++;
}
}
void FactorAnalysisStat::substractSpeakerStats(){
if (verbose) cout <<"(FactorAnalysisStat) Compute and Substract Speaker FA Stats... " << endl;
RealVector <double> AUX1;AUX1.setSize(_supervsize); AUX1.setAllValues(0.0); double *aux1=AUX1.getArray();
// Compute Occupations and Statistics
unsigned long loc=0;
unsigned long sent=0;
XLine *pline; String *pFile; fileList.rewind();
double *N_h=_matN_h.getArray();
double *S_X_h=_matS_X_h.getArray();
while((pline=fileList.getLine())!=NULL) {
while((pFile=pline->getElement())!=NULL) {
this->getMplusDY(AUX1,*pFile);
for(unsigned long k=0;k<_mixsize;k++)
for (unsigned long i=0;i<_vsize;i++)
S_X_h[sent*_supervsize+(k*_vsize+i)]-= N_h[sent*_mixsize+k]*aux1[i+k*_vsize];
sent++;
}
loc++;
}
if (verboseLevel >1) cout << "(FactorAnalysisStat) Done "<<endl;
};
void bahamut::senders::Akill::Send(User* u, XLine* x)
{
if (x->IsRegex() || x->HasNickOrReal())
{
if (!u)
{
/* No user (this akill was just added), and contains nick and/or realname. Find users that match and ban them */
for (user_map::const_iterator it = UserListByNick.begin(); it != UserListByNick.end(); ++it)
if (x->GetManager()->Check(it->second, x))
this->Send(it->second, x);
return;
}
XLine *old = x;
if (old->GetManager()->HasEntry("*@" + u->host))
return;
/* We can't akill x as it has a nick and/or realname included, so create a new akill for *@host */
x = Serialize::New<XLine *>();
x->SetMask("*@" + u->host);
x->SetBy(old->GetBy());
x->SetExpires(old->GetExpires());
x->SetReason(old->GetReason());
x->SetID(old->GetID());
old->GetManager()->AddXLine(x);
Log(Config->GetClient("OperServ"), "akill") << "AKILL: Added an akill for " << x->GetMask() << " because " << u->GetMask() << "#" << u->realname << " matches " << old->GetMask();
}
/* ZLine if we can instead */
if (x->GetUser() == "*")
{
cidr a(x->GetHost());
if (a.valid())
{
IRCD->Send<messages::SZLine>(u, x);
return;
}
}
// Calculate the time left before this would expire, capping it at 2 days
time_t timeleft = x->GetExpires() - Anope::CurTime;
if (timeleft > 172800)
timeleft = 172800;
Uplink::Send("AKILL", x->GetHost(), x->GetUser(), timeleft, x->GetBy(), Anope::CurTime, x->GetReason());
}
//-------------------------------------------------------------------------
int labelNGram(Config& config)
{
if (config.existsParam("debug"))debug=true; else debug=false;
if (config.existsParam("verbose"))verbose=true; else verbose=false;
String extOutputLabel=".sym.lbl"; // the extension of the output files
if (config.existsParam("saveLabelFileExtension")) extOutputLabel=config.getParam("saveLabelFileExtension");
String pathOutput="./"; // the path of the output files
if (config.existsParam("labelOutputPath")) pathOutput=config.getParam("labelOutputPath");
String extSymbol=".sym"; // the extension of the symbol files
if (config.existsParam("symbolFileExtension")) extSymbol=config.getParam("symbolFileExtension");
String pathSymbol="./";
if (config.existsParam("symbolPath")) pathSymbol=config.getParam("symbolPath");
String formatSymbol="ascii";
if (config.existsParam("symbolFormat")) pathSymbol=config.getParam("symbolFormat");
String NGramFilename=config.getParam("NGramFilename");
unsigned long NGramOrder=3;
if (config.existsParam("NGramOrder")) NGramOrder=config.getParam("NGramOrder").toLong();
unsigned long NGramSelected=16;
if (config.existsParam("NGramSelected")) NGramSelected=config.getParam("NGramSelected").toLong();
NGram NGramTable(NGramOrder,NGramSelected);
NGramTable.load(NGramFilename,config); // Load the NGRAM table, selecting the NGramSelected first
String inputFilename=config.getParam("inputFilename");
String labelSelectedFrames=config.getParam("labelSelectedFrames");
XLine inputFileList;
try{
if (inputFilename.endsWith(".lst")){ // input is file containing a list of filenames
XList tmp(inputFilename,config);
inputFileList=tmp.getAllElements();
}
else inputFileList.addElement(inputFilename); // a single filename
String *p;
while ((p=inputFileList.getElement())){
String& filename=*p;
if (verbose)
cout <<"labelNGram file["<<filename<<"] Table["<<NGramFilename<<"] Order["<<NGramOrder<<"] Selected["<<NGramSelected<<"]"<<endl;
SegServer segServer;
LabelServer labelServer;
loadClusterFile(filename,segServer,labelServer,config);
long codeSelectedFrame=labelServer.getLabelIndexByString(labelSelectedFrames); // Get the index of the selected cluster
if (codeSelectedFrame==-1){ // No data for this model !!!!!!!!!!!!!!
cout << " WARNING - NO DATA with the label["<<labelSelectedFrames<<"] in file ["<<filename<<"]"<<endl;
exit(0);
}
SegCluster& cluster=segServer.getCluster(codeSelectedFrame); // Gives the cluster of the selected/used segments
ULongVector tabS;
unsigned long nbSym=loadSymbol(pathSymbol+filename+extSymbol,formatSymbol,tabS,config); // Read the stream of symbols
SegServer segServerOutput;
SegCluster& clusterOut=segServerOutput.createCluster(0,labelSelectedFrames,cluster.sourceName());
//
computeLabelNGram(NGramTable,cluster,clusterOut,tabS,nbSym);
//
if (verbose){
cout <<"File["<<filename<<"]" <<endl;
cout << "Output the new label file in ["<<pathOutput+filename+extOutputLabel <<"]"<<endl;
}
outputLabelFile(clusterOut,pathOutput+filename+extOutputLabel,config);
} // end file loop
} // fin try
catch (Exception& e)
{
cout << e.toString().c_str() << endl;
}
return 0;
}
int saveApost(Config &config)
{
bool writeAllFeature=true; // Output a vector for all input vectors (selected and not selected vectors) - DEFAULT=on
if (config.existsParam("writeAllFeatures")) writeAllFeature=config.getParam("writeAllFeatures").toBool(); // Define if all the feature (selected or not) should be written
String modelname = config.getParam("inputModelFilename");
String inputFeatureFileName =config.getParam("inputFeatureFilename"); // input feature - could be a simple feature file or a list of filenames
XLine inputFeatureFileNameList; // The (feature) input filename list
if (inputFeatureFileName.endsWith(".lst")){ // If the file parameter is the name of a XList file
XList inputFileNameXList(inputFeatureFileName,config); // Read the filename list file
inputFeatureFileNameList=inputFileNameXList.getAllElements(); // And put the filename in a list if the file is a list of feature filenames
}
else { // It was a simple feature file and not a filename list
inputFeatureFileNameList.addElement(inputFeatureFileName); // add the filename in the list
}
try{
// read UBM
MixtureServer _ms(config);
StatServer _ss(config);
_ms.loadMixtureGD(config.getParam("inputWorldFilename"));
MixtureGD & UBM=_ms.getMixtureGD((unsigned long) 0);
MixtureGDStat &acc=_ss.createAndStoreMixtureStat(UBM);
unsigned long _vsize=UBM.getVectSize();
unsigned long _mixsize=UBM.getDistribCount();
// Loop over the list of feature files
String *file;
String labelSelectedFrames;
unsigned long codeSelectedFrame;
while ((file=inputFeatureFileNameList.getElement())!= NULL){
String & featureFilename=(*file);
FeatureServer fs(config,featureFilename);
FeatureServer fs_out(config,featureFilename);
SegServer segmentsServer;
LabelServer labelServer;
initializeClusters(featureFilename,segmentsServer,labelServer,config);
verifyClusterFile(segmentsServer,fs,config);
labelSelectedFrames=config.getParam("labelSelectedFrames");
codeSelectedFrame=labelServer.getLabelIndexByString(labelSelectedFrames);
SegCluster& selectedSegments=segmentsServer.getCluster(codeSelectedFrame);
// Compute Occupations and Statistics
acc.resetOcc();
Seg *seg;
selectedSegments.rewind();
String currentSource="";
while((seg=selectedSegments.getSeg())!=NULL){
unsigned long begin=seg->begin()+fs.getFirstFeatureIndexOfASource(seg->sourceName()); // Idx of the first frame of the current file in the feature server
if (currentSource!=seg->sourceName()) {
currentSource=seg->sourceName();
if (verbose)cout << "Processing speaker["<<currentSource<<"]"<< endl;
}
fs.seekFeature(begin);
Feature f;
for (unsigned long idxFrame=0;idxFrame<seg->length();idxFrame++){
fs.readFeature(f);
acc.computeAndAccumulateOcc(f);
RealVector <double> aPost=acc.getOccVect();
Feature tmpF;
for(unsigned long k=0;k<_mixsize;k++) {
tmpF[k]=aPost[k];
}
fs_out.addFeature(f);
}
}
// Writing apost probabilities to file
cout << "Writing to: " << featureFilename << endl;
FeatureFileWriter w(featureFilename, config); // build a featurefile writer to output the features (real features)
SegServer fakeSegServer;
if (writeAllFeature) { // Output all the features- feature count id the same SegServer fakeSegServer; // Create a new fake segment server
fakeSegServer.createCluster(0); // Create a new cluster
SegCluster& fakeSeg=fakeSegServer.getCluster(0); // Get the cluster
fakeSeg.add(fakeSegServer.createSeg(0,fs_out.getFeatureCount(),codeSelectedFrame, labelSelectedFrames,featureFilename)); // Add a segment with all the features
outputFeatureFile(config,fs_out,fakeSeg,w); // output all the features - giving the same file length
}
else
outputFeatureFile(config,fs_out,selectedSegments, w); // Output only the selected features - giving a shorter output
}
}
catch (Exception& e){cout << e.toString().c_str() << endl;}
return 0;
}
// Training of client Speakers
// Input: Xlist Format: ID_Client Seg1 Seg2 ..
// Output: ALIZE_MixtureServer (binaire) + GMM / Client (binary)
int TrainTarget(Config& config)
{
String inputClientListFileName = config.getParam("targetIdList");
String inputWorldFilename = config.getParam("inputWorldFilename");
String outputSERVERFilename = "";
if (config.existsParam("mixtureServer")) outputSERVERFilename =config.getParam("mixtureServer");
bool initByClient=false; // In this case, the init model is read from the file
if (config.existsParam("initByClient")) initByClient=config.getParam("initByClient").toBool();
bool saveEmptyModel=false;
if (config.existsParam("saveEmptyModel")) saveEmptyModel=config.getParam("saveEmptyModel").toBool();
// label for selected frames - Only the frames associated with this label, in the label files, will be used
bool fixedLabelSelectedFrame=true;
String labelSelectedFrames;
if (config.existsParam("useIdForSelectedFrame")) // the ID of each speaker is used as labelSelectedFrame ?
fixedLabelSelectedFrame=(config.getParam("useIdForSelectedFrame").toBool()==false);
if (fixedLabelSelectedFrame) // the label is decided by the command line and is unique for the run
labelSelectedFrames=config.getParam("labelSelectedFrames");
bool modelData=false;
if (config.existsParam("useModelData")) modelData=config.getParam("useModelData").toBool();
String initModelS=inputWorldFilename;
if (modelData) if (config.existsParam("initModel")) initModelS=config.getParam("initModel"); // Use a specific model for Em init
bool outputAdaptParam=false;
if (config.existsParam("superVector")) outputAdaptParam=true;
bool NAP=false;
Matrix <double> ChannelMatrix;
if (config.existsParam("NAP")) {
if (verbose) cout<< "Removing channel effect with NAP from " << config.getParam("NAP") << " of size: [";
NAP=true; // enable NAP
ChannelMatrix.load(config.getParam("NAP"),config); //get Channel Matrix from args and load in a Matrix object
if (verbose) cout << ChannelMatrix.rows() << "," <<ChannelMatrix.cols() << "]" << endl;
}
bool saveCompleteServer=false;
try{
XList inputClientList(inputClientListFileName,config); // read the Id + filenames for each client
XLine * linep;
inputClientList.getLine(0);
MixtureServer ms(config);
StatServer ss(config, ms);
if (verbose) cout << "TrainTarget - Load world model [" << inputWorldFilename<<"]"<<endl;
MixtureGD& world = ms.loadMixtureGD(inputWorldFilename);
MixtureGD& initModel =ms.loadMixtureGD(initModelS);
if (verbose) cout <<"Use["<<initModelS<<"] for initializing EM"<<endl;
// *********** Target loop *****************
while ((linep=inputClientList.getLine()) != NULL){ // linep gives the XLine with the Id of a given client and the list of files
String *id=linep->getElement(); // Get the Client ID (id)
XLine featureFileListp=linep->getElements(); // Get the list of feature file for the client (end of the line)
if (verbose) cout << "Train model ["<<*id<<"]"<<endl;
if (!fixedLabelSelectedFrame){ // the ID is used as label for selecting the frame
labelSelectedFrames=*id;
if (verbose) cout <<*id<<" is used for label selected frames"<<endl;
}
FeatureServer fs(config,featureFileListp); // Reading the features (from several files)
SegServer segmentsServer; // Create the segment server for managing the segments/clusters
LabelServer labelServer; // Create the lable server, for indexing the segments/clusters
initializeClusters(featureFileListp,segmentsServer,labelServer,config); // Reading the segmentation files for each feature input file
verifyClusterFile(segmentsServer,fs,config); // Verify if the segments ending before the end of the feature files...
MixtureGD & adaptedMixture = ms.duplicateMixture(world,DUPL_DISTRIB); // Creating final as a copy of the world model
MixtureGD & clientMixture= ms.duplicateMixture(world,DUPL_DISTRIB);
if (initByClient){ // During trainig data statistic estimation by EM,
clientMixture= ms.loadMixtureGD(*id); // the client model is used for initalization
adaptedMixture=clientMixture;
}
long codeSelectedFrame=labelServer.getLabelIndexByString(labelSelectedFrames); // Get the index of the cluster with in interest audio segments
if (codeSelectedFrame==-1){ // No data for this model !!!!!!!!!!!!!!
cout << " WARNING - NO DATA FOR TRAINING ["<<*id<<"]";
if (saveEmptyModel){
cout <<" World model is returned"<<endl; // In this case, the client model is the world model
if (verbose) cout << "Save client model ["<<*id<<"]" << endl;
adaptedMixture.save(*id, config); // Save the client model
}
}
else{
SegCluster& selectedSegments=segmentsServer.getCluster(codeSelectedFrame); // Gives the cluster of the selected/used segments
if (!initByClient) ms.setMixtureId(clientMixture,*id); // Set the client model Id
if (modelData) modelBasedadaptModel(config,ss,ms,fs,selectedSegments,world,clientMixture,initModel); // EM algo with MAP criterion
else adaptModel(config,ss,ms,fs,selectedSegments,world,clientMixture); // EM algo with MAP criterion
if (NAP) {
if (verbose) cout << "NAP on SVs" << endl;
computeNap(clientMixture,ChannelMatrix);
}
if (outputAdaptParam) {
RealVector<double> v;
getSuperVector(v,world,clientMixture,config);
String out=config.getParam("saveVectorFilesPath")+*id+config.getParam("vectorFilesExtension");
Matrix <double> vv=(Matrix<double>)v;
vv.save(out,config);
}
if (!outputAdaptParam) {
if (verbose) cout << "Save client model ["<<*id<<"]" << endl;
clientMixture.save(*id, config); // Save the client model
}
if (!saveCompleteServer){
long tid=ms.getMixtureIndex(*id); // TO BE SUPPRESSED BY
ms.deleteMixtures(tid,tid); // ADDING a delete on a mixture pointor
ms.deleteUnusedDistribs();
}
}
} // end of the the target loop
// Save the complete mixture server
// TODO
} // fin try
catch (Exception& e)
{
cout << e.toString().c_str() << endl;
}
return 0;
}
// Training of client Speakers
// The same than TrainTarget but train simultaneoulsy 1 model for each cluster (set of segments with the same label)
// found in the input files labels.
// One option in order to save the n models as a modification of the world model - save disk space
int TrainTargetByLabel(Config& config)
{
String inputClientListFileName = config.getParam("targetIdList");
String inputWorldFilename = config.getParam("inputWorldFilename");
String outputSERVERFilename = config.getParam("mixtureServer");
// label for selected frames - Only the frames associated with this label, in the label files, will be used
//bool fixedLabelSelectedFrame;
bool initByClient=false;
bool aprioriWorld=true;
if (config.existsParam("initByClient")) initByClient=true;
if (config.existsParam("aprioriClient")){
aprioriWorld=false;
initByClient=true;
}
bool saveCompleteServer=false;
bool outputAdaptParam=false;
if (config.existsParam("outputAdaptParam")) outputAdaptParam=config.getParam("outputAdaptParam").toBool();
try{
XList inputClientList(inputClientListFileName,config); // read the Id + filenames for each client
XLine *linep;
inputClientList.getLine(0);
MixtureServer ms(config);
StatServer ss(config, ms);
if (verbose) cout << "TrainTarget - by label opption - Load world model [" << inputWorldFilename<<"]"<<endl;
MixtureGD& world = ms.loadMixtureGD(inputWorldFilename);
// *********** Target loop *****************
while ((linep=inputClientList.getLine()) != NULL){ // linep gives the XLine with the Id of a given client and the list of files
String clientId=(*linep->getElement()); // Get the Client ID (clientId)
XLine featureFileListp=linep->getElements(); // Get the list of feature file for the client (end of the line)
FeatureServer fs(config,featureFileListp); // Reading the features (from several files)
if (verbose) cout << "Train label models for client ["<<clientId<<"]"<<endl;
MixtureGD &clientGModel=ms.createMixtureGD();
if (initByClient) {
if (verbose) cout << "Load client model [" << clientId <<"]"<<endl;
clientGModel = ms.loadMixtureGD(clientId); //not necessary to load client model
}
SegServer segmentsServer; // Create the segment server for managing the segments/clusters
LabelServer labelServer; // Create the lable server, for indexing the segments/clusters
initializeClusters(featureFileListp,segmentsServer,labelServer,config); // Reading the segmentation files for each feature input file
verifyClusterFile(segmentsServer,fs,config); // Verify if the segments ending before the end of the feature files...
for (unsigned long codeSelectedFrame=0;codeSelectedFrame<segmentsServer.getClusterCount();codeSelectedFrame++){ // For each cluster
String clientIdByLabel=clientId+"_"+labelServer.getLabel(codeSelectedFrame).getString(); // Build the model name for the client and the label
if (verbose) cout << "Train labeldependent model ["<<clientIdByLabel<<"]"<<endl;
SegCluster& selectedSegments=segmentsServer.getCluster(codeSelectedFrame); // Gives the cluster of the selected/used segments
MixtureGD & clientMixture = ms.duplicateMixture(world,DUPL_DISTRIB); // Creating clientMixture as a copy of the world model
ms.setMixtureId(clientMixture,clientIdByLabel); // Set the client model Id
if (initByClient) // During trainig data statistic estimation by EM,
clientMixture=clientGModel; // the global client model is used for initalization
if (aprioriWorld) // EM algo with MAP criterion
adaptModel(config,ss,ms,fs,selectedSegments,world,clientMixture); // A priori info is the world model
else adaptModel(config,ss,ms,fs,selectedSegments,clientGModel,clientMixture);// A priori info is the client model-by default initByClient is also set
if (!outputAdaptParam) {
if (verbose) cout << "Save client model ["<<clientIdByLabel<<"]" << endl;
clientMixture.save(clientIdByLabel, config); // Save the client model
}
if (!saveCompleteServer){
long tid=ms.getMixtureIndex(clientIdByLabel); // TO BE SUPPRESSED BY
ms.deleteMixtures(tid,tid); // ADDING a delete on a mixture pointor
ms.deleteUnusedDistribs();
}
}
if (!saveCompleteServer){
long tid=ms.getMixtureIndex(clientId); // TO BE SUPPRESSED BY
ms.deleteMixtures(tid,tid); // ADDING a delete on a mixture pointor
ms.deleteUnusedDistribs();
} // end of the the label loop fr a speaker
} // end of the the target loop
// Save the complete mixture server
// TODO
} // fin try
catch (Exception& e)
{
cout << e.toString().c_str() << endl;
}
return 0;
}
//-----------------------------------------------------------------------------------------------------------------------------------------------------------
int TrainTargetFA(Config& config)
{
String inputClientListFileName = config.getParam("targetIdList");
String inputWorldFilename = config.getParam("inputWorldFilename");
String outputSERVERFilename = "";
if (config.existsParam("mixtureServer")) outputSERVERFilename =config.getParam("mixtureServer");
bool initByClient=false; // In this case, the init model is read from the file
if (config.existsParam("initByClient")) initByClient=config.getParam("initByClient").toBool();
bool saveEmptyModel=false;
if (config.existsParam("saveEmptyModel")) saveEmptyModel=config.getParam("saveEmptyModel").toBool();
// label for selected frames - Only the frames associated with this label, in the label files, will be used
bool fixedLabelSelectedFrame=true;
String labelSelectedFrames;
if (config.existsParam("useIdForSelectedFrame")) // the ID of each speaker is used as labelSelectedFrame ?
fixedLabelSelectedFrame=(config.getParam("useIdForSelectedFrame").toBool()==false);
if (fixedLabelSelectedFrame) // the label is decided by the command line and is unique for the run
labelSelectedFrames=config.getParam("labelSelectedFrames");
bool modelData=false;
if (config.existsParam("useModelData")) modelData=config.getParam("useModelData").toBool();
String initModelS=inputWorldFilename;
if (modelData) if (config.existsParam("initModel")) initModelS=config.getParam("initModel"); // Use a specific model for Em init
bool outputAdaptParam=false;
if (config.existsParam("superVectors")) outputAdaptParam=true;
Matrix <double> ChannelMatrix;
if (verbose) cout<< "EigenMAP and Eigenchannel with [" << config.getParam("initChannelMatrix") << "] of size: [";
ChannelMatrix.load(config.getParam("initChannelMatrix"),config); //get Channel Matrix from args and load in a Matrix object
if (verbose) cout << ChannelMatrix.rows() << "," <<ChannelMatrix.cols() << "]" << endl;
bool varAdapt=false;
if (config.existsParam("FAVarAdapt")) varAdapt=true;
bool saveCompleteServer=false;
try{
XList inputClientList(inputClientListFileName,config); // read the Id + filenames for each client
XLine * linep;
inputClientList.getLine(0);
MixtureServer ms(config);
StatServer ss(config, ms);
if (verbose) cout << "(TrainTarget) Factor Analysis - Load world model [" << inputWorldFilename<<"]"<<endl;
MixtureGD& world = ms.loadMixtureGD(inputWorldFilename);
if (verbose) cout <<"(TrainTarget) Use["<<initModelS<<"] for initializing EM"<<endl;
// *********** Target loop *****************
while ((linep=inputClientList.getLine()) != NULL){ // linep gives the XLine with the Id of a given client and the list of files
String *id=linep->getElement(); // Get the Client ID (id)
XLine featureFileListp=linep->getElements(); // Get the list of feature file for the client (end of the line)
if (verbose) cout << "(TrainTarget) Train model ["<<*id<<"]"<<endl;
FeatureServer fs(config,featureFileListp); // Reading the features (from several files)
SegServer segmentsServer; // Create the segment server for managing the segments/clusters
LabelServer labelServer; // Create the lable server, for indexing the segments/clusters
initializeClusters(featureFileListp,segmentsServer,labelServer,config); // Reading the segmentation files for each feature input file
verifyClusterFile(segmentsServer,fs,config); // Verify if the segments ending before the end of the feature files...
MixtureGD & adaptedMixture = ms.duplicateMixture(world,DUPL_DISTRIB); // Creating final as a copy of the world model
MixtureGD & clientMixture= ms.duplicateMixture(world,DUPL_DISTRIB);
long codeSelectedFrame=labelServer.getLabelIndexByString(labelSelectedFrames); // Get the index of the cluster with in interest audio segments
if (codeSelectedFrame==-1){ // No data for this model !!!!!!!!!!!!!!
cout << " WARNING - NO DATA FOR TRAINING ["<<*id<<"]";
if (saveEmptyModel){
cout <<" World model is returned"<<endl; // In this case, the client model is the world model
if (verbose) cout << "Save client model ["<<*id<<"]" << endl;
adaptedMixture.save(*id, config); // Save the client model
}
}
else{
SegCluster& selectedSegments=segmentsServer.getCluster(codeSelectedFrame); // Gives the cluster of the selected/used segments
/// **** Factor Analysis Stuff
XList faNdx;
faNdx.addLine()=featureFileListp;
FactorAnalysisStat FA(faNdx,fs,config); // give all features to FA stats
//FA.computeAndAccumulateGeneralFAStats(selectedSegments,fs,config);
for(int i=0;i<config.getParam("nbTrainIt").toLong();i++){
if (verbose) cout << "------ Iteration ["<<i<<"] ------"<<endl;
FA.computeAndAccumulateGeneralFAStats(selectedSegments,fs,config);
/*if (!varAdapt) FA.getTrueSpeakerModel(clientMixture,linep->getElement(1));
else FA.getFactorAnalysisModel(clientMixture,linep->getElement(1));
if (verbose) cout << "LLK for model["<<*id<<"] at it["<<i-1<<"]="<<FA.getLLK(selectedSegments,clientMixture,fs,config) << endl; */
FA.estimateAndInverseL(config);
FA.substractSpeakerStats();
FA.getXEstimate();
FA.substractChannelStats();
FA.getYEstimate();
}
MixtureGD & sessionMixture= ms.duplicateMixture(world,DUPL_DISTRIB);
bool saveSessionModel=false;
if (config.existsParam("saveSessionModel")) saveSessionModel=true;
if (saveSessionModel) FA.getSessionModel(sessionMixture,linep->getElement(1));
if (!varAdapt) FA.getTrueSpeakerModel(clientMixture,linep->getElement(1)); // basically compute M_s_h=M+Dy_s and get a model
else FA.getFactorAnalysisModel(clientMixture,linep->getElement(1)); // get FA variance adapted model
if (verbose) cout << "Final LLK for model["<<*id<<"]="<<FA.getLLK(selectedSegments,clientMixture,fs,config) << endl;
/// **** End of FA
if (!outputAdaptParam) {
if (verbose) cout << "Save client model ["<<*id<<"]" << endl;
clientMixture.save(*id, config); // Save the client model
if (saveSessionModel) {
String sessionfile=*id+".session";
if (verbose) cout << "Save session model ["<<sessionfile<<"]" << endl;
sessionMixture.save(sessionfile,config);
}
}
if (!saveCompleteServer){
long tid=ms.getMixtureIndex(*id); // TO BE SUPPRESSED BY
ms.deleteMixtures(tid,tid); // ADDING a delete on a mixture pointor
ms.deleteUnusedDistribs();
}
}
}
} // fin try
catch (Exception& e) {cout << e.toString().c_str() << endl;}
return 0;
}
//-----------------------------------------------------------------------------------------------------------------------------------------------------------
int TrainTargetJFA(Config& config)
{
String inputClientListFileName = config.getParam("targetIdList");
String inputWorldFilename = config.getParam("inputWorldFilename");
String outputSERVERFilename = "";
if (config.existsParam("mixtureServer")) outputSERVERFilename =config.getParam("mixtureServer");
bool initByClient=false; // In this case, the init model is read from the file
if (config.existsParam("initByClient")) initByClient=config.getParam("initByClient").toBool();
bool saveEmptyModel=false;
if (config.existsParam("saveEmptyModel")) saveEmptyModel=config.getParam("saveEmptyModel").toBool();
// label for selected frames - Only the frames associated with this label, in the label files, will be used
bool fixedLabelSelectedFrame=true;
String labelSelectedFrames;
if (config.existsParam("useIdForSelectedFrame")) // the ID of each speaker is used as labelSelectedFrame ?
fixedLabelSelectedFrame=(config.getParam("useIdForSelectedFrame").toBool()==false);
if (fixedLabelSelectedFrame) // the label is decided by the command line and is unique for the run
labelSelectedFrames=config.getParam("labelSelectedFrames");
bool modelData=false;
if (config.existsParam("useModelData")) modelData=config.getParam("useModelData").toBool();
String initModelS=inputWorldFilename;
if (modelData) if (config.existsParam("initModel")) initModelS=config.getParam("initModel"); // Use a specific model for Em init
bool outputAdaptParam=false;
if (config.existsParam("superVectors")) outputAdaptParam=true;
try{
XList inputClientList(inputClientListFileName,config); // read the Id + filenames for each client
XLine * linep;
inputClientList.getLine(0);
MixtureServer ms(config);
StatServer ss(config, ms);
if (verbose) cout << "(TrainTarget) Joint Factor Analysis - Load world model [" << inputWorldFilename<<"]"<<endl;
MixtureGD& world = ms.loadMixtureGD(inputWorldFilename);
if (verbose) cout <<"(TrainTarget) Use["<<initModelS<<"] for initializing EM"<<endl;
//LOAD JFA MAtrices
Matrix<double> U, V;
DoubleVector D;
//Initialise EC matrix
if(config.existsParam("eigenChannelMatrix")){
String uName = config.getParam("matrixFilesPath") + config.getParam("eigenChannelMatrix") + config.getParam("loadMatrixFilesExtension");
U.load (uName, config);
if (verboseLevel >=1) cout << "(TrainTargetJFA) Init EC matrix from "<< config.getParam("eigenChannelMatrix") <<" from EigenChannel Matrix: "<<", rank: ["<<U.rows() << "] sv size: [" << U.cols() <<"]"<<endl;
}
else{
unsigned long sS = world.getVectSize() * world.getDistribCount();
U.setDimensions(1,sS);
U.setAllValues(0.0);
if (verboseLevel >=1) cout << "(TrainTargetJFA) Init EC matrix to 0"<<endl;
}
//Initialise EV matrix
if(config.existsParam("eigenVoiceMatrix")){
String vName = config.getParam("matrixFilesPath") + config.getParam("eigenVoiceMatrix") + config.getParam("loadMatrixFilesExtension");
V.load (vName, config);
if (verboseLevel >=1) cout << "(TrainTargetJFA) Init EV matrix from "<< config.getParam("eigenVoiceMatrix") <<" from EigenVoice Matrix: "<<", rank: ["<<V.rows() << "] sv size: [" << V.cols() <<"]"<<endl;
}
else{
unsigned long sS = world.getVectSize() * world.getDistribCount();
V.setDimensions(1,sS);
V.setAllValues(0.0);
if (verboseLevel >=1) cout << "(TrainTargetJFA) Init EV matrix to 0"<<endl;
}
//Initialise D matrix
if(config.existsParam("DMatrix")){
String dName = config.getParam("matrixFilesPath") + config.getParam("DMatrix") + config.getParam("loadMatrixFilesExtension");
Matrix<double> tmpD(dName, config);
if( (tmpD.rows() != 1) || ( tmpD.cols() != world.getVectSize()*world.getDistribCount() ) ){
throw Exception("Incorrect dimension of D Matrix",__FILE__,__LINE__);
}
else{
D.setSize(world.getVectSize()*world.getDistribCount());
D.setAllValues(0.0);
for(unsigned long i=0; i<world.getVectSize()*world.getDistribCount(); i++){
D[i] = tmpD(0,i);
}
if (verboseLevel >=1) cout << "(TrainTargetJFA) Init D matrix from "<<config.getParam("DMatrix")<<endl;
}
}
else{
unsigned long sS = world.getVectSize() * world.getDistribCount();
D.setSize(sS);
D.setAllValues(0.0);
if (verboseLevel >1) cout << "(TrainTargetJFA) Init D matrix to 0"<<endl;
}
// *********** Target loop *****************
while ((linep=inputClientList.getLine()) != NULL){ // linep gives the XLine with the Id of a given client and the list of files
String *id=linep->getElement(); // Get the Client ID (id)
XLine featureFileListp=linep->getElements(); // Get the list of feature file for the client (end of the line)
if (verbose) cout << "(TrainTarget) Train model ["<<*id<<"]"<<endl;
XList ndx; ndx.addLine() = featureFileListp;
JFAAcc jfaAcc(ndx,config,"TrainTarget");
//Load V, U and D from existing matrices.
jfaAcc.loadEV(V, config); jfaAcc.loadEC(U, config); jfaAcc.loadD(D);
//Initialize VU matrix
jfaAcc.initVU();
FeatureServer fs(config,featureFileListp); // Reading the features (from several files)
SegServer segmentsServer; // Create the segment server for managing the segments/clusters
LabelServer labelServer; // Create the lable server, for indexing the segments/clusters
initializeClusters(featureFileListp,segmentsServer,labelServer,config); // Reading the segmentation files for each feature input file
verifyClusterFile(segmentsServer,fs,config); // Verify if the segments ending before the end of the feature files...
MixtureGD & adaptedMixture = ms.duplicateMixture(world,DUPL_DISTRIB); // Creating final as a copy of the world model
MixtureGD & clientMixture= ms.duplicateMixture(world,DUPL_DISTRIB);
long codeSelectedFrame=labelServer.getLabelIndexByString(labelSelectedFrames); // Get the index of the cluster with in interest audio segments
if (codeSelectedFrame==-1){ // No data for this model !!!!!!!!!!!!!!
cout << " WARNING - NO DATA FOR TRAINING ["<<*id<<"]";
if (saveEmptyModel){
cout <<" World model is returned"<<endl; // In this case, the client model is the world model
if (verbose) cout << "Save client model ["<<*id<<"]" << endl;
adaptedMixture.save(*id, config); // Save the client model
}
}
else{
SegCluster& selectedSegments=segmentsServer.getCluster(codeSelectedFrame); // Gives the cluster of the selected/used segments
//Compute the JFA statistics
jfaAcc.computeAndAccumulateJFAStat(selectedSegments,fs,config);
//Estimate X and Y in one time for each speaker
jfaAcc.storeAccs();
jfaAcc.estimateVUEVUT(config);
jfaAcc.estimateAndInverseL_VU(config);
jfaAcc.substractMplusDZ(config);
jfaAcc.estimateYX();
//Reinitialise the accumulators
jfaAcc.resetTmpAcc();
jfaAcc.restoreAccs();
//Split X and Y estimates
jfaAcc.splitYX();
//Substract speaker and channel statistics M + VUYX
jfaAcc.substractMplusVUYX();
//Estimate Z for each speaker
jfaAcc.estimateZ();
//Reinitialise the accumulators
jfaAcc.resetTmpAcc();
jfaAcc.restoreAccs();
bool varAdapt = false;
if((config.existsParam("varAdapt")) && ( config.getParam("varAdapt").toBool() )){
varAdapt = true;
}
DoubleVector clientSV(jfaAcc.getSvSize(), jfaAcc.getSvSize());
clientSV.setSize(jfaAcc.getSvSize());
DoubleVector clientModel(jfaAcc.getSvSize(), jfaAcc.getSvSize());
clientModel.setSize(jfaAcc.getSvSize());
bool saveMixture = true;
if((config.existsParam("saveMixture")) && !( config.getParam("saveMixture").toBool() )) saveMixture = false;
bool saveSuperVector = true;
if((config.existsParam("saveSuperVector")) && !( config.getParam("saveSuperVector").toBool() )) saveSuperVector = false;
bool saveX = false;
bool saveY = false;
bool saveZ = false;
if(config.existsParam("saveX")) saveX = config.getParam("saveX").toBool();
if(config.existsParam("saveY")) saveY = config.getParam("saveY").toBool();
if(config.existsParam("saveZ")) saveZ = config.getParam("saveZ").toBool();
String xExtension = ".x"; String yExtension = ".y"; String zExtension = ".z";
if(config.existsParam("xExtension")) xExtension = config.getParam("xExtension");
if(config.existsParam("yExtension")) yExtension = config.getParam("yExtension");
if(config.existsParam("zExtension")) zExtension = config.getParam("zExtension");
jfaAcc.getVYplusDZ(clientSV, 0);
jfaAcc.getMplusVYplusDZ(clientModel, 0);
//WARNING !!!!! only the SuperVector model is divided by the UBM Co-Variance.
for(unsigned long i=0; i<jfaAcc.getSvSize(); i++){
clientSV[i] *= jfaAcc.getUbmInvVar()[i];
}
//Create the ClientMixture to save if required
if(saveMixture){
svToModel(clientModel, clientMixture);
clientMixture.save(*id, config);
}
if(saveSuperVector){
String svPath=config.getParam("saveVectorFilesPath");
String svExt=config.getParam("vectorFilesExtension");
String svFile=svPath+*id+svExt;
((Matrix<double>)clientSV).save(svFile,config);
}
String svPath=config.getParam("saveVectorFilesPath");
if(saveX){
String xFile=svPath+*id+xExtension;
jfaAcc.saveX(xFile,config);
}
if(saveY){
String yFile=svPath+*id+yExtension;
jfaAcc.saveY(yFile,config);
}
if(saveZ){
String zFile=svPath+*id+zExtension;
jfaAcc.saveZ(zFile,config);
}
long tid=ms.getMixtureIndex(*id);
ms.deleteMixtures(tid,tid);
ms.deleteUnusedDistribs();
}
}
} // fin try
catch (Exception& e) {cout << e.toString().c_str() << endl;}
return 0;
}
//-----------------------------------------------------------------------------------------------------------------------------------------------------------
int TrainTargetLFA(Config& config)
{
String inputClientListFileName = config.getParam("targetIdList");
String inputWorldFilename = config.getParam("inputWorldFilename");
String outputSERVERFilename = "";
if (config.existsParam("mixtureServer")) outputSERVERFilename =config.getParam("mixtureServer");
bool initByClient=false; // In this case, the init model is read from the file
if (config.existsParam("initByClient")) initByClient=config.getParam("initByClient").toBool();
bool saveEmptyModel=false;
if (config.existsParam("saveEmptyModel")) saveEmptyModel=config.getParam("saveEmptyModel").toBool();
// label for selected frames - Only the frames associated with this label, in the label files, will be used
bool fixedLabelSelectedFrame=true;
String labelSelectedFrames;
if (config.existsParam("useIdForSelectedFrame")) // the ID of each speaker is used as labelSelectedFrame ?
fixedLabelSelectedFrame=(config.getParam("useIdForSelectedFrame").toBool()==false);
if (fixedLabelSelectedFrame) // the label is decided by the command line and is unique for the run
labelSelectedFrames=config.getParam("labelSelectedFrames");
bool modelData=false;
if (config.existsParam("useModelData")) modelData=config.getParam("useModelData").toBool();
String initModelS=inputWorldFilename;
if (modelData) if (config.existsParam("initModel")) initModelS=config.getParam("initModel"); // Use a specific model for Em init
bool outputAdaptParam=false;
if (config.existsParam("superVectors")) outputAdaptParam=true;
try{
XList inputClientList(inputClientListFileName,config); // read the Id + filenames for each client
XLine * linep;
inputClientList.getLine(0);
MixtureServer ms(config);
StatServer ss(config, ms);
if (verbose) cout << "(TrainTarget) Latent Factor Analysis - Load world model [" << inputWorldFilename<<"]"<<endl;
MixtureGD& world = ms.loadMixtureGD(inputWorldFilename);
if (verbose) cout <<"(TrainTarget) Use["<<initModelS<<"] for initializing EM"<<endl;
//LOAD JFA MAtrices
unsigned long svsize=world.getDistribCount()*world.getVectSize();
Matrix<double> U, V;
DoubleVector D(svsize,svsize);
//Initialise EC matrix
if(config.existsParam("eigenChannelMatrix")){
String uName = config.getParam("matrixFilesPath") + config.getParam("eigenChannelMatrix") + config.getParam("loadMatrixFilesExtension");
U.load (uName, config);
if (verboseLevel >=1) cout << "(TrainTargetLFA) Init EC matrix from "<< config.getParam("eigenChannelMatrix") <<" from EigenChannel Matrix: "<<", rank: ["<<U.rows() << "] sv size: [" << U.cols() <<"]"<<endl;
}
else{
U.setDimensions(1,svsize);
U.setAllValues(0.0);
if (verboseLevel >1) cout << "(TrainTargetLFA) Init EC matrix to 0"<<endl;
}
V.setDimensions(1,svsize);
V.setAllValues(0.0);
if (verboseLevel >=1) cout << "(TrainTargetLFA) Init EV matrix to 0"<<endl;
//Initialise the D matrix for MAP adaptation
for(unsigned long i=0; i<world.getDistribCount(); i++){
for(unsigned long j = 0; j<world.getVectSize(); j++){
D[i*world.getVectSize()+j] = sqrt(1.0/(world.getDistrib(i).getCovInv(j)*config.getParam("regulationFactor").toDouble()));
}
}
// *********** Target loop *****************
while ((linep=inputClientList.getLine()) != NULL){ // linep gives the XLine with the Id of a given client and the list of files
String *id=linep->getElement(); // Get the Client ID (id)
XLine featureFileListp=linep->getElements(); // Get the list of feature file for the client (end of the line)
if (verbose) cout << "(TrainTargetLFA) Train model ["<<*id<<"]"<<endl;
XList ndx; ndx.addLine() = featureFileListp;
JFAAcc jfaAcc(ndx,config,"TrainTarget");
//Charger les matrices V, U et D a partir des objets matrice existant.
jfaAcc.loadEV(V, config); jfaAcc.loadEC(U, config); jfaAcc.loadD(D);
//Initialise VU matrix
jfaAcc.initVU();
FeatureServer fs(config,featureFileListp); // Reading the features (from several files)
SegServer segmentsServer; // Create the segment server for managing the segments/clusters
LabelServer labelServer; // Create the lable server, for indexing the segments/clusters
initializeClusters(featureFileListp,segmentsServer,labelServer,config); // Reading the segmentation files for each feature input file
verifyClusterFile(segmentsServer,fs,config); // Verify if the segments ending before the end of the feature files...
MixtureGD & adaptedMixture = ms.duplicateMixture(world,DUPL_DISTRIB); // Creating final as a copy of the world model
MixtureGD & clientMixture= ms.duplicateMixture(world,DUPL_DISTRIB);
long codeSelectedFrame=labelServer.getLabelIndexByString(labelSelectedFrames); // Get the index of the cluster with in interest audio segments
if (codeSelectedFrame==-1){ // No data for this model !!!!!!!!!!!!!!
cout << " WARNING - NO DATA FOR TRAINING ["<<*id<<"]";
if (saveEmptyModel){
cout <<" World model is returned"<<endl; // In this case, the client model is the world model
if (verbose) cout << "Save client model ["<<*id<<"]" << endl;
adaptedMixture.save(*id, config); // Save the client model
}
}
else{
SegCluster& selectedSegments=segmentsServer.getCluster(codeSelectedFrame); // Gives the cluster of the selected/used segments
//Compute the JFA statistics
jfaAcc.computeAndAccumulateJFAStat(selectedSegments,fs,config);
//Estimate X and Y in one time for each speaker
jfaAcc.storeAccs();
jfaAcc.estimateVUEVUT(config);
jfaAcc.estimateAndInverseL_VU(config);
jfaAcc.substractMplusDZ(config);
jfaAcc.estimateYX();
//Reinitialise the accumulators
jfaAcc.resetTmpAcc();
jfaAcc.restoreAccs();
//Split X and Y estimates
jfaAcc.splitYX();
//Substract speaker and channel statistics M + VUYX
jfaAcc.substractMplusVUYX();
//Estimate Z for each speaker
double tau = config.getParam("regulationFactor").toLong();
jfaAcc.estimateZMAP(tau);
//Reinitialise the accumulators
jfaAcc.resetTmpAcc();
jfaAcc.restoreAccs();
bool varAdapt = false;
if((config.existsParam("varAdapt")) && ( config.getParam("varAdapt").toBool() )){
varAdapt = true;
}
DoubleVector clientModel(jfaAcc.getSvSize(), jfaAcc.getSvSize());
clientModel.setSize(jfaAcc.getSvSize());
jfaAcc.getMplusVYplusDZ(clientModel, 0);
//Create the ClientMixture
svToModel(clientModel, clientMixture);
clientMixture.save(*id, config);
long tid=ms.getMixtureIndex(*id);
ms.deleteMixtures(tid,tid);
ms.deleteUnusedDistribs();
}
}
} // fin try
catch (Exception& e) {cout << e.toString().c_str() << endl;}
return 0;
}
/* add a client connection to the sockets list */
void UserManager::AddUser(int socket, ListenSocket* via, irc::sockets::sockaddrs* client, irc::sockets::sockaddrs* server)
{
/* NOTE: Calling this one parameter constructor for User automatically
* allocates a new UUID and places it in the hash_map.
*/
LocalUser* New = NULL;
try
{
New = new LocalUser(socket, client, server);
}
catch (...)
{
ServerInstance->Logs->Log("USERS", DEFAULT,"*** WTF *** Duplicated UUID! -- Crack smoking monkeys have been unleashed.");
ServerInstance->SNO->WriteToSnoMask('a', "WARNING *** Duplicate UUID allocated!");
return;
}
UserIOHandler* eh = &New->eh;
/* Give each of the modules an attempt to hook the user for I/O */
FOREACH_MOD(I_OnHookIO, OnHookIO(eh, via));
if (eh->GetIOHook())
{
try
{
eh->GetIOHook()->OnStreamSocketAccept(eh, client, server);
}
catch (CoreException& modexcept)
{
ServerInstance->Logs->Log("SOCKET", DEBUG,"%s threw an exception: %s", modexcept.GetSource(), modexcept.GetReason());
}
}
ServerInstance->Logs->Log("USERS", DEBUG,"New user fd: %d", socket);
this->unregistered_count++;
/* The users default nick is their UUID */
New->nick.assign(New->uuid, 0, ServerInstance->Config->Limits.NickMax);
(*(this->clientlist))[New->nick] = New;
New->registered = REG_NONE;
New->signon = ServerInstance->Time() + ServerInstance->Config->dns_timeout;
New->lastping = 1;
ServerInstance->Users->AddLocalClone(New);
ServerInstance->Users->AddGlobalClone(New);
New->localuseriter = this->local_users.insert(local_users.end(), New);
if ((this->local_users.size() > ServerInstance->Config->SoftLimit) || (this->local_users.size() >= (unsigned int)ServerInstance->SE->GetMaxFds()))
{
ServerInstance->SNO->WriteToSnoMask('a', "Warning: softlimit value has been reached: %d clients", ServerInstance->Config->SoftLimit);
this->QuitUser(New,"No more connections allowed");
return;
}
/*
* First class check. We do this again in FullConnect after DNS is done, and NICK/USER is recieved.
* See my note down there for why this is required. DO NOT REMOVE. :) -- w00t
*/
New->SetClass();
/*
* Check connect class settings and initialise settings into User.
* This will be done again after DNS resolution. -- w00t
*/
New->CheckClass();
if (New->quitting)
return;
/*
* even with bancache, we still have to keep User::exempt current.
* besides that, if we get a positive bancache hit, we still won't f**k
* them over if they are exempt. -- w00t
*/
New->exempt = (ServerInstance->XLines->MatchesLine("E",New) != NULL);
if (BanCacheHit *b = ServerInstance->BanCache->GetHit(New->GetIPString()))
{
if (!b->Type.empty() && !New->exempt)
{
/* user banned */
ServerInstance->Logs->Log("BANCACHE", DEBUG, "BanCache: Positive hit for " + New->GetIPString());
if (!ServerInstance->Config->MoronBanner.empty())
New->WriteServ("NOTICE %s :*** %s", New->nick.c_str(), ServerInstance->Config->MoronBanner.c_str());
this->QuitUser(New, b->Reason);
return;
}
else
{
ServerInstance->Logs->Log("BANCACHE", DEBUG, "BanCache: Negative hit for " + New->GetIPString());
}
}
else
{
if (!New->exempt)
{
XLine* r = ServerInstance->XLines->MatchesLine("Z",New);
if (r)
{
r->Apply(New);
return;
}
}
}
if (!ServerInstance->SE->AddFd(eh, FD_WANT_FAST_READ | FD_WANT_EDGE_WRITE))
{
ServerInstance->Logs->Log("USERS", DEBUG,"Internal error on new connection");
this->QuitUser(New, "Internal error handling connection");
}
/* NOTE: even if dns lookups are *off*, we still need to display this.
* BOPM and other stuff requires it.
*/
New->WriteServ("NOTICE Auth :*** Looking up your hostname...");
if (ServerInstance->Config->RawLog)
New->WriteServ("NOTICE Auth :*** Raw I/O logging is enabled on this server. All messages, passwords, and commands are being recorded.");
FOREACH_MOD(I_OnSetUserIP,OnSetUserIP(New));
if (New->quitting)
return;
FOREACH_MOD(I_OnUserInit,OnUserInit(New));
if (ServerInstance->Config->NoUserDns)
{
New->WriteServ("NOTICE %s :*** Skipping host resolution (disabled by server administrator)", New->nick.c_str());
New->dns_done = true;
}
else
{
New->StartDNSLookup();
}
}
