This file contains function prototypes. More...

This graph shows which files directly or indirectly include this file:

Functions
float	ComputeAccuracy (int n, gk_fkv_t *list)

float	ComputeMean (int n, float *values)

float	ComputeMedianRFP (int n, gk_fkv_t *list)

float	ComputeROCn (int n, int maxN, gk_fkv_t *list)

float	ComputeStdDev (int n, float *values)

void	decodeblock (unsigned char in, unsigned char out)

void	encodeblock (unsigned char in, unsigned char out)

void	errexit (char *,...)

void	gk_AllocMatrix (void ***, size_t, size_t, size_t)

void	gk_array2csr (size_t n, size_t range, int array, int ptr, int *ind)
	Converts an element-based set membership into a CSR-format set-based membership. More...

void	gk_ckvsortd (size_t, gk_ckv_t *)

void	gk_ckvsorti (size_t, gk_ckv_t *)

double	gk_CPUSeconds (void)

void	gk_csortd (size_t, char *)

void	gk_csorti (size_t, char *)

void	gk_csr_CompactColumns (gk_csr_t *mat)

float	gk_csr_ComputeSimilarity (gk_csr_t *mat, int i1, int i2, int what, int simtype)

void	gk_csr_ComputeSquaredNorms (gk_csr_t *mat, int what)

void	gk_csr_ComputeSums (gk_csr_t *mat, int what)

gk_csr_t *	gk_csr_Create ()

void	gk_csr_CreateIndex (gk_csr_t *mat, int what)

gk_csr_t *	gk_csr_Dup (gk_csr_t *mat)

gk_csr_t *	gk_csr_ExtractPartition (gk_csr_t mat, int part, int pid)

gk_csr_t *	gk_csr_ExtractRows (gk_csr_t mat, int nrows, int rind)

gk_csr_t *	gk_csr_ExtractSubmatrix (gk_csr_t *mat, int rstart, int nrows)

void	gk_csr_Free (gk_csr_t **mat)

void	gk_csr_FreeContents (gk_csr_t *mat)

int	gk_csr_GetSimilarRows (gk_csr_t mat, int nqterms, int qind, float qval, int simtype, int nsim, float minsim, gk_fkv_t hits, int _imarker, gk_fkv_t i_cand)

void	gk_csr_Init (gk_csr_t *mat)

gk_csr_t *	gk_csr_LowFilter (gk_csr_t *mat, int what, int norm, float fraction)

void	gk_csr_Normalize (gk_csr_t *mat, int what, int norm)

gk_csr_t *	gk_csr_Prune (gk_csr_t *mat, int what, int minf, int maxf)

gk_csr_t *	gk_csr_Read (char *filename, int format, int readvals, int numbering)

void	gk_csr_Scale (gk_csr_t *mat, int type)

void	gk_csr_SortIndices (gk_csr_t *mat, int what)

gk_csr_t **	gk_csr_Split (gk_csr_t mat, int color)

gk_csr_t *	gk_csr_TopKPlusFilter (gk_csr_t *mat, int what, int topk, float keepval)

void	gk_csr_Write (gk_csr_t mat, char filename, int format, int writevals, int numbering)

gk_csr_t *	gk_csr_ZScoreFilter (gk_csr_t *mat, int what, float zscore)

int	gk_dexists (char *)

int	gk_dfkvkselect (size_t, int, gk_fkv_t *)

void	gk_dkvsortd (size_t, gk_dkv_t *)

void	gk_dkvsorti (size_t, gk_dkv_t *)

double *	gk_dreadfilebin (char fname, ssize_t r_nelmnts)

void	gk_dsortd (size_t, double *)

void	gk_dsorti (size_t, double *)

void	gk_errexit (int signum, char *,...)

void	gk_fclose (FILE *)

int	gk_fexists (char *)

void	gk_find_frequent_itemsets (int ntrans, ssize_t tranptr, int tranind, int minfreq, int maxfreq, int minlen, int maxlen, void(process_itemset)(void stateptr, int nitems, int itemind, int ntrans, int tranind), void *stateptr)

void	gk_fkvsortd (size_t, gk_fkv_t *)

void	gk_fkvsorti (size_t, gk_fkv_t *)

float	gk_flog2 (float)

FILE *	gk_fopen (char , char , const char *)

float *	gk_freadfilebin (char fname, ssize_t r_nelmnts)

void	gk_free (void **ptr1,...)

void	gk_FreeMatrix (void ***, size_t, size_t)

void	gk_freepdbf (pdbf *p)
	Frees the memory of a pdbf structure. More...

void	gk_freetokenslist (gk_Tokens_t *tokens)

void	gk_fsortd (size_t, float *)

void	gk_fsorti (size_t, float *)

size_t	gk_fwritefilebin (char fname, size_t n, float a)

char *	gk_getbasename (char *path)

size_t	gk_GetCurMemoryUsed ()

char *	gk_getextname (char *path)

char *	gk_getfilename (char *path)

void	gk_getfilestats (char fname, size_t r_nlines, size_t r_ntokens, size_t r_max_nlntokens, size_t *r_nbytes)

intmax_t	gk_getfsize (char *)
	Returns the size of the file in bytes. More...

gk_idx_t	gk_getline (char *lineptr, size_t n, FILE *stream)

size_t	gk_GetMaxMemoryUsed ()

char *	gk_getpathname (char *path)

int	gk_GetStringID (gk_StringMap_t strmap, char key)

void	gk_gkmcoreAdd (gk_mcore_t mcore, int type, size_t nbytes, void ptr)

gk_mcore_t *	gk_gkmcoreCreate ()

void	gk_gkmcoreDel (gk_mcore_t mcore, void ptr)

void	gk_gkmcoreDestroy (gk_mcore_t **r_mcore, int showstats)

void	gk_gkmcorePop (gk_mcore_t *mcore)

void	gk_gkmcorePush (gk_mcore_t *mcore)

void	gk_graph_ComputeBestFOrdering (gk_graph_t graph, int v, int type, int32_t r_perm, int32_t *r_iperm)

void	gk_graph_ComputeBestFOrdering0 (gk_graph_t graph, int v, int type, int32_t r_perm, int32_t *r_iperm)

void	gk_graph_ComputeBFSOrdering (gk_graph_t graph, int v, int32_t r_perm, int32_t *r_iperm)

gk_graph_t *	gk_graph_Create ()

gk_graph_t *	gk_graph_Dup (gk_graph_t *graph)

gk_graph_t *	gk_graph_ExtractSubgraph (gk_graph_t *graph, int vstart, int nvtxs)

int	gk_graph_FindComponents (gk_graph_t graph, int32_t cptr, int32_t *cind)

void	gk_graph_Free (gk_graph_t **graph)

void	gk_graph_FreeContents (gk_graph_t *graph)

void	gk_graph_Init (gk_graph_t *graph)

gk_graph_t *	gk_graph_Read (char *filename, int format, int isfewgts, int isfvwgts, int isfvsizes)

gk_graph_t *	gk_graph_Reorder (gk_graph_t graph, int32_t perm, int32_t *iperm)

void	gk_graph_SingleSourceShortestPaths (gk_graph_t graph, int v, void *r_sps)

void	gk_graph_Write (gk_graph_t graph, char filename, int format)

gk_i2cc2i_t *	gk_i2cc2i_create_common (char *alphabet)
	This function creates the localizations for the various sequences. More...

void	gk_i32kvsortd (size_t, gk_i32kv_t *)

void	gk_i32kvsorti (size_t, gk_i32kv_t *)

int32_t *	gk_i32readfile (char fname, gk_idx_t r_nlines)

int32_t *	gk_i32readfilebin (char fname, ssize_t r_nelmnts)

void	gk_i64kvsortd (size_t, gk_i64kv_t *)

void	gk_i64kvsorti (size_t, gk_i64kv_t *)

int64_t *	gk_i64readfile (char fname, gk_idx_t r_nlines)

int64_t *	gk_i64readfilebin (char fname, ssize_t r_nelmnts)

void	gk_idxkvsortd (size_t, gk_idxkv_t *)

void	gk_idxkvsorti (size_t, gk_idxkv_t *)

void	gk_idxsortd (size_t, gk_idx_t *)

void	gk_idxsorti (size_t, gk_idx_t *)

int	gk_ifkvkselect (size_t, int, gk_fkv_t *)

void	gk_ikvsortd (size_t, gk_ikv_t *)

void	gk_ikvsorti (size_t, gk_ikv_t *)

void	gk_isortd (size_t, int *)

void	gk_isorti (size_t, int *)

int	gk_ispow2 (int)

int	gk_log2 (int)

void *	gk_malloc (size_t nbytes, char *msg)

void	gk_malloc_cleanup (int showstats)

int	gk_malloc_init ()

void	gk_mcoreAdd (gk_mcore_t mcore, int type, size_t nbytes, void ptr)

gk_mcore_t *	gk_mcoreCreate (size_t coresize)

void	gk_mcoreDel (gk_mcore_t mcore, void ptr)

void	gk_mcoreDestroy (gk_mcore_t **r_mcore, int showstats)

void *	gk_mcoreMalloc (gk_mcore_t *mcore, size_t nbytes)

void	gk_mcorePop (gk_mcore_t *mcore)

void	gk_mcorePush (gk_mcore_t *mcore)

int	gk_mkpath (char *)

void	gk_NonLocalExit_Handler (int signum)

void	gk_randinit (uint64_t)

uint32_t	gk_randint32 (void)

uint64_t	gk_randint64 (void)

void	gk_RandomPermute (size_t, int *, int)

char **	gk_readfile (char fname, gk_idx_t r_nlines)

pdbf *	gk_readpdbfile (char *fname)
	Reads a pdb file into a pdbf structure. More...

void *	gk_realloc (void oldptr, size_t nbytes, char msg)

int	gk_rmpath (char *)

int	gk_rw_PageRank (gk_csr_t mat, float lamda, float eps, int max_niter, float pr)

void	gk_seq_init (gk_seq_t *seq)

gk_seq_t *	gk_seq_ReadGKMODPSSM (char *file_name)
	This function reads a pssm in the format of gkmod pssm. More...

void	gk_set_exit_on_error (int value)

void	gk_SetSignalHandlers ()

void	gk_showcorruption (pdbf *p)
	Decodes the corruption bitswitch and prints any problems. More...

void	gk_sigthrow (int signum)

int	gk_sigtrap ()

int	gk_siguntrap ()

void	gk_skvsortd (size_t, gk_skv_t *)

void	gk_skvsorti (size_t, gk_skv_t *)

time_t	gk_str2time (char *str)
	Converts a date/time string into its equivalent time_t value. More...

int	gk_strcasecmp (char s1, char s2)
	Case insensitive string comparison. More...

char *	gk_strchr_replace (char str, char fromlist, char *tolist)
	Replaces certain characters in a string. More...

char *	gk_strdup (char *orgstr)
	Duplicates a string. More...

char *	gk_strerror (int errnum)
	Thread-safe implementation of strerror() More...

char *	gk_strhprune (char , char )
	Prunes characters from the beginning of the string. More...

int	gk_strrcmp (char s1, char s2)
	Compare two strings in revere order. More...

int	gk_strstr_replace (char str, char pattern, char replacement, char options, char **new_str)
	Regex-based search-and-replace function. More...

void	gk_strtokenize (char line, char delim, gk_Tokens_t *tokens)

char *	gk_strtolower (char *)
	Converts a string to lower case. More...

char *	gk_strtoupper (char *)
	Converts a string to upper case. More...

char *	gk_strtprune (char , char )
	Prunes characters from the end of the string. More...

char	gk_threetoone (char *res)
	Converts three-letter amino acid codes to one-leter codes. More...

char *	gk_time2str (time_t time)
	Converts a time_t time into a string. More...

void	gk_UnsetSignalHandlers ()

gk_wclock_t	gk_WClockSeconds (void)

void	gk_writealphacarbons (pdbf p, char fname)
	Writes out all the alpha carbon atoms of a structure. More...

void	gk_writebackbone (pdbf p, char fname)
	Writes out all the backbone atoms of a structure in pdb format. More...

void	gk_writefullatom (pdbf p, char fname)
	Writes all atoms in p in pdb-format to file fname. More...

void	gk_zkvsortd (size_t, gk_zkv_t *)

void	gk_zkvsorti (size_t, gk_zkv_t *)

void	GKDecodeBase64 (int nbytes, unsigned char inbuffer, unsigned char outbuffer)

void	GKEncodeBase64 (int nbytes, unsigned char inbuffer, unsigned char outbuffer)

gk_HTable_t *	HTable_Create (int nelements)

void	HTable_Delete (gk_HTable_t *htable, int key)

void	HTable_Destroy (gk_HTable_t *htable)

int	HTable_GetNext (gk_HTable_t htable, int key, int val, int type)

int	HTable_HFunction (int nelements, int key)

void	HTable_Insert (gk_HTable_t *htable, int key, int val)

void	HTable_Reset (gk_HTable_t *htable)

void	HTable_Resize (gk_HTable_t *htable, int nelements)

int	HTable_Search (gk_HTable_t *htable, int key)

int	HTable_SearchAndDelete (gk_HTable_t *htable, int key)

int	omp_get_dynamic (void)

int	omp_get_max_threads (void)

int	omp_get_nested (void)

int	omp_get_num_procs (void)

int	omp_get_num_threads (void)

int	omp_get_thread_num (void)

int	omp_in_parallel (void)

void	omp_set_dynamic (int num_threads)

void	omp_set_nested (int nested)

void	omp_set_num_threads (int num_threads)

void	PrintBackTrace ()

Detailed Description

This file contains function prototypes.

Date: Started 3/27/2007

Author: George

Version

$Id: gk_proto.h 12591 2012-09-01 19:03:15Z karypis $

Definition in file gk_proto.h.

Function Documentation

◆ ComputeAccuracy()

float ComputeAccuracy	(	int	n,
		gk_fkv_t *	list
	)

Definition at line 16 of file evaluate.c.

◆ ComputeMean()

float ComputeMean	(	int	n,
		float *	values
	)

Definition at line 107 of file evaluate.c.

◆ ComputeMedianRFP()

float ComputeMedianRFP	(	int	n,
		gk_fkv_t *	list
	)

Definition at line 81 of file evaluate.c.

◆ ComputeROCn()

float ComputeROCn	(	int	n,
		int	maxN,
		gk_fkv_t *	list
	)

Definition at line 47 of file evaluate.c.

◆ ComputeStdDev()

float ComputeStdDev	(	int	n,
		float *	values
	)

Definition at line 121 of file evaluate.c.

◆ decodeblock()

void decodeblock	(	unsigned char *	in,
		unsigned char *	out
	)

Definition at line 46 of file b64.c.

◆ encodeblock()

void encodeblock	(	unsigned char *	in,
		unsigned char *	out
	)

Definition at line 28 of file b64.c.

◆ errexit()

void errexit	(	char *	f_str,
			...
	)

This function prints an error message and exits

Definition at line 54 of file error.c.

◆ gk_AllocMatrix()

void gk_AllocMatrix	(	void ***	r_matrix,
		size_t	elmlen,
		size_t	ndim1,
		size_t	ndim2
	)

Define the set of memory allocation routines for each data type

This function allocates a two-dimensional matrix.

Definition at line 53 of file memory.c.

◆ gk_array2csr()

void gk_array2csr	(	size_t	n,
		size_t	range,
		int *	array,
		int *	ptr,
		int *	ind
	)

Converts an element-based set membership into a CSR-format set-based membership.

For example, it takes an array such as part[] that stores where each element belongs to and returns a pair of arrays (pptr[], pind[]) that store in CSF format the list of elements belonging in each partition.

Parameters

n	the number of elements in the array (e.g., # of vertices)
range	the cardinality of the set (e.g., # of partitions)
array	the array that stores the per-element set membership
ptr	the array that will store the starting indices in ind for the elements of each set. This is filled by the routine and its size should be at least range+1.
ind	the array that stores consecutively which elements belong to each set. The size of this array should be n.

Definition at line 62 of file GKlib/util.c.

◆ gk_ckvsortd()

void gk_ckvsortd	(	size_t	n,
		gk_ckv_t *	base
	)

Sorts an array of gk_ckv_t in decreasing order

Definition at line 146 of file sort.c.

◆ gk_ckvsorti()

void gk_ckvsorti	(	size_t	n,
		gk_ckv_t *	base
	)

Sorts an array of gk_ckv_t in increasing order

Definition at line 135 of file sort.c.

◆ gk_CPUSeconds()

double gk_CPUSeconds ( void )

Definition at line 36 of file timers.c.

◆ gk_csortd()

void gk_csortd	(	size_t	n,
		char *	base
	)

Sorts an array of chars in decreasing order

Definition at line 34 of file sort.c.

◆ gk_csorti()

void gk_csorti	(	size_t	n,
		char *	base
	)

Sorts an array of chars in increasing order

Definition at line 23 of file sort.c.

◆ gk_csr_CompactColumns()

void gk_csr_CompactColumns ( gk_csr_t * mat )

Compacts the column-space of the matrix by removing empty columns. As a result of the compaction, the column numbers are renumbered. The compaction operation is done in place and only affects the row-based representation of the matrix. The new columns are ordered in decreasing frequency.

Parameters

mat	the matrix whose empty columns will be removed.

Definition at line 1098 of file csr.c.

◆ gk_csr_ComputeSimilarity()

float gk_csr_ComputeSimilarity	(	gk_csr_t *	mat,
		int	i1,
		int	i2,
		int	what,
		int	simtype
	)

Computes the similarity between two rows/columns

Parameters

mat	the matrix itself. The routine assumes that the indices are sorted in increasing order.
i1	is the first row/column,
i2	is the second row/column,
what	is either GK_CSR_ROW or GK_CSR_COL indicating the type of objects between the similarity will be computed,
simtype	is the type of similarity and is one of GK_CSR_COS, GK_CSR_JAC, GK_CSR_MIN, GK_CSR_AMIN

Returns: the similarity between the two rows/columns.

Definition at line 1694 of file csr.c.

◆ gk_csr_ComputeSquaredNorms()

void gk_csr_ComputeSquaredNorms	(	gk_csr_t *	mat,
		int	what
	)

Computes the squared of the norms of the rows/columns

Parameters

mat	the matrix itself,
what	is either GK_CSR_ROW or GK_CSR_COL indicating which squared norms to compute.

Definition at line 1643 of file csr.c.

◆ gk_csr_ComputeSums()

void gk_csr_ComputeSums	(	gk_csr_t *	mat,
		int	what
	)

Computes the sums of the rows/columns

Parameters

mat	the matrix itself,
what	is either GK_CSR_ROW or GK_CSR_COL indicating which sums to compute.

Definition at line 1597 of file csr.c.

◆ gk_csr_Create()

gk_csr_t* gk_csr_Create ( )

Allocate memory for a CSR matrix and initializes it

Returns: the allocated matrix. The various fields are set to NULL.

Definition at line 19 of file csr.c.

◆ gk_csr_CreateIndex()

void gk_csr_CreateIndex	(	gk_csr_t *	mat,
		int	what
	)

Creates a row/column index from the column/row data.

Parameters

mat	the matrix itself,
what	is either GK_CSR_ROW or GK_CSR_COL indicating which index will be created.

Definition at line 1223 of file csr.c.

◆ gk_csr_Dup()

gk_csr_t* gk_csr_Dup ( gk_csr_t * mat )

Returns a copy of a matrix.

Parameters

mat	is the matrix to be duplicated.

Returns: the newly created copy of the matrix.

Definition at line 80 of file csr.c.

◆ gk_csr_ExtractPartition()

gk_csr_t* gk_csr_ExtractPartition	(	gk_csr_t *	mat,
		int *	part,
		int	pid
	)

Returns a submatrix corresponding to a specified partitioning of rows.

Parameters

mat	is the original matrix.
part	is the partitioning vector of the rows.
pid	is the partition ID that will be extracted.

Returns: the row structure of the newly created submatrix.

Definition at line 230 of file csr.c.

◆ gk_csr_ExtractRows()

gk_csr_t* gk_csr_ExtractRows	(	gk_csr_t *	mat,
		int	nrows,
		int *	rind
	)

Returns a submatrix containing a certain set of rows.

Parameters

mat	is the original matrix.
nrows	is the number of rows to extract.
rind	is the set of row numbers to extract.

Returns: the row structure of the newly created submatrix.

Definition at line 191 of file csr.c.

◆ gk_csr_ExtractSubmatrix()

gk_csr_t* gk_csr_ExtractSubmatrix	(	gk_csr_t *	mat,
		int	rstart,
		int	nrows
	)

Returns a submatrix containint a set of consecutive rows.

Parameters

mat	is the original matrix.
rstart	is the starting row.
nrows	is the number of rows from rstart to extract.

Returns: the row structure of the newly created submatrix.

Definition at line 135 of file csr.c.

◆ gk_csr_Free()

void gk_csr_Free ( gk_csr_t ** mat )

Frees all the memory allocated for matrix.

Parameters

mat	is the matrix to be freed.

Definition at line 48 of file csr.c.

◆ gk_csr_FreeContents()

void gk_csr_FreeContents ( gk_csr_t * mat )

Frees only the memory allocated for the matrix's different fields and sets them to NULL.

Parameters

mat	is the matrix whose contents will be freed.

Definition at line 63 of file csr.c.

◆ gk_csr_GetSimilarRows()

int gk_csr_GetSimilarRows	(	gk_csr_t *	mat,
		int	nqterms,
		int *	qind,
		float *	qval,
		int	simtype,
		int	nsim,
		float	minsim,
		gk_fkv_t *	hits,
		int *	i_marker,
		gk_fkv_t *	i_cand
	)

Finds the n most similar rows (neighbors) to the query using cosine similarity.

Parameters

mat	the matrix itself
nqterms	is the number of columns in the query
qind	is the list of query columns
qval	is the list of correspodning query weights
simtype	is the type of similarity and is one of GK_CSR_COS, GK_CSR_JAC, GK_CSR_MIN, GK_CSR_AMIN
nsim	is the maximum number of requested most similar rows. If -1 is provided, then everything is returned unsorted.
minsim	is the minimum similarity of the requested most similar rows
hits	is the result set. This array should be at least of length nsim.
i_marker	is an array of size equal to the number of rows whose values are initialized to -1. If NULL is provided then this array is allocated and freed internally.
i_cand	is an array of size equal to the number of rows. If NULL is provided then this array is allocated and freed internally.

Returns: the number of identified most similar rows, which can be smaller than the requested number of nnbrs in those cases in which there are no sufficiently many neighbors.

Definition at line 1866 of file csr.c.

◆ gk_csr_Init()

void gk_csr_Init ( gk_csr_t * mat )

Initializes the matrix

Parameters

mat	is the matrix to be initialized.

Definition at line 36 of file csr.c.

◆ gk_csr_LowFilter()

gk_csr_t* gk_csr_LowFilter	(	gk_csr_t *	mat,
		int	what,
		int	norm,
		float	fraction
	)

Eliminates certain entries from the rows/columns of the matrix. The filtering takes place by keeping only the highest weight entries whose sum accounts for a certain fraction of the overall weight of the row/column.

Parameters

mat	the matrix to be prunned,
what	indicates if the rows (GK_CSR_ROW) or the columns (GK_CSR_COL) of the matrix will be prunned,
norm	indicates the norm that will be used to aggregate the weights and possible values are 1 or 2,
fraction	is the fraction of the overall norm that will be retained by the kept entries.

Returns: the filtered matrix consisting only of its row-based structure. The input matrix is not modified.

Definition at line 759 of file csr.c.

◆ gk_csr_Normalize()

void gk_csr_Normalize	(	gk_csr_t *	mat,
		int	what,
		int	norm
	)

Normalizes the rows/columns of the matrix to be unit length.

Parameters

mat	the matrix itself,
what	indicates what will be normalized and is obtained by specifying GK_CSR_ROW, GK_CSR_COL, GK_CSR_ROW\|GK_CSR_COL.
norm	indicates what norm is to normalize to, 1: 1-norm, 2: 2-norm

Definition at line 1319 of file csr.c.

◆ gk_csr_Prune()

gk_csr_t* gk_csr_Prune	(	gk_csr_t *	mat,
		int	what,
		int	minf,
		int	maxf
	)

Prunes certain rows/columns of the matrix. The prunning takes place by analyzing the row structure of the matrix. The prunning takes place by removing rows/columns but it does not affect the numbering of the remaining rows/columns.

Parameters

mat	the matrix to be prunned,
what	indicates if the rows (GK_CSR_ROW) or the columns (GK_CSR_COL) of the matrix will be prunned,
minf	is the minimum number of rows (columns) that a column (row) must be present in order to be kept,
maxf	is the maximum number of rows (columns) that a column (row) must be present at in order to be kept.

Returns: the prunned matrix consisting only of its row-based structure. The input matrix is not modified.

Definition at line 669 of file csr.c.

◆ gk_csr_Read()

gk_csr_t* gk_csr_Read	(	char *	filename,
		int	format,
		int	readvals,
		int	numbering
	)

Reads a CSR matrix from the supplied file and stores it the matrix's forward structure.

Parameters

filename	is the file that stores the data.
format	is either GK_CSR_FMT_METIS, GK_CSR_FMT_CLUTO, GK_CSR_FMT_CSR, GK_CSR_FMT_BINROW, GK_CSR_FMT_BINCOL specifying the type of the input format. The GK_CSR_FMT_CSR does not contain a header line, whereas the GK_CSR_FMT_BINROW is a binary format written by gk_csr_Write() using the same format specifier.
readvals	is either 1 or 0, indicating if the CSR file contains values or it does not. It only applies when GK_CSR_FMT_CSR is used.
numbering	is either 1 or 0, indicating if the numbering of the indices start from 1 or 0, respectively. If they start from 1, they are automatically decreamented during input so that they will start from 0. It only applies when GK_CSR_FMT_CSR is used.

Returns: the matrix that was read.

Definition at line 349 of file csr.c.

◆ gk_csr_Scale()

void gk_csr_Scale	(	gk_csr_t *	mat,
		int	type
	)

Applies different row scaling methods.

Parameters

mat	the matrix itself,
type	indicates the type of row scaling. Possible values are: GK_CSR_MAXTF, GK_CSR_SQRT, GK_CSR_LOG, GK_CSR_IDF, GK_CSR_MAXTF2.

Definition at line 1386 of file csr.c.

◆ gk_csr_SortIndices()

void gk_csr_SortIndices	(	gk_csr_t *	mat,
		int	what
	)

Sorts the indices in increasing order

Parameters

mat	the matrix itself,
what	is either GK_CSR_ROW or GK_CSR_COL indicating which set of indices to sort.

Definition at line 1146 of file csr.c.

◆ gk_csr_Split()

gk_csr_t** gk_csr_Split	(	gk_csr_t *	mat,
		int *	color
	)

Splits the matrix into multiple sub-matrices based on the provided color array.

Parameters

mat	is the original matrix.
color	is an array of size equal to the number of non-zeros in the matrix (row-wise structure). The matrix is split into as many parts as the number of colors. For meaningfull results, the colors should be numbered consecutively starting from 0.

Returns: an array of matrices for each supplied color number.

Definition at line 277 of file csr.c.

◆ gk_csr_TopKPlusFilter()

gk_csr_t* gk_csr_TopKPlusFilter	(	gk_csr_t *	mat,
		int	what,
		int	topk,
		float	keepval
	)

Eliminates certain entries from the rows/columns of the matrix. The filtering takes place by keeping only the highest weight top-K entries along each row/column and those entries whose weight is greater than a specified value.

Parameters

mat	the matrix to be prunned,
what	indicates if the rows (GK_CSR_ROW) or the columns (GK_CSR_COL) of the matrix will be prunned,
topk	is the number of the highest weight entries to keep.
keepval	is the weight of a term above which will be kept. This is used to select additional terms past the first topk.

Returns: the filtered matrix consisting only of its row-based structure. The input matrix is not modified.

Definition at line 901 of file csr.c.

◆ gk_csr_Write()

void gk_csr_Write	(	gk_csr_t *	mat,
		char *	filename,
		int	format,
		int	writevals,
		int	numbering
	)

Writes the row-based structure of a matrix into a file.

Parameters

mat	is the matrix to be written,
filename	is the name of the output file.
format	is one of: GK_CSR_FMT_CLUTO, GK_CSR_FMT_CSR, GK_CSR_FMT_BINROW, GK_CSR_FMT_BINCOL.
writevals	is either 1 or 0 indicating if the values will be written or not. This is only applicable when GK_CSR_FMT_CSR is used.
numbering	is either 1 or 0 indicating if the internal 0-based numbering will be shifted by one or not during output. This is only applicable when GK_CSR_FMT_CSR is used.

Definition at line 591 of file csr.c.

◆ gk_csr_ZScoreFilter()

gk_csr_t* gk_csr_ZScoreFilter	(	gk_csr_t *	mat,
		int	what,
		float	zscore
	)

Eliminates certain entries from the rows/columns of the matrix. The filtering takes place by keeping only the terms whose contribution to the total length of the document is greater than a user-splied multiple over the average.

This routine assumes that the vectors are normalized to be unit length.

Parameters

mat	the matrix to be prunned,
what	indicates if the rows (GK_CSR_ROW) or the columns (GK_CSR_COL) of the matrix will be prunned,
zscore	is the multiplicative factor over the average contribution to the length of the document.

Returns: the filtered matrix consisting only of its row-based structure. The input matrix is not modified.

Definition at line 1031 of file csr.c.

◆ gk_dexists()

int gk_dexists ( char * )

Definition at line 35 of file fs.c.

◆ gk_dfkvkselect()

int gk_dfkvkselect	(	size_t	n,
		int	topk,
		gk_fkv_t *	cand
	)

This function puts the 'topk' largest values in the beginning of the array

Definition at line 20 of file fkvkselect.c.

◆ gk_dkvsortd()

void gk_dkvsortd	(	size_t	n,
		gk_dkv_t *	base
	)

Sorts an array of gk_fkv_t in decreasing order

Definition at line 278 of file sort.c.

◆ gk_dkvsorti()

void gk_dkvsorti	(	size_t	n,
		gk_dkv_t *	base
	)

Sorts an array of gk_dkv_t in increasing order

Definition at line 267 of file sort.c.

◆ gk_dreadfilebin()

double* gk_dreadfilebin	(	char *	fname,
		ssize_t *	r_nelmnts
	)

This function reads the contents of a binary file and returns it in the form of an array of double.

Parameters

fname	is the name of the file
r_nlines	is the number of lines in the file. If it is NULL, this information is not returned.

Definition at line 354 of file GKlib/io.c.

◆ gk_dsortd()

void gk_dsortd	(	size_t	n,
		double *	base
	)

Sorts an array of doubles in decreasing order

Definition at line 100 of file sort.c.

◆ gk_dsorti()

void gk_dsorti	(	size_t	n,
		double *	base
	)

Sorts an array of doubles in increasing order

Definition at line 89 of file sort.c.

◆ gk_errexit()

void gk_errexit	(	int	signum,
		char *	f_str,
			...
	)

This function prints an error message and raises a signum signal

Definition at line 77 of file error.c.

◆ gk_fclose()

void gk_fclose ( FILE * )

Definition at line 44 of file GKlib/io.c.

◆ gk_fexists()

int gk_fexists ( char * )

Definition at line 21 of file fs.c.

◆ gk_find_frequent_itemsets()

void gk_find_frequent_itemsets	(	int	ntrans,
		ssize_t *	tranptr,
		int *	tranind,
		int	minfreq,
		int	maxfreq,
		int	minlen,
		int	maxlen,
		void()(void stateptr, int nitems, int itemind, int ntrans, int tranind)	process_itemset,
		void *	stateptr
	)

◆ gk_fkvsortd()

void gk_fkvsortd	(	size_t	n,
		gk_fkv_t *	base
	)

Sorts an array of gk_fkv_t in decreasing order

Definition at line 256 of file sort.c.

◆ gk_fkvsorti()

void gk_fkvsorti	(	size_t	n,
		gk_fkv_t *	base
	)

Sorts an array of gk_fkv_t in increasing order

Definition at line 245 of file sort.c.

◆ gk_flog2()

float gk_flog2 ( float )

Definition at line 104 of file GKlib/util.c.

◆ gk_fopen()

FILE* gk_fopen	(	char *	,
		char *	,
		const char *
	)

Definition at line 24 of file GKlib/io.c.

◆ gk_freadfilebin()

float* gk_freadfilebin	(	char *	fname,
		ssize_t *	r_nelmnts
	)

This function reads the contents of a binary file and returns it in the form of an array of float.

Parameters

fname	is the name of the file
r_nlines	is the number of lines in the file. If it is NULL, this information is not returned.

Definition at line 292 of file GKlib/io.c.

◆ gk_free()

void gk_free	(	void **	ptr1,
			...
	)

Definition at line 202 of file memory.c.

◆ gk_FreeMatrix()

void gk_FreeMatrix	(	void ***	r_matrix,
		size_t	ndim1,
		size_t	ndim2
	)

This function frees a two-dimensional matrix.

Definition at line 79 of file memory.c.

◆ gk_freepdbf()

void gk_freepdbf ( pdbf * p )

Frees the memory of a pdbf structure.

This function takes a pdbf pointer and frees all the memory below it.

Parameters

p	is the pdbf structure to be freed.

Definition at line 105 of file pdb.c.

◆ gk_freetokenslist()

void gk_freetokenslist ( gk_Tokens_t * tokens )

Definition at line 73 of file tokenizer.c.

◆ gk_fsortd()

void gk_fsortd	(	size_t	n,
		float *	base
	)

Sorts an array of floats in decreasing order

Definition at line 78 of file sort.c.

◆ gk_fsorti()

void gk_fsorti	(	size_t	n,
		float *	base
	)

Sorts an array of floats in increasing order

Definition at line 67 of file sort.c.

◆ gk_fwritefilebin()

size_t gk_fwritefilebin	(	char *	fname,
		size_t	n,
		float *	a
	)

This function writes the contents of an array into a binary file.

Parameters

fname	is the name of the file
n	the number of elements in the array.
a	the array to be written out.

Definition at line 331 of file GKlib/io.c.

◆ gk_getbasename()

char* gk_getbasename ( char * path )

Definition at line 135 of file fs.c.

◆ gk_GetCurMemoryUsed()

size_t gk_GetCurMemoryUsed ( )

Definition at line 233 of file memory.c.

◆ gk_getextname()

char* gk_getextname ( char * path )

Definition at line 159 of file fs.c.

◆ gk_getfilename()

char* gk_getfilename ( char * path )

Definition at line 173 of file fs.c.

◆ gk_getfilestats()

void gk_getfilestats	(	char *	fname,
		size_t *	r_nlines,
		size_t *	r_ntokens,
		size_t *	r_max_nlntokens,
		size_t *	r_nbytes
	)

This function gets some basic statistics about the file.

Parameters

fname	is the name of the file
r_nlines	is the number of lines in the file. If it is NULL, this information is not returned.
r_ntokens	is the number of tokens in the file. If it is NULL, this information is not returned.
r_max_nlntokens	is the maximum number of tokens in any line in the file. If it is NULL this information is not returned.
r_nbytes	is the number of bytes in the file. If it is NULL, this information is not returned.

Definition at line 79 of file fs.c.

◆ gk_getfsize()

intmax_t gk_getfsize ( char * filename )

Returns the size of the file in bytes.

This function returns the size of a file as a 64 bit integer. If there were any errors in stat'ing the file, -1 is returned.

Note: That due to the -1 return code, the maximum file size is limited to 63 bits (which I guess is okay for now).

Definition at line 55 of file fs.c.

◆ gk_getline()

gk_idx_t gk_getline	(	char **	lineptr,
		size_t *	n,
		FILE *	stream
	)

This function is the GKlib implementation of glibc's getline() function.

Returns: -1 if the EOF has been reached, otherwise it returns the number of bytes read.

Definition at line 57 of file GKlib/io.c.

◆ gk_GetMaxMemoryUsed()

size_t gk_GetMaxMemoryUsed ( )

Definition at line 246 of file memory.c.

◆ gk_getpathname()

char* gk_getpathname ( char * path )

◆ gk_GetStringID()

int gk_GetStringID	(	gk_StringMap_t *	strmap,
		char *	key
	)

Definition at line 519 of file string.c.

◆ gk_gkmcoreAdd()

void gk_gkmcoreAdd	(	gk_mcore_t *	mcore,
		int	type,
		size_t	nbytes,
		void *	ptr
	)

Adds a memory allocation at the end of the list. This is the gkmcore version.

Definition at line 309 of file mcore.c.

◆ gk_gkmcoreCreate()

gk_mcore_t* gk_gkmcoreCreate ( )

This function creates an mcore. This version is used for gkmcore.

Definition at line 43 of file mcore.c.

◆ gk_gkmcoreDel()

void gk_gkmcoreDel	(	gk_mcore_t *	mcore,
		void *	ptr
	)

This function deletes the mop associated with the supplied pointer. The mop has to be a heap allocation, otherwise it fails violently. This is the gkmcore version.

Definition at line 373 of file mcore.c.

◆ gk_gkmcoreDestroy()

void gk_gkmcoreDestroy	(	gk_mcore_t **	r_mcore,
		int	showstats
	)

This function destroys an mcore. This version is for gkmcore.

Definition at line 103 of file mcore.c.

◆ gk_gkmcorePop()

void gk_gkmcorePop ( gk_mcore_t * mcore )

This function frees all mops since the last push. This version is for poping the gkmcore and it uses free instead of gk_free.

Definition at line 237 of file mcore.c.

◆ gk_gkmcorePush()

void gk_gkmcorePush ( gk_mcore_t * mcore )

This function sets a marker in the stack of malloc ops to be used subsequently for freeing purposes. This is the gkmcore version.

Definition at line 187 of file mcore.c.

◆ gk_graph_ComputeBestFOrdering()

void gk_graph_ComputeBestFOrdering	(	gk_graph_t *	graph,
		int	v,
		int	type,
		int32_t **	r_perm,
		int32_t **	r_iperm
	)

This function computes a permutation of the vertices based on a best-first-traversal. It can be used for re-ordering the graph to reduce its bandwidth for better cache locality.

Parameters

Definition at line 887 of file GKlib/graph.c.

◆ gk_graph_ComputeBestFOrdering0()

void gk_graph_ComputeBestFOrdering0	(	gk_graph_t *	graph,
		int	v,
		int	type,
		int32_t **	r_perm,
		int32_t **	r_iperm
	)

This function computes a permutation of the vertices based on a best-first-traversal. It can be used for re-ordering the graph to reduce its bandwidth for better cache locality.

Parameters

Definition at line 762 of file GKlib/graph.c.

◆ gk_graph_ComputeBFSOrdering()

void gk_graph_ComputeBFSOrdering	(	gk_graph_t *	graph,
		int	v,
		int32_t **	r_perm,
		int32_t **	r_iperm
	)

This function computes a permutation of the vertices based on a breadth-first-traversal. It can be used for re-ordering the graph to reduce its bandwidth for better cache locality. The algorithm used is a simplified version of the method used to find the connected components.

Parameters

Definition at line 665 of file GKlib/graph.c.

◆ gk_graph_Create()

gk_graph_t* gk_graph_Create ( )

Allocate memory for a graph and initializes it

Returns: the allocated graph. The various fields are set to NULL.

Definition at line 19 of file GKlib/graph.c.

◆ gk_graph_Dup()

gk_graph_t* gk_graph_Dup ( gk_graph_t * graph )

Returns a copy of a graph.

Parameters

graph is the graph to be duplicated.

Returns: the newly created copy of the graph.

Definition at line 340 of file GKlib/graph.c.

◆ gk_graph_ExtractSubgraph()

gk_graph_t* gk_graph_ExtractSubgraph	(	gk_graph_t *	graph,
		int	vstart,
		int	nvtxs
	)

Returns a subgraph containing a set of consecutive vertices.

Parameters

graph	is the original graph.
vstart	is the starting vertex.
nvtxs	is the number of vertices from vstart to extract.

Returns: the newly created subgraph.

Definition at line 391 of file GKlib/graph.c.

◆ gk_graph_FindComponents()

int gk_graph_FindComponents	(	gk_graph_t *	graph,
		int32_t *	cptr,
		int32_t *	cind
	)

This function finds the connected components in a graph.

Parameters

graph	is the graph structure
cptr	is the ptr structure of the CSR representation of the components. The length of this vector must be graph->nvtxs+1.
cind	is the indices structure of the CSR representation of the components. The length of this vector must be graph->nvtxs.

Returns: the number of components that it found.

Note: The cptr and cind parameters can be NULL, in which case only the number of connected components is returned.

Definition at line 572 of file GKlib/graph.c.

◆ gk_graph_Free()

void gk_graph_Free ( gk_graph_t ** graph )

Frees all the memory allocated for a graph.

Parameters

graph is the graph to be freed.

Definition at line 48 of file GKlib/graph.c.

◆ gk_graph_FreeContents()

void gk_graph_FreeContents ( gk_graph_t * graph )

Frees only the memory allocated for the graph's different fields and sets them to NULL.

Parameters

graph is the graph whose contents will be freed.

Definition at line 63 of file GKlib/graph.c.

◆ gk_graph_Init()

void gk_graph_Init ( gk_graph_t * graph )

Initializes the graph.

Parameters

graph is the graph to be initialized.

Definition at line 36 of file GKlib/graph.c.

◆ gk_graph_Read()

gk_graph_t* gk_graph_Read	(	char *	filename,
		int	format,
		int	isfewgts,
		int	isfvwgts,
		int	isfvsizes
	)

Reads a sparse graph from the supplied file

Parameters

filename	is the file that stores the data.
format	is the graph format. The supported values are: GK_GRAPH_FMT_METIS.
isfewgts	is 1 if the edge-weights should be read as floats
isfvwgts	is 1 if the vertex-weights should be read as floats
isfvsizes	is 1 if the vertex-sizes should be read as floats

Returns: the graph that was read.

Definition at line 85 of file GKlib/graph.c.

◆ gk_graph_Reorder()

gk_graph_t* gk_graph_Reorder	(	gk_graph_t *	graph,
		int32_t *	perm,
		int32_t *	iperm
	)

Returns a graph that has been reordered according to the permutation.

Parameters

Definition at line 460 of file GKlib/graph.c.

◆ gk_graph_SingleSourceShortestPaths()

void gk_graph_SingleSourceShortestPaths	(	gk_graph_t *	graph,
		int	v,
		void **	r_sps
	)

This function computes the single-source shortest path lengths from the root node to all the other nodes in the graph. If the graph is not connected then, the sortest part to the vertices in the other components is -1.

Parameters

Definition at line 1084 of file GKlib/graph.c.

◆ gk_graph_Write()

void gk_graph_Write	(	gk_graph_t *	graph,
		char *	filename,
		int	format
	)

Writes a graph into a file.

Parameters

graph	is the graph to be written,
filename	is the name of the output file.
format	is one of GK_GRAPH_FMT_METIS specifying the format of the output file.

Definition at line 277 of file GKlib/graph.c.

◆ gk_i2cc2i_create_common()

gk_i2cc2i_t* gk_i2cc2i_create_common ( char * alphabet )

This function creates the localizations for the various sequences.

Parameters

string i.e amino acids, nucleotides, sequences

Returns: gk_i2cc2i_t variable

Definition at line 48 of file seq.c.

◆ gk_i32kvsortd()

void gk_i32kvsortd	(	size_t	n,
		gk_i32kv_t *	base
	)

Sorts an array of gk_i32kv_t in decreasing order

Definition at line 190 of file sort.c.

◆ gk_i32kvsorti()

void gk_i32kvsorti	(	size_t	n,
		gk_i32kv_t *	base
	)

Sorts an array of gk_i32kv_t in increasing order

Definition at line 179 of file sort.c.

◆ gk_i32readfile()

int32_t* gk_i32readfile	(	char *	fname,
		gk_idx_t *	r_nlines
	)

This function reads the contents of a file and returns it in the form of an array of int32_t.

Parameters

fname	is the name of the file
r_nlines	is the number of lines in the file. If it is NULL, this information is not returned.

Definition at line 139 of file GKlib/io.c.

◆ gk_i32readfilebin()

int32_t* gk_i32readfilebin	(	char *	fname,
		ssize_t *	r_nelmnts
	)

This function reads the contents of a binary file and returns it in the form of an array of int32_t.

Parameters

fname	is the name of the file
r_nlines	is the number of lines in the file. If it is NULL, this information is not returned.

Definition at line 214 of file GKlib/io.c.

◆ gk_i64kvsortd()

void gk_i64kvsortd	(	size_t	n,
		gk_i64kv_t *	base
	)

Sorts an array of gk_i64kv_t in decreasing order

Definition at line 212 of file sort.c.

◆ gk_i64kvsorti()

void gk_i64kvsorti	(	size_t	n,
		gk_i64kv_t *	base
	)

Sorts an array of gk_i64kv_t in increasing order

Definition at line 201 of file sort.c.

◆ gk_i64readfile()

int64_t* gk_i64readfile	(	char *	fname,
		gk_idx_t *	r_nlines
	)

This function reads the contents of a file and returns it in the form of an array of int64_t.

Parameters

fname	is the name of the file
r_nlines	is the number of lines in the file. If it is NULL, this information is not returned.

Definition at line 177 of file GKlib/io.c.

◆ gk_i64readfilebin()

int64_t* gk_i64readfilebin	(	char *	fname,
		ssize_t *	r_nelmnts
	)

This function reads the contents of a binary file and returns it in the form of an array of int64_t.

Parameters

fname	is the name of the file
r_nlines	is the number of lines in the file. If it is NULL, this information is not returned.

Definition at line 253 of file GKlib/io.c.

◆ gk_idxkvsortd()

void gk_idxkvsortd	(	size_t	n,
		gk_idxkv_t *	base
	)

Sorts an array of gk_idxkv_t in decreasing order

Definition at line 322 of file sort.c.

◆ gk_idxkvsorti()

void gk_idxkvsorti	(	size_t	n,
		gk_idxkv_t *	base
	)

Sorts an array of gk_idxkv_t in increasing order

Definition at line 311 of file sort.c.

◆ gk_idxsortd()

void gk_idxsortd	(	size_t	n,
		gk_idx_t *	base
	)

Sorts an array of gk_idx_t in decreasing order

Definition at line 122 of file sort.c.

◆ gk_idxsorti()

void gk_idxsorti	(	size_t	n,
		gk_idx_t *	base
	)

Sorts an array of gk_idx_t in increasing order

Definition at line 111 of file sort.c.

◆ gk_ifkvkselect()

int gk_ifkvkselect	(	size_t	n,
		int	topk,
		gk_fkv_t *	cand
	)

This function puts the 'topk' smallest values in the beginning of the array

Definition at line 84 of file fkvkselect.c.

◆ gk_ikvsortd()

void gk_ikvsortd	(	size_t	n,
		gk_ikv_t *	base
	)

Sorts an array of gk_ikv_t in decreasing order

Definition at line 168 of file sort.c.

◆ gk_ikvsorti()

void gk_ikvsorti	(	size_t	n,
		gk_ikv_t *	base
	)

Sorts an array of gk_ikv_t in increasing order

Definition at line 157 of file sort.c.

◆ gk_isortd()

void gk_isortd	(	size_t	n,
		int *	base
	)

Sorts an array of integers in decreasing order

Definition at line 56 of file sort.c.

◆ gk_isorti()

void gk_isorti	(	size_t	n,
		int *	base
	)

Sorts an array of integers in increasing order

Definition at line 45 of file sort.c.

◆ gk_ispow2()

int gk_ispow2 ( int )

Definition at line 95 of file GKlib/util.c.

◆ gk_log2()

int gk_log2 ( int )

Definition at line 83 of file GKlib/util.c.

◆ gk_malloc()

void* gk_malloc	(	size_t	nbytes,
		char *	msg
	)

This function is my wrapper around malloc that provides the following enhancements over malloc: It always allocates one byte of memory, even if 0 bytes are requested. This is to ensure that checks of returned values do not lead to NULL due to 0 bytes requested. It zeros-out the memory that is allocated. This is for a quick init of the underlying datastructures.

Definition at line 140 of file memory.c.

◆ gk_malloc_cleanup()

void gk_malloc_cleanup ( int showstats )

This function frees the memory that has been allocated since the last call to gk_malloc_init().

Definition at line 118 of file memory.c.

◆ gk_malloc_init()

int gk_malloc_init ( )

This function initializes tracking of heap allocations.

Definition at line 99 of file memory.c.

◆ gk_mcoreAdd()

void gk_mcoreAdd	(	gk_mcore_t *	mcore,
		int	type,
		size_t	nbytes,
		void *	ptr
	)

Adds a memory allocation at the end of the list.

Definition at line 265 of file mcore.c.

◆ gk_mcoreCreate()

gk_mcore_t* gk_mcoreCreate ( size_t coresize )

This function creates an mcore

Definition at line 18 of file mcore.c.

◆ gk_mcoreDel()

void gk_mcoreDel	(	gk_mcore_t *	mcore,
		void *	ptr
	)

This function deletes the mop associated with the supplied pointer. The mop has to be a heap allocation, otherwise it fails violently.

Definition at line 345 of file mcore.c.

◆ gk_mcoreDestroy()

void gk_mcoreDestroy	(	gk_mcore_t **	r_mcore,
		int	showstats
	)

This function destroys an mcore.

Definition at line 67 of file mcore.c.

◆ gk_mcoreMalloc()

void* gk_mcoreMalloc	(	gk_mcore_t *	mcore,
		size_t	nbytes
	)

This function allocate space from the core/heap

Definition at line 140 of file mcore.c.

◆ gk_mcorePop()

void gk_mcorePop ( gk_mcore_t * mcore )

This function frees all mops since the last push

Definition at line 198 of file mcore.c.

◆ gk_mcorePush()

void gk_mcorePush ( gk_mcore_t * mcore )

This function sets a marker in the stack of malloc ops to be used subsequently for freeing purposes

Definition at line 175 of file mcore.c.

◆ gk_mkpath()

int gk_mkpath ( char * )

Definition at line 207 of file fs.c.

◆ gk_NonLocalExit_Handler()

void gk_NonLocalExit_Handler ( int signum )

Definition at line 166 of file error.c.

◆ gk_randinit()

void gk_randinit ( uint64_t seed )

Create the various random number functions

GKlib's built in random number generator for portability across different architectures

Definition at line 71 of file random.c.

◆ gk_randint32()

uint32_t gk_randint32 ( void )

Definition at line 125 of file random.c.

◆ gk_randint64()

uint64_t gk_randint64 ( void )

Definition at line 84 of file random.c.

◆ gk_RandomPermute()

void gk_RandomPermute	(	size_t	,
		int *	,
		int
	)

Definition at line 20 of file GKlib/util.c.

◆ gk_readfile()

char** gk_readfile	(	char *	fname,
		gk_idx_t *	r_nlines
	)

This function reads the contents of a text file and returns it in the form of an array of strings.

Parameters

fname	is the name of the file
r_nlines	is the number of lines in the file. If it is NULL, this information is not returned.

Definition at line 103 of file GKlib/io.c.

◆ gk_readpdbfile()

pdbf* gk_readpdbfile ( char * fname )

Reads a pdb file into a pdbf structure.

This function allocates a pdbf structure and reads the file fname into that structure.

Parameters

fname is the file name to be read

Returns: A filled pdbf structure.

Definition at line 132 of file pdb.c.

◆ gk_realloc()

void* gk_realloc	(	void *	oldptr,
		size_t	nbytes,
		char *	msg
	)

Definition at line 172 of file memory.c.

◆ gk_rmpath()

int gk_rmpath ( char * )

Definition at line 219 of file fs.c.

◆ gk_rw_PageRank()

int gk_rw_PageRank	(	gk_csr_t *	mat,
		float	lamda,
		float	eps,
		int	max_niter,
		float *	pr
	)

Computes the (personalized) page-rank of the vertices in a graph.

Parameters

mat	is the matrix storing the graph.
lamda	is the restart probability.
eps	is the error tolerance for convergance.
max_niter	is the maximum number of allowed iterations.
pr	on entry stores the restart distribution of the vertices. This allows for the computation of personalized page-rank scores by appropriately setting that parameter. On return, pr stores the computed page ranks.

Returns: the number of iterations that were performed.

Definition at line 29 of file rw.c.

◆ gk_seq_init()

void gk_seq_init ( gk_seq_t * seq )

Definition at line 27 of file seq.c.

◆ gk_seq_ReadGKMODPSSM()

gk_seq_t* gk_seq_ReadGKMODPSSM ( char * filename )

This function reads a pssm in the format of gkmod pssm.

Parameters

file_name is the name of the pssm file

Returns: gk_seq_t

Definition at line 83 of file seq.c.

◆ gk_set_exit_on_error()

void gk_set_exit_on_error ( int value )

This function sets the gk_exit_on_error variable

Definition at line 43 of file error.c.

◆ gk_SetSignalHandlers()

void gk_SetSignalHandlers ( )

Definition at line 145 of file error.c.

◆ gk_showcorruption()

void gk_showcorruption ( pdbf * p )

Decodes the corruption bitswitch and prints any problems.

Due to the totally unreliable nature of the pdb format, reading a pdb file stores a corruption bitswitch, and this function decodes that switch and prints the result on stdout.

Parameters

p	is the pdb structure to write out.
fname	is the file name to be written.

Definition at line 435 of file pdb.c.

◆ gk_sigthrow()

void gk_sigthrow ( int signum )

This function is the custome signal handler, which all it does is to perform a longjump to the most recent saved environment

Definition at line 135 of file error.c.

◆ gk_sigtrap()

int gk_sigtrap ( )

This function sets a number of signal handlers and sets the return point of a longjmp

Definition at line 98 of file error.c.

◆ gk_siguntrap()

int gk_siguntrap ( )

This function sets the handlers for the signals to their default handlers

Definition at line 116 of file error.c.

◆ gk_skvsortd()

void gk_skvsortd	(	size_t	n,
		gk_skv_t *	base
	)

Sorts an array of gk_skv_t in decreasing order

Definition at line 300 of file sort.c.

◆ gk_skvsorti()

void gk_skvsorti	(	size_t	n,
		gk_skv_t *	base
	)

Sorts an array of gk_skv_t in increasing order

Definition at line 289 of file sort.c.

◆ gk_str2time()

time_t gk_str2time ( char * str )

Converts a date/time string into its equivalent time_t value.

This function takes date and/or time specification and converts it in the equivalent time_t representation. The conversion is done using the strptime() function. The format that gk_str2time() understands is mm/dd/yyyy hh:mm:ss, in which the hours are in military time.

Parameters

str	is the date/time string to be converted.

Returns: If the conversion was successful it returns the time, otherwise it returns -1.

Definition at line 499 of file string.c.

◆ gk_strcasecmp()

int gk_strcasecmp	(	char *	s1,
		char *	s2
	)

Case insensitive string comparison.

This function compares two strings for equality by ignoring the case of the strings.

Warning: This function is not equivalent to a case-insensitive strcmp() function, as it does not return ordering information.

Todo:: Remove the above warning.

Parameters

s1	is the first string to be compared.
s2	is the second string to be compared.

Return values

1	if the strings are identical,
0	otherwise.

Definition at line 405 of file string.c.

◆ gk_strchr_replace()

char* gk_strchr_replace	(	char *	str,
		char *	fromlist,
		char *	tolist
	)

Replaces certain characters in a string.

This function takes a string and replaces all the characters in the fromlist with the corresponding characters from the tolist. That is, each occurence of fromlist[i] is replaced by tolist[i]. If the tolist is shorter than fromlist, then the corresponding characters are deleted. The modifications on str are done in place. It tries to provide a functionality similar to Perl's tr// function.

Parameters

str	is the string whose characters will be replaced.
fromlist	is the set of characters to be replaced.
tolist	is the set of replacement characters .

Returns: A pointer to str itself.

Definition at line 37 of file string.c.

◆ gk_strdup()

char* gk_strdup ( char * orgstr )

Duplicates a string.

This function is a replacement for C's standard strdup() function. The key differences between the two are that gk_strdup():

uses the dynamic memory allocation routines of GKlib.
it correctly handles NULL input strings.

The string that is returned must be freed by gk_free().

Parameters

orgstr is the string that will be duplicated.

Returns: A pointer to the newly created string.

See also: gk_free()

Definition at line 372 of file string.c.

◆ gk_strerror()

char* gk_strerror ( int errnum )

Thread-safe implementation of strerror()

Definition at line 175 of file error.c.

◆ gk_strhprune()

char* gk_strhprune	(	char *	str,
		char *	rmlist
	)

Prunes characters from the beginning of the string.

This function removes any starting characters that are included in the rmlist. The trimming stops at the first character that is not in rmlist. This function can be used to removed leading spaces, tabs, etc. This is a distructive operation as it modifies the string.

Parameters

str	is the string that will be trimmed.
rmlist	contains the set of characters that will be removed.

Returns: A pointer to str itself.

See also: gk_strtprune()

Definition at line 291 of file string.c.

◆ gk_strrcmp()

int gk_strrcmp	(	char *	s1,
		char *	s2
	)

Compare two strings in revere order.

This function is similar to strcmp but it performs the comparison as if the two strings were reversed.

Parameters

s1	is the first string to be compared.
s2	is the second string to be compared.

Return values

-1,0,1,if the s1 < s2, s1 == s2, or s1 > s2.

Definition at line 433 of file string.c.

◆ gk_strstr_replace()

int gk_strstr_replace	(	char *	str,
		char *	pattern,
		char *	replacement,
		char *	options,
		char **	new_str
	)

Regex-based search-and-replace function.

This function is a C implementation of Perl's s// regular-expression based substitution function.

Parameters

str	is the input string on which the operation will be performed.
pattern	is the regular expression for the pattern to be matched for substitution.
replacement	is the replacement string, in which the possible captured pattern substrings are referred to as $1, $2, ..., $9. The entire matched pattern is refered to as $0.
options	is a string specified options for the substitution operation. Currently the `"i"` (case insensitive) and `"g"` (global substitution) are supported.
new_str	is a reference to a pointer that will store a pointer to the newly created string that results from the substitutions. This string is allocated via gk_malloc() and needs to be freed using gk_free(). The string is returned even if no substitutions were performed.

Returns: If successful, it returns 1 + the number of substitutions that were performed. Thus, if no substitutions were performed, the returned value will be 1. Otherwise it returns 0. In case of error, a meaningful error message is returned in newstr, which also needs to be freed afterwards.

Definition at line 94 of file string.c.

◆ gk_strtokenize()

void gk_strtokenize	(	char *	line,
		char *	delim,
		gk_Tokens_t *	tokens
	)

Definition at line 22 of file tokenizer.c.

◆ gk_strtolower()

char* gk_strtolower ( char * str )

Converts a string to lower case.

This function converts a string to lower case. This operation modifies the string itself.

Parameters

str	is the string whose case will be changed.

Returns: A pointer to str itself.

See also: gk_strtoupper()

Definition at line 348 of file string.c.

◆ gk_strtoupper()

char* gk_strtoupper ( char * str )

Converts a string to upper case.

This function converts a string to upper case. This operation modifies the string itself.

Parameters

str	is the string whose case will be changed.

Returns: A pointer to str itself.

See also: gk_strtolower()

Definition at line 328 of file string.c.

◆ gk_strtprune()

char* gk_strtprune	(	char *	str,
		char *	rmlist
	)

Prunes characters from the end of the string.

This function removes any trailing characters that are included in the rmlist. The trimming stops at the last character (i.e., first character from the end) that is not in rmlist. This function can be used to removed trailing spaces, newlines, etc. This is a distructive operation as it modifies the string.

Parameters

str	is the string that will be trimmed.
rmlist	contains the set of characters that will be removed.

Returns: A pointer to str itself.

See also: gk_strhprune()

Definition at line 254 of file string.c.

◆ gk_threetoone()

char gk_threetoone ( char * res )

Converts three-letter amino acid codes to one-leter codes.

This function takes a three letter * and converts it to a single

Parameters

res	is the three-letter code to be converted.

Returns: A representing the amino acid.

Definition at line 24 of file pdb.c.

◆ gk_time2str()

char* gk_time2str ( time_t time )

Converts a time_t time into a string.

This function takes a time_t-specified time and returns a string-formated representation of the corresponding time. The format of the string is mm/dd/yyyy hh:mm:ss, in which the hours are in military time.

Parameters

time	is the time to be converted.

Returns: It returns a pointer to a statically allocated string that is over-written in successive calls of this function. If the conversion failed, it returns NULL.

Definition at line 470 of file string.c.

◆ gk_UnsetSignalHandlers()

void gk_UnsetSignalHandlers ( )

Definition at line 155 of file error.c.

◆ gk_WClockSeconds()

gk_wclock_t gk_WClockSeconds ( void )

Definition at line 19 of file timers.c.

◆ gk_writealphacarbons()

void gk_writealphacarbons	(	pdbf *	p,
		char *	fname
	)

Writes out all the alpha carbon atoms of a structure.

This function takes a pdbf structure p and writes only the alpha carbon atoms to a filename fname.

Parameters

p	is the pdb structure to write out.
fname	is the file name to be written.

Definition at line 413 of file pdb.c.

◆ gk_writebackbone()

void gk_writebackbone	(	pdbf *	p,
		char *	fname
	)

Writes out all the backbone atoms of a structure in pdb format.

This function takes a pdbf structure p and writes only the backbone atoms to a filename fname.

Parameters

p	is the pdb structure to write out.
fname	is the file name to be written.

Definition at line 392 of file pdb.c.

◆ gk_writefullatom()

void gk_writefullatom	(	pdbf *	p,
		char *	fname
	)

Writes all atoms in p in pdb-format to file fname.

This function takes a pdbf structure and writes out all the atom information to file fname.

Parameters

p	is the pdbf structure to write out
fname	is the file name to be written

Definition at line 371 of file pdb.c.

◆ gk_zkvsortd()

void gk_zkvsortd	(	size_t	n,
		gk_zkv_t *	base
	)

Sorts an array of gk_zkv_t in decreasing order

Definition at line 234 of file sort.c.

◆ gk_zkvsorti()

void gk_zkvsorti	(	size_t	n,
		gk_zkv_t *	base
	)

Sorts an array of gk_zkv_t in increasing order

Definition at line 223 of file sort.c.

◆ GKDecodeBase64()

void GKDecodeBase64	(	int	nbytes,
		unsigned char *	inbuffer,
		unsigned char *	outbuffer
	)

Definition at line 85 of file b64.c.

◆ GKEncodeBase64()

void GKEncodeBase64	(	int	nbytes,
		unsigned char *	inbuffer,
		unsigned char *	outbuffer
	)

Definition at line 64 of file b64.c.

◆ HTable_Create()

gk_HTable_t* HTable_Create ( int nelements )

Definition at line 16 of file htable.c.

◆ HTable_Delete()

void HTable_Delete	(	gk_HTable_t *	htable,
		int	key
	)

Definition at line 107 of file htable.c.

◆ HTable_Destroy()

void HTable_Destroy ( gk_HTable_t * htable )

Definition at line 235 of file htable.c.

◆ HTable_GetNext()

int HTable_GetNext	(	gk_HTable_t *	htable,
		int	key,
		int *	val,
		int	type
	)

Definition at line 162 of file htable.c.

◆ HTable_HFunction()

int HTable_HFunction	(	int	nelements,
		int	key
	)

Definition at line 244 of file htable.c.

◆ HTable_Insert()

void HTable_Insert	(	gk_HTable_t *	htable,
		int	key,
		int	val
	)

Definition at line 74 of file htable.c.

◆ HTable_Reset()

void HTable_Reset ( gk_HTable_t * htable )

Definition at line 33 of file htable.c.

◆ HTable_Resize()

void HTable_Resize	(	gk_HTable_t *	htable,
		int	nelements
	)

Definition at line 46 of file htable.c.

◆ HTable_Search()

int HTable_Search	(	gk_HTable_t *	htable,
		int	key
	)

Definition at line 135 of file htable.c.

◆ HTable_SearchAndDelete()

int HTable_SearchAndDelete	(	gk_HTable_t *	htable,
		int	key
	)

Definition at line 200 of file htable.c.

◆ omp_get_dynamic()

int omp_get_dynamic ( void )

◆ omp_get_max_threads()

int omp_get_max_threads ( void )

◆ omp_get_nested()

int omp_get_nested ( void )

◆ omp_get_num_procs()

int omp_get_num_procs ( void )

◆ omp_get_num_threads()

int omp_get_num_threads ( void )

◆ omp_get_thread_num()

int omp_get_thread_num ( void )

◆ omp_in_parallel()

int omp_in_parallel ( void )

◆ omp_set_dynamic()

void omp_set_dynamic ( int num_threads )

◆ omp_set_nested()

void omp_set_nested ( int nested )

◆ omp_set_num_threads()

void omp_set_num_threads ( int num_threads )

◆ PrintBackTrace()

void PrintBackTrace ( )

Definition at line 198 of file error.c.

Functions

Detailed Description

Function Documentation

◆ ComputeAccuracy()

◆ ComputeMean()

◆ ComputeMedianRFP()

◆ ComputeROCn()

◆ ComputeStdDev()

◆ decodeblock()

◆ encodeblock()

◆ errexit()

◆ gk_AllocMatrix()

◆ gk_array2csr()

◆ gk_ckvsortd()

◆ gk_ckvsorti()

◆ gk_CPUSeconds()

◆ gk_csortd()

◆ gk_csorti()

◆ gk_csr_CompactColumns()

◆ gk_csr_ComputeSimilarity()

◆ gk_csr_ComputeSquaredNorms()

◆ gk_csr_ComputeSums()

◆ gk_csr_Create()

◆ gk_csr_CreateIndex()

◆ gk_csr_Dup()

◆ gk_csr_ExtractPartition()

◆ gk_csr_ExtractRows()

◆ gk_csr_ExtractSubmatrix()

◆ gk_csr_Free()

◆ gk_csr_FreeContents()

◆ gk_csr_GetSimilarRows()

◆ gk_csr_Init()

◆ gk_csr_LowFilter()

◆ gk_csr_Normalize()

◆ gk_csr_Prune()

◆ gk_csr_Read()

◆ gk_csr_Scale()

◆ gk_csr_SortIndices()

◆ gk_csr_Split()

◆ gk_csr_TopKPlusFilter()

◆ gk_csr_Write()

◆ gk_csr_ZScoreFilter()

◆ gk_dexists()

◆ gk_dfkvkselect()

◆ gk_dkvsortd()

◆ gk_dkvsorti()

◆ gk_dreadfilebin()

◆ gk_dsortd()

◆ gk_dsorti()

◆ gk_errexit()

◆ gk_fclose()

◆ gk_fexists()

◆ gk_find_frequent_itemsets()

◆ gk_fkvsortd()

◆ gk_fkvsorti()

◆ gk_flog2()

◆ gk_fopen()

◆ gk_freadfilebin()

◆ gk_free()

◆ gk_FreeMatrix()

◆ gk_freepdbf()

◆ gk_freetokenslist()

◆ gk_fsortd()

◆ gk_fsorti()

◆ gk_fwritefilebin()

◆ gk_getbasename()

◆ gk_GetCurMemoryUsed()

◆ gk_getextname()

◆ gk_getfilename()

◆ gk_getfilestats()

◆ gk_getfsize()

◆ gk_getline()

◆ gk_GetMaxMemoryUsed()

◆ gk_getpathname()

◆ gk_GetStringID()

◆ gk_gkmcoreAdd()

◆ gk_gkmcoreCreate()

◆ gk_gkmcoreDel()

◆ gk_gkmcoreDestroy()

◆ gk_gkmcorePop()