-
Notifications
You must be signed in to change notification settings - Fork 0
/
kmeans.c
351 lines (288 loc) · 9.49 KB
/
kmeans.c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
#define PY_SSIZE_T_CLEAN
#include <Python.h> //for linux
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <math.h>
int find_closets_cluster(double *data_point);
void algorithm();
void free_array_2d(double **arr, int r);
void set_equal_array_2d(double **new, double **current, int r, int c);
void zero_array_2d(double **arr, int r, int c);
void calculate_new_centroids();
double **allocate_array_2d(int r, int c);
double calculate_distance_squared(double *a, double *b);
void error_occurred(int condition);
void set_clusters();
int k, max_iter, num_rows, d = 1;
int *num_elements_in_cluster;
double **data_points, **centroids, **new_centroids, **clusters;
double epsilon;
/*
we get PyObject that represents 2 dimensional array and we convert it
to 2 dimensional double array (C object)
*/
void convertPython2DArray(PyObject *python_arr,double **arr,int rows,int columns){
int i=0,j=0;
PyObject *current_list,*item;
for (i = 0; i < rows; i++) {
current_list = PyList_GetItem(python_arr, i);
for(j=0;j<columns;j++){
item = PyList_GetItem(current_list, j);
if (!PyFloat_Check(item)){
arr[i][j] = 0.0;
}
arr[i][j] = PyFloat_AsDouble(item);
}
}
}
/*
we get 2 dimensional array and num of rows and columns,
we return 2 dimensional PyObject
*/
PyObject* createPyObjectFrom2DArray(double **arr,int rows,int columns){
int i=0,j=0;
PyObject *current,*result,*val;
result = PyList_New(k);
for(i=0;i<rows;i++){
current = PyList_New(columns);
for(j=0;j<columns;j++){
val = PyFloat_FromDouble(arr[i][j]);
PyList_SetItem(current, j, val);
}
PyList_SetItem(result, i, current);
}
return result;
}
/*
* This actually defines the geo function using a wrapper C API function
* The wrapping function needs a PyObject* self argument.
* This is a requirement for all functions and methods in the C API.
* It has input PyObject *args from Python.
*/
/*
We perceive the variables from the Python program, initialize the
*/
static PyObject* fit(PyObject *self, PyObject *args){
PyObject *py_centroids,*py_data_points,*result;
/* This parses the Python arguments into a double (d) variable named z and int (i) variable named n*/
if(!PyArg_ParseTuple(args, "iiiidOO", &k, &max_iter,&d,&num_rows,&epsilon,&py_centroids,&py_data_points)) {
return NULL; /* In the CPython API, a NULL value is never valid for a
PyObject* so it is used to signal that an error has occurred. */
}
// printf("K=%d, max_iter=%d, num_rows=%d, d=%d, epsilon=%f\n",k,max_iter,num_rows,d,epsilon);
centroids = allocate_array_2d(k, d);
data_points = allocate_array_2d(num_rows, d);
convertPython2DArray(py_centroids,centroids,k,d);
convertPython2DArray(py_data_points,data_points,num_rows,d);
algorithm();
result = createPyObjectFrom2DArray(centroids,k,d);
free_array_2d(centroids, k);
return result;
}
/*
* This array tells Python what methods this module has.
* We will use it in the next structure
*/
static PyMethodDef capiMethods[] = {
{"fit", /* the Python method name that will be used */
(PyCFunction) fit, /* the C-function that implements the Python function and returns static PyObject* */
METH_VARARGS, /* flags indicating parametersaccepted for this function */
PyDoc_STR("algorithm to find the clusters")}, /* The docstring for the function */
{NULL, NULL, 0, NULL} /* The last entry must be all NULL as shown to act as a
sentinel. Python looks for this entry to know that all
of the functions for the module have been defined. */
};
/* This initiates the module using the above definitions. */
static struct PyModuleDef moduledef = {
PyModuleDef_HEAD_INIT,
"mykmeanssp", /* name of module */
NULL, /* module documentation, may be NULL */
-1, /* size of per-interpreter state of the module, or -1 if the module keeps state in global variables. */
capiMethods /* the PyMethodDef array from before containing the methods of the extension */
};
/*
* The PyModuleDef structure, in turn, must be passed to the interpreter in the module’s initialization function.
* The initialization function must be named PyInit_name(), where name is the name of the module and should match
* what we wrote in struct PyModuleDef.
* This should be the only non-static item defined in the module file
*/
PyMODINIT_FUNC
PyInit_mykmeanssp(void)
{
PyObject *m;
m = PyModule_Create(&moduledef);
if (!m) {
return NULL;
}
return m;
}
/*
this function handles error,
it checks the condition if it is not true it will print "An Error Has Occurred" and exit
*/
void error_occurred(int condition) {
if (condition == 1) {
printf("An Error Has Occurred\n");
exit(1);
}
}
/*
the function gets data points and calculates the closest cluster
by the smallest squared distance from a centroid
it returns the index of the cluster in the array
*/
int find_closets_cluster(double *data_point) {
double *difference, *current_mu, sum, min_sum;
int index = 0, i;
difference = calloc(k, sizeof(int *));
error_occurred(difference == NULL);
for (i = 0; i < k; i++) {
current_mu = centroids[i];
sum = calculate_distance_squared(current_mu, data_point);
difference[i] = sum;
}
min_sum = difference[0];
for (i = 0; i < k; i++) {
if (difference[i] < min_sum) {
min_sum = difference[i];
index = i;
}
}
free(difference);
return index;
}
/*
the function iterate through all the data points and assigns each to the closest cluster
it calculates the num elements in each cluster
*/
void set_clusters() {
int index, i, a;
for (i = 0; i < k; i++) {
num_elements_in_cluster[i] = 0;
}
for (i = 0; i < num_rows; i++) {
index = find_closets_cluster(data_points[i]);
for (a = 0; a < d; a++) {
clusters[index][a] += data_points[i][a];
}
num_elements_in_cluster[index]++;
}
}
/*
the function calculate the centroids by the new clusters that was already calculated
*/
void calculate_new_centroids() {
double *cluster;
int len_cluster, i, j;
for (i = 0; i < k; i++) {
cluster = clusters[i];
len_cluster = num_elements_in_cluster[i];
for (j = 0; j < d; j++) {
new_centroids[i][j] = cluster[j] / len_cluster;
}
}
}
/*
the function calculate the squared distance of two points
*/
double calculate_distance_squared(double *a, double *b) {
double distance = 0;
int i = d;
while (i--) {
distance += pow(a[i] - b[i], 2);
}
return distance;
}
/*
pre condition: centroids and data points arrays already initialized
every iteration it calculates the clusters and then updates the centroids
when the difference squared distance of all the centroids is smaller than epsilon we stop and return the centroids
*/
void algorithm() {
int i, j;
double *sum_diff_centroids;
double diff, sq_diff, max;
num_elements_in_cluster = calloc(k, sizeof(int));
clusters = allocate_array_2d(k, d);
new_centroids = allocate_array_2d(k, d);
sum_diff_centroids = calloc(k, sizeof(double));
while (max_iter > 0) {
for (j = 0; j < k; j++) {
sum_diff_centroids[j] = 0;
}
zero_array_2d(new_centroids, k, d);
zero_array_2d(clusters, k, d);
error_occurred(num_elements_in_cluster == NULL);
set_clusters();
calculate_new_centroids();
max_iter--;
error_occurred(sum_diff_centroids == NULL);
for (i = 0; i < k; i++) {
diff = calculate_distance_squared(centroids[i], new_centroids[i]);
sq_diff = pow(diff, 0.5);
sum_diff_centroids[i] = sq_diff;
}
max = sum_diff_centroids[0];
for (i = 0; i < k; i++) {
if (sum_diff_centroids[i] > max) {
max = sum_diff_centroids[i];
}
}
set_equal_array_2d(centroids, new_centroids, k, d);
if (max <= epsilon) {
max_iter = 0;
}
}
free(sum_diff_centroids);
free_array_2d(new_centroids, k);
free_array_2d(data_points, num_rows);
free_array_2d(clusters, k);
free(num_elements_in_cluster);
}
/*
the function allocate memory for 2 dimensional array of type double
*/
double **allocate_array_2d(int r, int c) {
double **arr;
arr = calloc(r, sizeof(double *));
error_occurred(arr == NULL);
while (r--) {
arr[r] = calloc(c, sizeof(double));
error_occurred(arr[r] == NULL);
}
return arr;
}
/*
the function free memory for 2 dimensional array
*/
void free_array_2d(double **arr, int r) {
while (r--) {
free(arr[r]);
}
free(arr);
}
/*
the function gets two 2 dimensional arrays and num of rows and columns
it sets the first array to be equal to the second array
*/
void set_equal_array_2d(double **new, double **current, int r, int c) {
int i, j;
for (i = 0; i < r; i++) {
for (j = 0; j < c; j++) {
new[i][j] = current[i][j];
}
}
}
/*
the function gets 2 dimensional array and num of rows and columns
it sets all the values of the array to zero
*/
void zero_array_2d(double **arr, int r, int c) {
int i, j;
for (i = 0; i < r; i++) {
for (j = 0; j < c; j++) {
arr[i][j] = 0;
}
}
}