effspm 0.2.6__cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (53) hide show
  1. effspm/__init__.py +11 -0
  2. effspm/_core.cpp +106 -0
  3. effspm/_effspm.cpp +609 -0
  4. effspm/_effspm.cpython-312-x86_64-linux-gnu.so +0 -0
  5. effspm/btminer/src/build_mdd.cpp +63 -0
  6. effspm/btminer/src/build_mdd.hpp +40 -0
  7. effspm/btminer/src/freq_miner.cpp +179 -0
  8. effspm/btminer/src/freq_miner.hpp +39 -0
  9. effspm/btminer/src/load_inst.cpp +200 -0
  10. effspm/btminer/src/load_inst.hpp +25 -0
  11. effspm/btminer/src/utility.cpp +65 -0
  12. effspm/btminer/src/utility.hpp +40 -0
  13. effspm/freq_miner.cpp +143 -0
  14. effspm/freq_miner.hpp +48 -0
  15. effspm/htminer/src/build_mdd.cpp +192 -0
  16. effspm/htminer/src/build_mdd.hpp +64 -0
  17. effspm/htminer/src/freq_miner.cpp +350 -0
  18. effspm/htminer/src/freq_miner.hpp +60 -0
  19. effspm/htminer/src/load_inst.cpp +394 -0
  20. effspm/htminer/src/load_inst.hpp +23 -0
  21. effspm/htminer/src/utility.cpp +72 -0
  22. effspm/htminer/src/utility.hpp +77 -0
  23. effspm/largebm/src/build_mdd.cpp +137 -0
  24. effspm/largebm/src/build_mdd.hpp +47 -0
  25. effspm/largebm/src/freq_miner.cpp +349 -0
  26. effspm/largebm/src/freq_miner.hpp +48 -0
  27. effspm/largebm/src/load_inst.cpp +230 -0
  28. effspm/largebm/src/load_inst.hpp +45 -0
  29. effspm/largebm/src/utility.cpp +45 -0
  30. effspm/largebm/src/utility.hpp +18 -0
  31. effspm/largehm/src/build_mdd.cpp +174 -0
  32. effspm/largehm/src/build_mdd.hpp +93 -0
  33. effspm/largehm/src/freq_miner.cpp +445 -0
  34. effspm/largehm/src/freq_miner.hpp +77 -0
  35. effspm/largehm/src/load_inst.cpp +357 -0
  36. effspm/largehm/src/load_inst.hpp +64 -0
  37. effspm/largehm/src/utility.cpp +38 -0
  38. effspm/largehm/src/utility.hpp +29 -0
  39. effspm/largepp/src/freq_miner.cpp +170 -0
  40. effspm/largepp/src/freq_miner.hpp +43 -0
  41. effspm/largepp/src/load_inst.cpp +219 -0
  42. effspm/largepp/src/load_inst.hpp +28 -0
  43. effspm/largepp/src/utility.cpp +34 -0
  44. effspm/largepp/src/utility.hpp +21 -0
  45. effspm/load_inst.cpp +252 -0
  46. effspm/load_inst.hpp +31 -0
  47. effspm/utility.cpp +55 -0
  48. effspm/utility.hpp +29 -0
  49. effspm-0.2.6.dist-info/METADATA +237 -0
  50. effspm-0.2.6.dist-info/RECORD +53 -0
  51. effspm-0.2.6.dist-info/WHEEL +6 -0
  52. effspm-0.2.6.dist-info/licenses/LICENSE +201 -0
  53. effspm-0.2.6.dist-info/top_level.txt +1 -0
@@ -0,0 +1,350 @@
1
+ #include <iostream>
2
+ #include <time.h>
3
+ #include "freq_miner.hpp"
4
+ #include "build_mdd.hpp"
5
+ #include "utility.hpp"
6
+
7
+ namespace htminer {
8
+ void Out_patt(vector<int>& seq, unsigned int freq);
9
+ void Extend_patt(Pattern& _patt);
10
+ void Mine_vec(unsigned int seq_ID, int pos, int num_found, vector<unsigned int>& ancest, vector<int>& items, unsigned int inod, int sgn);
11
+
12
+ unsigned long long int num_patt = 0;
13
+
14
+ vector<bool> ilist;
15
+ vector<bool> slist;
16
+
17
+ vector<Pattern> pot_patt;
18
+ vector<VPattern> pot_vpatt;
19
+ vector<unsigned int> last_strpnt;
20
+ vector<unsigned int> ancest_base;
21
+ vector<int> DFS_numfound;
22
+
23
+ Pattern _patt;
24
+ VPattern _vpatt;
25
+
26
+ int itmset_size;
27
+ int last_neg;
28
+
29
+ bool ilist_nempty;
30
+
31
+ void Freq_miner() {
32
+ collectedPatterns.clear();
33
+ vector<int> list;
34
+
35
+ for (int i = 0; i < L; ++i) {
36
+ if (DFS[i].freq >= theta) {
37
+ list.push_back(-i-1);
38
+ if (itmset_exists)
39
+ list.push_back(i+1);
40
+ }
41
+ }
42
+
43
+ for (int i = 0; i < DFS.size(); ++i)
44
+ DFS[i].list = list;
45
+
46
+ while (!DFS.empty() && give_time(clock() - start_time) < time_limit) {
47
+ if (DFS.back().freq >= theta)
48
+ Extend_patt(DFS.back());
49
+ else {
50
+ DFS.pop_back();
51
+ if (!VDFS.empty() && VDFS.back().ass_patt == DFS.size())
52
+ VDFS.pop_back();
53
+ }
54
+ }
55
+ }
56
+
57
+ void Extend_patt(Pattern& _pattern) {
58
+
59
+ swap(_patt, _pattern);
60
+ DFS.pop_back();
61
+
62
+ slist = vector<bool>(L, 0);
63
+ ilist_nempty = 0;
64
+
65
+ if (itmset_exists) {
66
+ ilist = vector<bool>(L, 0);
67
+ for (vector<int>::iterator it = _patt.list.begin(); it != _patt.list.end(); ++it) {
68
+ if (*it < 0)
69
+ slist[-(*it) - 1] = 1;
70
+ else {
71
+ ilist[(*it) - 1] = 1;
72
+ ilist_nempty = 1;
73
+ }
74
+ }
75
+ }
76
+ else {
77
+ for (vector<int>::iterator it = _patt.list.begin(); it != _patt.list.end(); ++it)
78
+ slist[-(*it) - 1] = 1;
79
+ }
80
+
81
+ last_neg = _patt.seq.size() - 1;
82
+ while (_patt.seq[last_neg] > 0)
83
+ --last_neg;
84
+ itmset_size = _patt.seq.size() - last_neg;
85
+
86
+ pot_patt = vector<Pattern>(L + L * ilist_nempty);
87
+ if (!CTree.empty())
88
+ pot_vpatt = vector<VPattern>(L + L * ilist_nempty);
89
+
90
+ last_strpnt = vector<unsigned int>(L, 0);
91
+
92
+ if (!VDFS.empty() && VDFS.back().ass_patt == DFS.size()) {
93
+ swap(_vpatt, VDFS.back());
94
+ VDFS.pop_back();
95
+ for (unsigned int pnt = 0; pnt < _vpatt.str_pnt.size(); ++pnt) {
96
+ if (_vpatt.str_pnt[pnt] < 0)
97
+ Mine_vec(_vpatt.seq_ID[pnt], -_vpatt.str_pnt[pnt], -1, ancest_base, CTree[_vpatt.seq_ID[pnt]].seq, 0, -1); //starting search from vpatt should start from 1 position ahead of pointer
98
+ else //-1:no need to check ancest for remaining itemset items
99
+ Mine_vec(_vpatt.seq_ID[pnt], _vpatt.str_pnt[pnt], -1, ancest_base, VTree[_vpatt.seq_ID[pnt]].seq, 0, 1);
100
+ }
101
+ }
102
+
103
+ vector<unsigned int> DFS_itm; //for initial itemset extention
104
+ vector<unsigned int> DFS_seq; //for initial itemset extention
105
+ if (ilist_nempty)
106
+ DFS_numfound.clear(); //tracks whether the current itemset is found anywhere along the search path, by counting how many of the current items in the itemset are found
107
+ for (unsigned int pnt = 0; pnt < _patt.str_pnt.size(); ++pnt) {
108
+ DFS_itm.push_back(_patt.str_pnt[pnt]);
109
+ while(!DFS_itm.empty()) {
110
+ unsigned int cur_sibl = DFS_itm.back();
111
+ DFS_itm.pop_back();
112
+ if (Tree[cur_sibl].itmset < 0) {
113
+ unsigned int carc = Tree[cur_sibl].chld;
114
+ Mine_vec(carc, 0, -1, CTree[carc].ancest, CTree[carc].seq, _patt.str_pnt[pnt], -1);
115
+ cur_sibl = CTree[carc].ancest.back();
116
+ while (cur_sibl != 0) {
117
+ Mine_vec(cur_sibl - 1, 0, -1, CTree[carc].ancest, VTree[cur_sibl - 1].seq, _patt.str_pnt[pnt], 1);
118
+ cur_sibl = VTree[cur_sibl - 1].sibl;
119
+ }
120
+ continue;
121
+ }
122
+ cur_sibl = Tree[cur_sibl].chld;
123
+ while (cur_sibl != 0) {
124
+ int cur_itm = Tree[cur_sibl].item;
125
+ if (cur_itm < 0) {
126
+ cur_itm = -cur_itm;
127
+ if (slist[cur_itm - 1]) {
128
+ pot_patt[cur_itm - 1].freq += Tree[cur_sibl].freq;
129
+ if (Tree[cur_sibl].chld != 0 || Tree[cur_sibl].itmset < 0)
130
+ pot_patt[cur_itm - 1].str_pnt.push_back(cur_sibl);
131
+ }
132
+ if (Tree[cur_sibl].chld != 0 || Tree[cur_sibl].itmset < 0) {
133
+ DFS_seq.push_back(cur_sibl);
134
+ if (ilist_nempty) {
135
+ if (cur_itm == -_patt.seq[last_neg])
136
+ DFS_numfound.push_back(1);
137
+ else
138
+ DFS_numfound.push_back(0);
139
+ }
140
+ }
141
+ }
142
+ else {
143
+ if (ilist[cur_itm - 1]) {
144
+ pot_patt[cur_itm + L - 1].freq += Tree[cur_sibl].freq;
145
+ if (Tree[cur_sibl].chld != 0 || Tree[cur_sibl].itmset < 0)
146
+ pot_patt[cur_itm + L - 1].str_pnt.push_back(cur_sibl);
147
+ }
148
+ if (Tree[cur_sibl].chld != 0 || Tree[cur_sibl].itmset < 0)
149
+ DFS_itm.push_back(cur_sibl);
150
+ }
151
+ cur_sibl = Tree[cur_sibl].sibl;
152
+ }
153
+ }
154
+ if (ilist_nempty) {
155
+ for (int i = 0; i < L; ++i) {
156
+ if (ilist[i])
157
+ last_strpnt[i] = pot_patt[i + L].str_pnt.size();
158
+ }
159
+ }
160
+ while(!DFS_seq.empty()) {
161
+ unsigned int cur_sibl = DFS_seq.back();
162
+ DFS_seq.pop_back();
163
+ int num_found = 0;
164
+ if (ilist_nempty) {
165
+ num_found = DFS_numfound.back();
166
+ DFS_numfound.pop_back();
167
+ }
168
+ if (Tree[cur_sibl].itmset < 0) {
169
+ unsigned int carc = Tree[cur_sibl].chld;
170
+ Mine_vec(carc, 0, num_found, CTree[carc].ancest, CTree[carc].seq, _patt.str_pnt[pnt], -1);
171
+ cur_sibl = CTree[carc].ancest.back();
172
+ while (cur_sibl != 0) {
173
+ Mine_vec(cur_sibl - 1, 0, num_found, CTree[carc].ancest, VTree[cur_sibl - 1].seq, _patt.str_pnt[pnt], 1);
174
+ cur_sibl = VTree[cur_sibl - 1].sibl;
175
+ }
176
+ continue;
177
+ }
178
+ cur_sibl = Tree[cur_sibl].chld;
179
+ while (cur_sibl != 0) {
180
+ int cur_itm = Tree[cur_sibl].item;
181
+ if (cur_itm > 0) {
182
+ if (num_found == itmset_size && ilist[cur_itm - 1] && (abs(Tree[Tree[cur_sibl].anct].itmset) < abs(Tree[_patt.str_pnt[pnt]].itmset) || !check_parent(Tree[cur_sibl].anct, _patt.str_pnt[pnt], last_strpnt[cur_itm - 1], pot_patt[cur_itm + L - 1].str_pnt))) {
183
+ pot_patt[cur_itm + L - 1].freq += Tree[cur_sibl].freq;
184
+ if (Tree[cur_sibl].chld != 0 || Tree[cur_sibl].itmset < 0)
185
+ pot_patt[cur_itm + L - 1].str_pnt.push_back(cur_sibl);
186
+ }
187
+ if (slist[cur_itm - 1] && abs(Tree[Tree[cur_sibl].anct].itmset) <= abs(Tree[_patt.str_pnt[pnt]].itmset)) {
188
+ pot_patt[cur_itm - 1].freq += Tree[cur_sibl].freq;
189
+ if (Tree[cur_sibl].chld != 0 || Tree[cur_sibl].itmset < 0)
190
+ pot_patt[cur_itm - 1].str_pnt.push_back(cur_sibl);
191
+ }
192
+ if (Tree[cur_sibl].chld != 0 || Tree[cur_sibl].itmset < 0) {
193
+ DFS_seq.push_back(cur_sibl);
194
+ if (ilist_nempty) {
195
+ if (num_found < itmset_size && cur_itm == abs(_patt.seq[last_neg + num_found]))
196
+ DFS_numfound.push_back(num_found + 1);
197
+ else
198
+ DFS_numfound.push_back(num_found);
199
+ }
200
+ }
201
+ }
202
+ else {
203
+ cur_itm = -cur_itm;
204
+ if (slist[cur_itm - 1] && abs(Tree[Tree[cur_sibl].anct].itmset) <= abs(Tree[_patt.str_pnt[pnt]].itmset)) {
205
+ pot_patt[cur_itm - 1].freq += Tree[cur_sibl].freq;
206
+ if (Tree[cur_sibl].chld != 0 || Tree[cur_sibl].itmset < 0)
207
+ pot_patt[cur_itm - 1].str_pnt.push_back(cur_sibl);
208
+ }
209
+ if (Tree[cur_sibl].chld != 0 || Tree[cur_sibl].itmset < 0) {
210
+ DFS_seq.push_back(cur_sibl);
211
+ if (ilist_nempty) {
212
+ if (cur_itm == -_patt.seq[last_neg])
213
+ DFS_numfound.push_back(1);
214
+ else
215
+ DFS_numfound.push_back(0);
216
+ }
217
+ }
218
+ }
219
+ cur_sibl = Tree[cur_sibl].sibl;
220
+ }
221
+ }
222
+ }
223
+
224
+ vector<int> ilistp;
225
+ vector<int> slistp;
226
+ for (vector<int>::iterator it = _patt.list.begin(); it != _patt.list.end(); ++it) {
227
+ if (*it > 0 && pot_patt[(*it) + L - 1].freq >= theta)
228
+ ilistp.push_back(*it);
229
+ else if (*it < 0 && pot_patt[-(*it) - 1].freq >= theta) {
230
+ if (itmset_exists)
231
+ slistp.push_back(-(*it));
232
+ ilistp.push_back(*it);
233
+ slistp.push_back(*it);
234
+ }
235
+ }
236
+
237
+ for (vector<int>::iterator it = ilistp.begin(); it != ilistp.end(); ++it) {
238
+ int p;
239
+ if (*it < 0)
240
+ p = -(*it) - 1;
241
+ else
242
+ p = (*it) - 1 + L;
243
+ DFS.emplace_back();
244
+ swap(DFS.back(), pot_patt[p]);
245
+ DFS.back().seq = _patt.seq;
246
+ DFS.back().seq.push_back(*it);
247
+ if (*it < 0)
248
+ DFS.back().list = slistp;
249
+ else
250
+ DFS.back().list = ilistp;
251
+ if (!CTree.empty() && !pot_vpatt[p].str_pnt.empty()) {
252
+ pot_vpatt[p].ass_patt = DFS.size() - 1;
253
+ VDFS.emplace_back();
254
+ swap(VDFS.back(), pot_vpatt[p]);
255
+ }
256
+ if (b_disp || b_write)
257
+ Out_patt(DFS.back().seq, DFS.back().freq);
258
+ htminer::collectedPatterns.emplace_back(DFS.back().seq);
259
+ ++num_patt;
260
+
261
+ }
262
+ }
263
+
264
+
265
+ void Mine_vec(unsigned int seq_ID, int pos, int num_found, vector<unsigned int>& ancest, vector<int>& items, unsigned int pnt, int sgn) {
266
+
267
+ vector<bool> found(L + L * ilist_nempty, 0);
268
+ int num_ext = 0;
269
+
270
+ if (num_found == -1) {
271
+ while (pos < items.size() && items[pos] > 0 && num_ext < _patt.list.size()) {
272
+ int cur_itm = items[pos];
273
+ if (ilist[cur_itm - 1] && !found[cur_itm + L - 1]) {
274
+ if (pos + 1 < items.size()) {
275
+ pot_vpatt[cur_itm + L - 1].seq_ID.push_back(seq_ID);
276
+ pot_vpatt[cur_itm + L - 1].str_pnt.push_back(sgn * (pos + 1));
277
+ }
278
+ ++pot_patt[cur_itm + L - 1].freq;
279
+ found[cur_itm + L - 1] = 1;
280
+ ++num_ext;
281
+ }
282
+ ++pos;
283
+ }
284
+ }
285
+
286
+ for (unsigned int k = pos; k < items.size() && num_ext < _patt.list.size(); ++k) {
287
+ int cur_itm = abs(items[k]);
288
+ if (items[k] < 0)
289
+ num_found = 0;
290
+ if (slist[cur_itm - 1] && !found[cur_itm - 1]) {
291
+ if (ancest.empty() || abs(Tree[ancest[cur_itm - 1]].itmset) <= abs(Tree[pnt].itmset)) {
292
+ if (k + 1 < items.size()) {
293
+ pot_vpatt[cur_itm - 1].seq_ID.push_back(seq_ID);
294
+ pot_vpatt[cur_itm - 1].str_pnt.push_back(sgn * (k + 1));
295
+ }
296
+ ++pot_patt[cur_itm - 1].freq;
297
+ }
298
+ found[cur_itm - 1] = 1;
299
+ ++num_ext;
300
+ }
301
+ if (num_found == itmset_size) {
302
+ if (ilist[cur_itm - 1] && !found[cur_itm + L - 1]) {
303
+ if (ancest.empty() || abs(Tree[ancest[cur_itm - 1]].itmset) < abs(Tree[pnt].itmset) || !check_parent(ancest[cur_itm - 1], pnt, last_strpnt[cur_itm - 1], pot_patt[cur_itm + L - 1].str_pnt)) {
304
+ if (k + 1 < items.size()) {
305
+ pot_vpatt[cur_itm + L - 1].seq_ID.push_back(seq_ID);
306
+ pot_vpatt[cur_itm + L - 1].str_pnt.push_back(sgn * (k + 1));
307
+ }
308
+ ++pot_patt[cur_itm + L - 1].freq;
309
+ }
310
+ found[cur_itm + L - 1] = 1;
311
+ ++num_ext;
312
+ }
313
+ }
314
+ else if (cur_itm == abs(_patt.seq[last_neg + num_found]))
315
+ ++num_found;
316
+ }
317
+ }
318
+
319
+
320
+ void Out_patt(vector<int>& seq, unsigned int freq) {
321
+
322
+ ofstream file_o;
323
+ if (b_write)
324
+ file_o.open(out_file, std::ios::app);
325
+
326
+ for (int ii = 0; ii < seq.size(); ii++) {
327
+ if (b_disp)
328
+ cout << seq[ii] << " ";
329
+ if (b_write)
330
+ file_o << seq[ii] << " ";
331
+ }
332
+ if (b_disp)
333
+ cout << endl;
334
+ if (b_write)
335
+ file_o << endl;
336
+
337
+ if (b_disp)
338
+ cout << "************** Freq: " << freq << endl;
339
+ if (b_write) {
340
+ file_o << "************** Freq: " << freq << endl;
341
+ file_o.close();
342
+ }
343
+
344
+ }
345
+
346
+
347
+ }
348
+
349
+
350
+
@@ -0,0 +1,60 @@
1
+ #pragma once
2
+
3
+ #include "load_inst.hpp"
4
+ #include "build_mdd.hpp"
5
+
6
+ namespace htminer {
7
+ void Freq_miner();
8
+
9
+ class Pattern {
10
+ public:
11
+
12
+ vector<int> seq;
13
+ vector<unsigned int> str_pnt;
14
+ vector<int> list;
15
+
16
+ unsigned long long int freq;
17
+
18
+ Pattern(int item) {
19
+ seq.push_back(item);
20
+ freq = 0;
21
+ }
22
+
23
+ Pattern(size_t _pnt, bool _res) {
24
+ str_pnt.reserve(_pnt);
25
+ freq = 0;
26
+ }
27
+
28
+ Pattern() {
29
+ freq = 0;
30
+ }
31
+
32
+
33
+ };
34
+
35
+ class VPattern {
36
+ public:
37
+
38
+ unsigned long long int ass_patt;
39
+
40
+ vector<int> str_pnt;
41
+ vector<unsigned int> seq_ID;
42
+
43
+ VPattern(unsigned long long int _patt) {
44
+ ass_patt = _patt;
45
+ }
46
+
47
+ VPattern(size_t _pnt, bool a) {
48
+ str_pnt.reserve(_pnt);
49
+ }
50
+
51
+ VPattern() {}
52
+ };
53
+
54
+
55
+ extern unsigned long long int num_patt;
56
+ extern vector<Pattern> DFS;
57
+ extern vector<VPattern> VDFS;
58
+
59
+ }
60
+