int main(int argc, char** argv) {
/**
* format: ./convpoly [*.csv]
*/
polygons p;
if(argc<2) {
readpolygons(&p);
printpolygons(&p);
}
else {
if(access(argv[1], F_OK) == -1) {
fprintf(stderr, "Error: file %s does not exist.\n", argv[1]);
exit(1);
}
/* Read and convert the csv file to a polygons type */
FILE* f;
f = fopen(argv[1], "r");
readcsv(f, &p);
fclose(f);
writepolygons(&p);
printf("Wrote to polygons/all.ply.\n");
}
return 0;
}
int main(int argc, char** argv) {
char* content;
Uint contentlen,i,j;
stringset_t *set, *set2, **csv;
int space;
content = readfile(&space, "casessmall.rtxt", &contentlen);
printf("read file of length: %d\n", contentlen);
set = tokensToStringset(&space, "\n", content, contentlen);
FREEMEMORY(&space, content);
printf("file contained %d lines\n", set->noofstrings);
for(i=0; i < set->noofstrings; i++) {
set2 = tokensToStringset(&space, ",", set->strings[i].str, set->strings[i].len);
for(j=0; j < set2->noofstrings; j++) {
printf("%s,", set2->strings[j].str);
}
printf("\n");
destructStringset(&space, set2);
}
destructStringset(&space, set);
printf("reading csv ...\n");
csv = readcsv(&space, "cases.rtxt", ",", &contentlen);
for (i=0; i < contentlen; i++) {
for(j=0; j < csv[i]->noofstrings; j++) {
printf("%s",csv[i]->strings[j].str);
}
destructStringset(&space, csv[i]);
printf("\n");
}
FREEMEMORY(&space, csv);
return EXIT_SUCCESS;
}
int
main(int argc, char** argv)
{
Uint noofvecs, i;
Sint optindex, c;
vector_t info;
#ifdef MEMMAN_H
Spacetable spacetab;
#endif
void *space = NULL;
char *url = NULL;
char *outpath = NULL;
char *pveclistfile = NULL;
char *alphabetfile = NULL;
char *vecext="vec";
char *seqext="seq";
struct prob_vec *p_vec;
IntSequence *sequence;
FAlphabet *alphabet;
stringset_t *tok;
stringset_t **fn;
#ifdef MEMMAN_H
initmemoryblocks(&spacetab, 1000);
space = &spacetab;
#endif
while(1) {
c=getopt_long(argc, argv, "f:a:o:", long_options, &optindex);
if (c==-1) break;
switch(c) {
case 'f':
pveclistfile = optarg;
break;
case 'a':
alphabetfile = optarg;
break;
case 'o':
outpath = optarg;
break;
default:
usage(argv[0]);
exit (EXIT_FAILURE);
}
}
if (pveclistfile==NULL || alphabetfile == NULL) {
usage(argv[0]);
exit (EXIT_FAILURE);
}
fn=readcsv(space, pveclistfile, ".", &noofvecs);
alphabet = loadCSValphabet(space, alphabetfile);
sortMapdomain(space, alphabet);
for(i=0; i<noofvecs; i++)
{
INITVECTOR(&info);
SETSTR(fn[i],0) = concatdelim(space, SETSTR(fn[i],0), vecext,
SETSTRLEN(fn[i],0), 3, '.');
p_vec = prob_vec_read (SETSTR(fn[i],0));
if (p_vec->mship == NULL)
prob_vec_expand(p_vec);
sequence = encode_prob_vec(space, alphabet, p_vec, 0, 0,
cantorchar, &info);
sequence->info = (Uint*) info.elements;
sequence->namelen= strlen(alphabetfile);
COPYSTR(space, sequence->alphabetname, alphabetfile,
strlen(alphabetfile));
/*this is a potential security risk*/
if (p_vec->compnd_len > 0) {
sequence->descrlen = p_vec->compnd_len-1;
COPYSTR(space, sequence->description, p_vec->compnd,
p_vec->compnd_len-1);
} else {
sequence->descrlen = 14;
COPYSTR(space, sequence->description, "descriptor n/a", 14);
}
sequence->urllen = SETSTRLEN(fn[i],0);
COPYSTR(space, sequence->url, SETSTR(fn[i],0),
SETSTRLEN(fn[i],0));
SETSTR(fn[i],0) = concatdelim(space, SETSTR(fn[i],0), seqext,
SETSTRLEN(fn[i],0), 3, '.');
SETSTRLEN(fn[i],0) += 4;
if (outpath) {
tok = tokensToStringset(space, "/", SETSTR(fn[i],0),
SETSTRLEN(fn[i],0));
COPYSTR(space, url, outpath, strlen(outpath));
url = concat(space, url, SETSTR(tok, tok->noofstrings-1),
strlen(url), SETSTRLEN(tok, tok->noofstrings-1));
saveSequence(sequence, url);
destructStringset(space, tok);
FREEMEMORY(space, url);
url = NULL;
} else {
saveSequence(sequence, SETSTR(fn[i],0));
}
destructSequence (space, sequence);
prob_vec_destroy (p_vec);
destructStringset (space, fn[i]);
progressBarVT("probability vectors converted", noofvecs-1, i, 25);
}
printf("\nexit.\n");
FREEMEMORY(space, fn);
destructAlphabet(space, alphabet);
return EXIT_SUCCESS;
}
int
main (int argc, char** argv)
{
Sint optindex, c;
unsigned char depictsw=0;
unsigned char wurst=1;
unsigned char gnuplot=0;
Uint i, j, noofseqs=0, nooffreqs=0, noofqueries=0;
Uint noofhits=100;
Uint substrlen = 10;
Uint minseeds = 5;
Uint maxmatches = 10000;
char *seq, *vec, *bin;
imbissinfo *imbiss;
void *space = NULL;
double *scores = NULL;
int swscores[2]={3,-2};
char *pveclistfile=NULL;
char *alphabetfile=NULL;
char *inputfile=NULL;
char *batchfile = NULL;
char *subfile=NULL;
char *reportfile = NULL;
int (*handler)
(void *, Matchtype *, IntSequence **, Uint,
Uint, void *) = allscores;
double (*filter)
(void *, Matchtype *, IntSequence *, IntSequence *,
Uint *, Uint, Uint, void *) = swconstfilter;
Matchtype* (*select)
(void *, Matchtype *, Uint k,
IntSequence *, IntSequence **, void *) = selectSW;
stringset_t **fn, **freq, *queryurl, **queries=NULL;
Suffixarray *arr = NULL;
IntSequence **sequences = NULL;
IntSequence *input = NULL;
FAlphabet *alphabet = NULL;
PairSint *matches=NULL;
Uint percent=0;
time_t startsuf, endsuf;
double difsuf, difmatch, difrank;
#ifdef MEMMAN_H
Spacetable spacetab;
initmemoryblocks(&spacetab, 100000);
space = &spacetab;
#endif
while(1)
{
c=getopt_long(argc, argv, "SAghFGBLM:D:r:m:x:n:p:b:s:a:q:l:c:dvw",
long_options, &optindex);
if (c==-1) break;
switch(c) {
case 'r':
reportfile=optarg;
break;
case 'v':
verbose_flag=1;
break;
case 'd':
depictsw = 1;
break;
case 's':
pveclistfile = optarg;
break;
case 'a':
alphabetfile = optarg;
break;
case 'q':
inputfile = optarg;
noofqueries = 1;
break;
case 'l':
substrlen = atoi(optarg);
break;
case 'c':
minseeds = atoi(optarg);
break;
case 'b':
batchfile = optarg;
break;
case 'p':
percent = atoi(optarg);
break;
case 'x':
subfile = optarg;
break;
case 'n':
noofhits = atoi(optarg);
break;
case 'w':
wurst = 0;
break;
case 'B':
filter = scorefilter;
select = selectBlastScore;
break;
case 'S':
filter = scorefilter;
select = selectScore;
break;
case 'A':
filter = swconstfilter;
select = selectSW;
break;
case 'F':
filter = scorefilter;
select = selectScoreSWconst;
break;
case 'G':
filter = scorefilter;
select = selectBlastScoreSWconst;
break;
case 'M':
swscores[0]=atoi(optarg);
break;
case 'L':
handler = latexscores;
break;
case 'D':
swscores[1]=atoi(optarg);
break;
case 'g':
gnuplot = 1;
break;
case 'm':
maxmatches=atoi(optarg);
break;
case 'h':
default:
usage(argv[0]);
exit (EXIT_FAILURE);
}
}
if (pveclistfile==NULL || (inputfile == NULL && batchfile==NULL)
|| alphabetfile == NULL) {
usage(argv[0]);
exit (EXIT_FAILURE);
}
imbiss = ALLOCMEMORY(space, NULL, imbissinfo, 1);
imbiss->reportfile = reportfile;
imbiss->swscores = swscores;
imbiss->noofhits = noofhits;
imbiss->minseeds = minseeds;
imbiss->wurst = wurst;
/*read batch file*/
if (batchfile) {
queries = readcsv(space, batchfile, "", &noofqueries);
}
/*read substitution matrix*/
if (subfile) {
freq=readcsv(space, subfile,",", &nooffreqs);
scores = ALLOCMEMORY(space, NULL, double, ((nooffreqs-1)*(nooffreqs-1)) );
for(i=1; i < nooffreqs; i++) {
for(j=1; j < nooffreqs; j++) {
if(strcmp(SETSTR(freq[i],j),"inf")==0){
MATRIX2D(scores, nooffreqs-1, i, j)=0;
}else{
MATRIX2D(scores, nooffreqs-1, i, j)=atof(SETSTR(freq[i],j));
}
}
}
}
/*read alphabet*/
if (alphabetfile != NULL) {
alphabet = loadCSValphabet(space, alphabetfile);
sortMapdomain(space, alphabet);
}
/*load sequence database*/
fn=readcsv(space, pveclistfile, "", &noofseqs);
sequences = ALLOCMEMORY(space, NULL, IntSequence *, noofseqs);
for(i=0; i < noofseqs; i++)
{
sequences[i] = loadSequence(space, SETSTR(fn[i],0));
}
for (i=0; i < noofseqs; i++) {
destructStringset(space, fn[i]);
}
FREEMEMORY(space, fn);
/*construct the suffix array*/
time (&startsuf);
arr = constructSufArr(space, sequences, noofseqs, NULL);
constructLcp(space, arr);
time (&endsuf);
difsuf = difftime (endsuf, startsuf);
/*do search*/
for (i=0; i < noofqueries; i++) {
/*get query form batchfile*/
if (queries) {
inputfile = SETSTR(queries[i],0);
}
/*typically only used with batchfile*/
if (percent != 0) {
substrlen =
((double)((double)input->length/100)*(double) percent);
}
input = loadSequence(space, inputfile);
//seq = printSequence (space, input, 60);
printf(">IMBISS order delivered\n");
//printf("%s\n",seq);
printf("%s\n", input->url);
//FREEMEMORY(space, seq);
time (&startsuf);
matches=sufSubstring(space, arr, input->sequence, input->length, substrlen);
time (&endsuf);
difmatch = difftime (endsuf, startsuf);
/*get prob vector url for salami/wurst*/
//printf("%.*s\n", 5, input->url + 58);
vec = malloc(sizeof(char)*66);
sprintf(vec, "/smallfiles/public/no_backup/bm/pdb_all_vec_6mer_struct/%5s.vec\0", input->url+56);
bin = malloc(sizeof(char)*54);
sprintf(bin, "/smallfiles/public/no_backup/bm/pdb_all_bin/%5s.bin\0", input->url+56);
queryurl = initStringset(space);
addString(space, queryurl, bin, strlen(bin));
addString(space, queryurl, vec, strlen(vec));
getimbissblast(space, input, sequences, noofseqs, alphabet, imbiss);
imbiss->query = queryurl;
imbiss->substrlen = substrlen;
imbiss->alphabet = alphabet;
/*if a substition file was given ...*/
if (subfile) {
imbiss->sub = createsubmatrix(scores, imbiss->score, nooffreqs-1);
}
/*match 'n' report*/
time (&startsuf);
imbiss->consensus = ALLOCMEMORY(space, NULL, Uint, (input->length-substrlen));
memset(imbiss->consensus, 0, (sizeof(Uint)*(input->length-substrlen)));
rankSufmatch(space, arr, matches, input->length-substrlen,
maxmatches, substrlen,
sequences, noofseqs, filter, select, handler,
input, imbiss, scores, depictsw);
if (gnuplot) {
consensus (space, imbiss->consensus, input->length-substrlen,
input, substrlen, imbiss);
}
time (&endsuf);
difrank = difftime (endsuf, startsuf);
printf ("Building the suffixtree has taken %f seconds.\n", difsuf);
printf ("Match the suffixtree has taken %f seconds.\n", difmatch);
printf ("Rank the suffixtree has taken %f seconds.\n", difrank);
/*partial cleanup*/
//destructStringset(space, queryurl);
destructSequence(space, input);
if(subfile) {
FREEMEMORY(space, imbiss->sub);
}
FREEMEMORY(space, imbiss->consensus);
FREEMEMORY(space, imbiss->score);
FREEMEMORY(space, matches);
free(bin);
free(vec);
}
/*final cleanup*/
for (i=0; i < noofseqs; i++) {
destructSequence(space, sequences[i]);
}
FREEMEMORY(space, sequences);
destructSufArr(space, arr);
#ifdef MEMMAN_H
activeblocks(space);
#endif
printf("Goodbye.\n");
return EXIT_SUCCESS;
}
int main (int argc, char *argv[])
{
if (argc < 3) {
fprintf (stderr, "usage: kmeans <data> <centroids>\n");
return 1;
}
// read data from file
int d_n=0, d_m=0;
double data [MAX_N][MAX_M] = { 0 };
readcsv (argv[1], data, &d_n, &d_m);
// read centroids from file
int c_n=0, c_m=0;
double centroids [MAX_N][MAX_M] = { 0 };
readcsv (argv[2], centroids, &c_n, &c_m);
if (d_n < c_n) {
fprintf (stderr, "error: More clusters than data\n");
return -1;
}
if (d_m != c_m) {
fprintf (stderr, "error: Data and centroid dimensionalities differ\n");
return -1;
}
int clusters [d_n]; // assigned cluster of datum
int cluster_pop [c_n]; // population of cluster
double data_sum [c_n][c_m]; // data column sum of cluster
double cent_last [c_n][c_m]; // centroids (last iteration)
int i, j, k;
double dist, dist_min, c;
//
// Iterate, converge absolutely
do //
{ // For each observation, calculate the distance from
// each centroid and assign it to the nearest one
memset (cluster_pop, 0, c_n*sizeof(int));
for (i = 0; i < d_n; ++i) {
dist_min = INFINITY;
// assume distance minimum is +inf
for (j = 0; j < c_n; ++j) {
dist = 0.0;
// calculate distance to centroid j
for (k = 0; k < d_m; ++k) {
c = data[i][k] - centroids[j][k];
dist += c * c;
}
// minimize distance
if (dist < dist_min) {
dist_min = dist;
clusters[i] = j;
}
}
// add element its nearest cluster
++cluster_pop[clusters[i]];
}
//
// Calculate new centroids
memset (data_sum, 0, c_n*c_m*sizeof(double));
// sum data columns per cluster
for (i = 0; i < d_n; ++i)
for (k = 0; k < d_m; ++k)
data_sum[clusters[i]][k] += data[i][k];
// centroid coordinate is mean of corresponding column
for (j = 0; j < c_n; ++j) {
for (k = 0; k < d_m; ++k) {
cent_last[j][k] = centroids[j][k];
centroids[j][k] = data_sum[j][k] / cluster_pop[j];
}
}
// displacement-->0 (n-->inf)
} while (norm(*centroids, c_n, c_m) - norm(*cent_last, c_n, c_m) > TOL);
//
// Printout final centroid coordinates
printf ("data (%ix%i)\n", d_n, d_m);
for (i = 0; i < d_n; ++i) {
printf ("%5i { ", i);
for (j = 0; j < d_m; ++j)
printf ("%7.3f ", data[i][j]);
printf ("} --> %2i\n", clusters[i]);
}
printf ("centroids (%ix%i)\n", c_n, c_m);
for (j = 0; j < c_n; ++j) {
printf ("%5i { ", j);
for (k = 0; k < c_m; ++k)
printf ("%7.3f ", centroids[j][k]);
printf ("}\n");
}
return 0;
}
