ProgMetadataSplit Class Reference

Inheritance diagram for ProgMetadataSplit:

Collaboration diagram for ProgMetadataSplit:

Public Member Functions
void	run ()
Public Member Functions inherited from XmippProgram
const char *	getParam (const char *param, int arg=0)
const char *	getParam (const char *param, const char *subparam, int arg=0)
int	getIntParam (const char *param, int arg=0)
int	getIntParam (const char *param, const char *subparam, int arg=0)
double	getDoubleParam (const char *param, int arg=0)
double	getDoubleParam (const char *param, const char *subparam, int arg=0)
float	getFloatParam (const char *param, int arg=0)
float	getFloatParam (const char *param, const char *subparam, int arg=0)
void	getListParam (const char *param, StringVector &list)
int	getCountParam (const char *param)
bool	checkParam (const char *param)
bool	existsParam (const char *param)
void	addParamsLine (const String &line)
void	addParamsLine (const char *line)
ParamDef *	getParamDef (const char *param) const
virtual void	quit (int exit_code=0) const
virtual int	tryRun ()
void	initProgress (size_t total, size_t stepBin=60)
void	setProgress (size_t value=0)
void	endProgress ()
void	processDefaultComment (const char *param, const char *left)
void	setDefaultComment (const char *param, const char *comment)
virtual void	initComments ()
void	setProgramName (const char *name)
void	addUsageLine (const char *line, bool verbatim=false)
void	clearUsage ()
void	addExampleLine (const char *example, bool verbatim=true)
void	addSeeAlsoLine (const char *seeAlso)
void	addKeywords (const char *keywords)
const char *	name () const
virtual void	usage (int verb=0) const
virtual void	usage (const String &param, int verb=2)
int	version () const
virtual void	show () const
virtual void	read (int argc, const char **argv, bool reportErrors=true)
virtual void	read (int argc, char **argv, bool reportErrors=true)
void	read (const String &argumentsLine)
	XmippProgram ()
	XmippProgram (int argc, const char **argv)
virtual	~XmippProgram ()

Protected Member Functions
void	defineParams ()
void	readParams ()
Protected Member Functions inherited from XmippProgram
void	defineCommons ()

Additional Inherited Members
Public Attributes inherited from XmippProgram
bool	doRun
bool	runWithoutArgs
int	verbose
	Verbosity level. More...
int	debug
Protected Attributes inherited from XmippProgram
int	errorCode
ProgramDef *	progDef
	Program definition and arguments parser. More...
std::map< String, CommentList >	defaultComments
int	argc
	Original command line arguments. More...
const char **	argv

Detailed Description

Definition at line 33 of file metadata_split.cpp.

Member Function Documentation

defineParams()

void ProgMetadataSplit::defineParams ( )

inlineprotectedvirtual

Function in which the param of each Program are defined.

Reimplemented from XmippProgram.

Definition at line 47 of file metadata_split.cpp.

    {
        addUsageLine("Split a metadata (randomly by default) in any number of equally sized output metadata.");
        addUsageLine("By default, only enabled entries in the input file will be written to the output files.");

        addParamsLine("    -i <inputSelfile>    : Input MetaData File");
        addParamsLine("  [ -n <parts=2> ] : Number of output MetaDatas");
        addParamsLine("  [ --oroot <rootname=\"\"> ] : Rootname for output MetaDatas");
        addParamsLine("                          : output will be rootname_<n>.xmd");
        addParamsLine("  [--dont_randomize ]     : Do not generate random groups");
        addParamsLine("  [--dont_sort ]          : Do not sort the outputs MetaData");
        addParamsLine("  [--dont_remove_disabled]: Do not remove disabled images from MetaData");
        addParamsLine("  [--use_correlation <fnCC=\"\"> <iter=100> <subset=16>] : Use correlation to a single volume.");
        addParamsLine("                             : This is the output of xmipp_reconstruct_significant");
        addParamsLine("                             : with a single reference volume.");
        addParamsLine("                             :+ The metadata is splitted by taking random subsets of the projection");
        addParamsLine("                             :+ directions and applying hierarchical clustering a number of");
        addParamsLine("                             :+ iterations on these random subsets");
        addParamsLine("  [-l <label=\"image\">] : sort using a label, default image");

        addSeeAlsoLine("reconstruct_significant");

        addExampleLine("Splits input.sel in two parts:", false);
        addExampleLine("  xmipp_metadata_split -i input.sel --oroot output_part");
        addExampleLine("Splits input.sel in 4 output files without randomizing input metdata:", false);
        addExampleLine("  xmipp_metadata_split -i input.sel -n 4 --dont_randomize");
        addExampleLine("Splits input.sel in two parts with sel extension:", false);
        addExampleLine("  xmipp_metadata_split -i input.sel --oroot output_part:sel");
    }

readParams()

void ProgMetadataSplit::readParams ( )

inlineprotectedvirtual

Function in which each program will read parameters that it need. If some error occurs the usage will be printed out.

Reimplemented from XmippProgram.

Definition at line 77 of file metadata_split.cpp.

    {
        fn_in = getParam("-i");
        N = getIntParam("-n");
        fn_root = getParam("--oroot");
        extension = fn_root.getFileFormat();
        fn_root = fn_root.removeFileFormat();
        if (fn_root.empty())
            fn_root = fn_in.withoutExtension();
        if (extension.empty())
            extension = "xmd";
        dont_randomize = checkParam("--dont_randomize");
        dont_sort      = checkParam("--dont_sort");
        dont_remove_disabled = checkParam("--dont_remove_disabled");
        use_cc = checkParam("--use_correlation");
        if (use_cc)
        {
            fn_cc = getParam("--use_correlation",0);
            AHCiter = getIntParam("--use_correlation",1);
            AHCsubset = getIntParam("--use_correlation",2);
            dont_randomize=true;
        }

        sortLabelStr = "objId"; //if not sort, by default is objId
        if(!dont_sort)
            sortLabelStr = getParam("-l");

        sortLabel = MDL::str2Label(sortLabelStr);
        if (sortLabel == MDL_UNDEFINED)
            REPORT_ERROR(ERR_MD_UNDEFINED, (String)"Unrecognized label '" + sortLabelStr + "'");
    }

run()

void ProgMetadataSplit::run ( )

inlinevirtual

This function will be start running the program. it also should be implemented by derived classes.

Reimplemented from XmippProgram.

Definition at line 110 of file metadata_split.cpp.

    {
        mdIn.read(fn_in);

        if (dont_randomize)
            mdPtr = &mdIn;
        else
        {
            mdPtr = new MetaDataVec();
            mdPtr->randomize(mdIn);
        }

        if (!dont_remove_disabled && mdPtr->containsLabel(MDL_ENABLED)) //remove disabled entries by default
            mdPtr->removeObjects(MDValueEQ(MDL_ENABLED, -1));

        size_t Num_images = mdPtr->size();
        size_t Num_groups = XMIPP_MIN(N, Num_images);
        if (!use_cc)
            mdPtr->split(Num_groups, mdVector); //Split MD in Num_groups
        else
        {
            Image<double> cc;
            cc.read(fn_cc);
            MultidimArray<double> &mcc=cc();
            auto Ndirs=(int)XSIZE(mcc);
            auto Nvols=(int)YSIZE(mcc);
            auto Nimgs=(int)ZSIZE(mcc);
            if (Nvols!=1)
                REPORT_ERROR(ERR_ARG_INCORRECT,"This program is meant only for cross-correlation volumes of a single reference model");
            Matrix2D<int> coocurrence(Nimgs,Nimgs);
            Matrix2D<double> X(Nimgs,AHCsubset);
            MultidimArray<int> permutation;
            AHCClassifier ahc;

            for (int iter=0; iter<AHCiter; ++iter)
            {
                // Take the correlation with random directions that will serve as probe functions
                randomPermutation(Ndirs,permutation);
                FOR_ALL_ELEMENTS_IN_MATRIX2D(X)
                    MAT_ELEM(X,i,j)=A3D_ELEM(mcc,i,0,A1D_ELEM(permutation,j));

                // Classify
                ahc.clusterData(X,(int)Num_groups);

                // Update coocurrence
                for (int i=0; i<Nimgs-1; ++i)
                    for (int j=i+1; j<Nimgs; ++j)
                        if (VEC_ELEM(ahc.clusterAssigned,i)==VEC_ELEM(ahc.clusterAssigned,j))
                        {
                            MAT_ELEM(coocurrence,i,j)+=1;
                            MAT_ELEM(coocurrence,j,i)=MAT_ELEM(coocurrence,i,j);
                        }
            }

            // Convert the coocurrence matrix into a distance matrix
            int maxCoocurrence=coocurrence.computeMax();
            Matrix2D<double> D;
            D.resizeNoCopy(coocurrence);
            FOR_ALL_ELEMENTS_IN_MATRIX2D(coocurrence)
                MAT_ELEM(D,i,j)=maxCoocurrence-MAT_ELEM(coocurrence,i,j);

            // Now perform the clustering with this distance
            ahc.clusterWithDistance(D,(int)Num_groups);

            // Construct output metadata
            MetaDataVec dummy;
            for (size_t k=0; k<Num_groups; ++k)
                mdVector.push_back(dummy);

            size_t i=0;
            for (auto& row : *mdPtr)
            {
                mdVector[VEC_ELEM(ahc.clusterAssigned,i)].addRow(dynamic_cast<MDRowVec&>(row));
                i++;
            }
        }

        //Write splitted groups to disk
        for (size_t i = 0; i < Num_groups; ++i)
        {
            fn_out.compose(fn_root, i + 1, extension);

            if(!dont_sort)
                mdPtr->sort(mdVector[i], sortLabel);
            else
                mdPtr = &(mdVector[i]);

            mdPtr->write(fn_out);
        }
    }

---

The documentation for this class was generated from the following file:

• xmipp/libraries/reconstruction/metadata_split.cpp