Help!
lguang0121
|
1#
lguang0121 发表于 2007-05-24 19:56
Help!
Help!
我想把C：/temp下的data.txt中的数据给 Kmeans运算，但是怎么才能把 c中的#include改的能用啊 ？ 问题 代码： #!C:/perl/bin/perl use CGI; $query = new CGI; print $query->header; print $query->start_html("Save and Restore"); #@@ print '<body bgcolor="#FFFFFF">'."\n" ; print '<div align="center">'."\n" ; print "\n" ; print '<p align="center"><marquee direction=left bgcolor="#66FF66"><font color="#6699FF"><font face="Comic Sans MS">Welcome to K-means</font></font></marquee></p>'."\n" ; print "\n" ; print '<p align="center"><marquee direction=right bgcolor="#66FF66"><font color="#6699FF"><font face="Comic Sans MS">Welcome to K-means'."\n" ; print ",</font></font></marquee></p>\n" ; #@@ print '<font color="#6699FF"><font face="Comic Sans MS">THE DATA</font></font>'."\n" ; #@@ # Here's where we take action on the previous request &save_parameters($query) if $query->param('action') eq 'SAVE'; $query = &restore_parameters($query) if $query->param('action') eq 'RESTORE'; # Here's where we create the form print $query->start_multipart_form; print '<table border="3" cellpadding="3" cellspacing="0" width="250">'."\n" ; print " <tr>\n" ; print ' <td bgcolor="#008080"><br>'."\n" ; print '<textarea name=comment rows=5 cols=29></textarea>'."\n"; sub rd_parameters { local($query) = @_; local($filename) = "C:/temp/data.txt"; if (open(FILE,">$filename")) { $query->read<FILE,$data>; print $data; close FILE; } } #@@ use Inline C => <<END_C; $data = **data; int c_function() { #ifndef K_MEANS_H #define K_MEANS_H int *k_means(float**, int, int, int, float, float**); #endif /***** ** kmeans.c ** Parameters ** - array of data points (float **data) ** - number of data points (int n) ** - dimension (int m) ** - desired number of clusters (int k) ** - error tolerance (float t) ** - used as the stopping criterion, i.e. when the sum of ** squared euclidean distance (standard error for k-means) ** of an iteration is within the tolerable range from that ** of the previous iteration, the clusters are considered ** "stable", and the function returns ** - a suggested value would be 0.0001 ** - output address for the final centroids (float **centroids) ** - user must make sure the memory is properly allocated, or ** pass the null pointer if not interested in the centroids */ #include <stdlib.h> #include <assert.h> #include <float.h> #include <math.h> #include <stdio.h> #include <string.h> int *k_means (float **data, int n, int m, int k, float t, float **centroids); int loadDataset (float ***data, int *num, int *dim, char *filename); int main (int argc, char **argv) { float **data; int num, dim, ncluster; float e = 0.0005; int *c; int i = 0; if (argc < 3) { printf ("Usage: kmeans -ncluster filename\n"); exit (-1); } if (loadDataset (&data, &num, &dim, argv[2]) != 0) { printf ("Error in loading dataset"); exit (-1); } printf("%d records with %d features has been loaded\n",num,dim); ncluster = atoi (argv[1]); c = k_means (data, num, dim, ncluster, e, 0); for (i = 0; i < num; i++) { fprintf (stderr,"data point %d is in cluster %d\n", i, c[i]); } free(c); for (i = 0; i < num; i++) { free(data[i]); } free(data); return (0); } int * k_means (float **data, int n, int m, int k, float t, float **centroids) { /* * output cluster label for each data point */ int *labels = (int *) malloc (n * sizeof (int)); int h, i, j; /* loop counters, of course :) */ int *counts = (int *) malloc (k * sizeof (int)); /* size of * each * cluster */ float old_error, error = DBL_MAX; /* sum of squared euclidean * distance */ float **c = centroids ? centroids : (float **) malloc (k * sizeof (float *)); float **c1 = (float **) malloc (k * sizeof (float *)); /* temp * centroids */ int nloop = 0; assert (data && k > 0 && k <= n && m > 0 && t >= 0); /* for debugging */ /**** ** initialization */ for (h = i = 0; i < k; h += n / k, i++) { c1[i] = (float *) malloc (m * sizeof (float)); if (!centroids) { c[i] = (float *) malloc (m * sizeof (float)); } /* * pick k points as initial centroids */ for (j = 0; j < m; j++) { c[i][j] = data[h][j]; }; } /**** ** main loop */ do { /* * save error from last step */ nloop++; printf ("Runing loop %d\n", nloop); old_error = error, error = 0; /* * clear old counts and temp centroids */ for (i = 0; i < k; counts[i++] = 0) { for (j = 0; j < m; c1[i][j++] = 0); } for (h = 0; h < n; h++) { /* * identify the closest cluster */ float min_distance = DBL_MAX; for (i = 0; i < k; i++) { float distance = 0; for (j = m; j-- > 0; distance += pow (data[h][j] - c[i][j], 2)); if (distance < min_distance) { labels[h] = i; min_distance = distance; } } /* * update size and temp centroid of the destination cluster */ for (j = m; j-- > 0; c1[labels[h]][j] += data[h][j]); counts[labels[h]]++; /* * update standard error */ error += min_distance; } for (i = 0; i < k; i++) { /* update all centroids */ for (j = 0; j < m; j++) { c[i][j] = counts[i] ? c1[i][j] / counts[i] : c1[i][j]; } } } while (fabs (error - old_error) > t); /**** ** housekeeping */ for (i = 0; i < k; i++) { if (!centroids) { free (c[i]); } free (c1[i]); } if (!centroids) { free (c); } free (c1); free (counts); // printf("End of K_means\n"); return labels; } #define MAX_LINE_SIZE 2000 int loadDataset (float ***DATA, int *NUM, int *DIM, char *filename) { FILE *fp; char line[MAX_LINE_SIZE]; int num = 0, dim = 0; int i=0; float **data; fp = fopen (filename, "r"); if (!fp) { printf ("Cannot open %s\n", filename); return (-1); } while(fgets(line,MAX_LINE_SIZE,fp)) { char* pstr,*word; if((line[0]=='\0')||(line[0]=='#')) continue; num++; if(dim==0) { pstr=line; word=strtok(pstr,"\t"); while((word=strtok(NULL,"\t"))) { dim++; } } } data=(float**)malloc(num*sizeof(float*)); for(i=0;i<num;i++) { data[i]=(float*)malloc(dim*sizeof(float)); } rewind(fp); num=0; dim=0; while(fgets(line,MAX_LINE_SIZE,fp)) { char* pstr,*word; if((line[0]=='\0')||(line[0]=='#')) continue; pstr=line; word=strtok(pstr,"\t"); dim=0; while((word=strtok(NULL,"\t"))) { data[num][dim]=atoi(word); dim++; } num++; } fclose (fp); *DATA = data; *NUM = num; *DIM = dim; return (0); } } END_C c_function; exit 0; #@@ print "<html>\n" ; print "\n" ; print "<head>\n" ; print "</head>\n" ; print "\n" ; #@@ print'<form action="w.pl" method="POST">'."\n"; print '<input type="submit" value="FILE UPLOAD">'."\n"; print"</form>"; #@@ print'<form action="R.pl" method="POST">'."\n"; print '<input type="submit" value="RESULT">'."\n"; print"</form>"; #@@ print "</body>\n" ; print "</html>\n" ; |