-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathData_Preparation_For_Learning.py
1805 lines (1623 loc) · 85.6 KB
/
Data_Preparation_For_Learning.py
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
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
import random
import shutil
import os
from alogrithms import mergeSort
from New_Query_Sampling import DivideTrainingSetIntoQueries,DivideTestingSetIntoQueries, Clustering_Products
def get_num_reviews_per_product_local(product_path):
#Count the number of reviews for a product
count = 0
with open(product_path, 'r') as filep:
for item in filep:
count+=1
return count
def Write_Number_Reviews_For_All_Categories():
category_source = "d:\Yassien_PhD\categories/"
product_base_directory = "d:\Yassien_PhD\Product_Reviews/"
destenation_directory = "d:/Yassien_PhD/Number_of_reviews_per_product/"
Categories = ["Industrial & Scientific", "Jewelry", "Arts, Crafts & Sewing", "Toys & Games", "Video Games","Computers & Accessories", "Software", "Cell Phones & Accessories", "Electronics"]
for category_name in Categories:
print("Processing "+category_name)
source_category_path = category_source + category_name + ".txt"
dest_cat_path = destenation_directory+ category_name + ".txt"
filehandle = open(dest_cat_path,'w')
with open(source_category_path, 'r') as filep:
for item in filep:
line = item.split('\t')
productid = line[0]
product_file_path = product_base_directory + productid + ".txt"
count = get_num_reviews_per_product_local(product_file_path)
filehandle.write(str(productid)+"\t"+str(count)+"\n")
filehandle.close()
return
def Query_Sampling_For_New_Experiment_Setup(training_products,products_num_revs_path,sampling_choice,query_size):
products_num_revs_dict = dict()
with open(products_num_revs_path, 'r') as filep:
for item in filep:
line = item.split('\t')
products_num_revs_dict[line[0]]=int(line[1])
final_training_set = []
index = 0
temp_list = []
for product_line in training_products:
produtid = str(product_line).split('\t')[0]
if sampling_choice == 1 or sampling_choice == 2:
count = products_num_revs_dict[produtid]
if sampling_choice == 2 and count >50:
pair = (index,count)
temp_list.append(pair)
elif sampling_choice == 1:
pair = (index, count)
temp_list.append(pair)
index+=1
#print("temp list before")
#print(temp_list)
mergeSort(temp_list)
temp_list.reverse()
#print("temp list after")
#print(temp_list)
total_num_prods = len(temp_list)
new_list = []
index_top = 0
num_taken_top = 0
index_bottom = total_num_prods-1
num_to_take_top = int(query_size / 2)
num_to_take_bottom = query_size - num_to_take_top
print("num_to_take_top "+str(num_to_take_top))
print("num_to_take_bottom " + str(num_to_take_bottom))
num_taken_bottom = num_to_take_bottom
# print("total_num_prods "+str(total_num_prods))
if sampling_choice == 2:
while len(new_list)<total_num_prods:
#print("index_top "+str(index_top))
#print("index_bottom "+str(index_bottom))
if num_taken_top <num_to_take_top:
if index_top==index_bottom:
new_list.append(temp_list[index_bottom])
break
new_list.append(temp_list[index_top])
num_taken_top+=1
index_top+=1
elif num_taken_bottom == num_to_take_bottom and num_taken_top == num_to_take_top:
num_taken_top = 0
if num_taken_bottom < num_to_take_bottom:
if index_top == index_bottom:
new_list.append(temp_list[index_bottom])
break
new_list.append(temp_list[index_bottom])
num_taken_bottom += 1
index_bottom -= 1
elif num_taken_bottom == num_to_take_bottom and num_taken_top == num_to_take_top:
num_taken_bottom = 0
temp_list = new_list
for i in range(len(temp_list)):
pro_index = temp_list[i][0]
final_training_set.append(training_products[pro_index])
return final_training_set
def Randomize_Product_List_and_Picktraining(source_category_path,category_name, training_ratio,local_destination,product_base_directory,drive,query_size,source_feature_vector_path):
print("Processing "+category_name)
index = 0
product_list = []
productid_index_dict = dict()
#Copying the original categories file for reference
shutil.copy2(source_category_path,local_destination)
with open(source_category_path, 'r') as filep:
for item in filep:
line = item.split('\t')
productid = line[0]
product_line = item#line[0]
productid_index_dict[productid]=index
product_list.append(product_line)
index += 1
total_num_products = len(product_list)
print("Total Num Products "+str(total_num_products))
num_training = int(total_num_products*training_ratio)
num_testing = total_num_products-num_training
print("num_training " + str(num_training))
print("num_testing " + str(num_testing))
training_products = []
testing_products = []
indo = 0
while len(product_list)>num_testing:
# Code randomization
choice = random.choice(list(product_list))
#choice = product_list[indo]
training_products.append(choice)
#print("Choice "+str(choice))
product_list.remove(choice)
#indo+=1
#print(product_list)
testing_products = product_list
old_total = len(training_products)+len(testing_products)
#Here we inject whatever query sampling we need
sampling_choice =1 #means arrange by number of reviews
products_num_revs_path = drive+"Yassien_PhD/Number_of_reviews_per_product/"+category_name+".txt"
#training_products=Query_Sampling_For_New_Experiment_Setup(training_products,products_num_revs_path,sampling_choice,query_size)
training_products = Clustering_Products(training_products,source_feature_vector_path,source_category_path,products_num_revs_path)
######################################################################################################################
print("Final num_training now " + str(len(training_products)))
print("Final num_testing was " + str(len(testing_products)))
final_total = len(training_products) + len(testing_products)
print("Final total "+str(final_total))
print("Old Total "+str(old_total))
print("Final after num_training " + str(len(training_products)))
if old_total !=final_total:
new_list = []
new_testing_count = int(len(training_products)*0.1)
print("new_testing_count " + str(new_testing_count))
for i in range(new_testing_count):
new_list.append(testing_products[i])
testing_products = new_list
print("AFter Adjustment***************************************************************")
print("Final after num_testing " + str(len(testing_products)))
final_total = len(training_products) + len(testing_products)
print("Final after total " + str(final_total))
print("Writing Files")
training_filepath = local_destination + "training.txt"
training_index_filepath = local_destination + "training_index.txt"
filehandle = open(training_filepath, 'w')
filehandle_index = open(training_index_filepath, 'w')
num_actually_written = 0
for product_line in training_products:
produtid= str(product_line).split('\t')[0]
product_file_path = product_base_directory+produtid+".txt"
#count = get_num_reviews_per_product_local(product_file_path)
#if count>50:
filehandle.write(product_line)
filehandle_index.write(str(productid_index_dict[produtid])+"\n")
num_actually_written+=1
filehandle.close()
filehandle_index.close()
print("Final num_actually_written " + str((num_actually_written)))
testing_filepath = local_destination + "testing.txt"
testing_index_filepath = local_destination + "testing_index.txt"
filehandle = open(testing_filepath, 'w')
filehandle_index = open(testing_index_filepath, 'w')
for product_line in testing_products:
produtid = str(product_line).split('\t')[0]
filehandle.write(product_line)
filehandle_index.write(str(productid_index_dict[produtid]) + "\n")
filehandle.close()
filehandle_index.close()
print("------------------------------------------------")
return
def Retreive_Train_Test_Per_Category(source_feature_vector_path,category_name,modified_categories,cat_desination_directory):
print("Processing "+category_name)
try:
os.stat(cat_desination_directory)
except:
os.mkdir(cat_desination_directory)
trainin_index_filepath = modified_categories + "training_index.txt"
testing_index_filepath = modified_categories + "testing_index.txt"
training_indices = []
testing_indices = []
with open(trainin_index_filepath, 'r') as filep:
for item in filep:
training_indices.append(int(item))
with open(testing_index_filepath, 'r') as filep:
for item in filep:
testing_indices.append(int(item))
feature_vect_dict = dict()
index = 0
with open(source_feature_vector_path, 'r') as filep:
for line in filep:
feature_vect_dict[index]=line
index+=1
num_products = len(feature_vect_dict)
training_feat_vec_filepath = cat_desination_directory + "training.txt"
filehandle = open(training_feat_vec_filepath, 'w')
for i in range(len(training_indices)):
feat_vec = feature_vect_dict[training_indices[i]]
filehandle.write(feat_vec)
filehandle.close()
testing_feat_vec_filepath = cat_desination_directory + "testing.txt"
filehandle = open(testing_feat_vec_filepath, 'w')
for i in range(len(testing_indices)):
feat_vec = feature_vect_dict[testing_indices[i]]
filehandle.write(feat_vec)
filehandle.close()
return
def PrepareCategoriesWithSalesRankRanking(sourceCategorypath,destinationCategorypath,category_name):
print("Processing " + category_name)
ranks = []
with open(sourceCategorypath, 'r') as filep:
for line in filep:
ranks.append(int(line.split(' ')[0]))
products = []
with open(destinationCategorypath, 'r') as filep:
for line in filep:
products.append(line.split('\t')[0])
if len(ranks)!=len(products):
print("Non Equal lengths ranks is "+str(len(ranks))+" products "+str(len(products)))
print(products)
filehandle = open(destinationCategorypath, 'w')
for i in range(len(products)):
filehandle.write(str(products[i])+"\t"+str(ranks[i])+"\n")
filehandle.close()
return
def Prepare_Training_Testing_Data_New_Experiment_Setup(query_size,drive):
#drive = "d:/"
category_source = drive+"Yassien_PhD\categories/"
categories_source=drive+"Yassien_PhD\Experiment_5\Categories/"
source_features_path = drive+"Yassien_PhD\Experiment_4\All_Categories_Data_25_Basic_Features_With_5_Time_Intervals/"
train_test_destination_stage1=drive+"Yassien_PhD\Experiment_5\Train_Test_Category_Stage_1/"
train_test_destination=drive+"Yassien_PhD\Experiment_5\Train_Test_Category_With_10_Time_Interval_TQ_Target/"
product_base_directory =drive+"Yassien_PhD\Product_Reviews/"
#Categories with large number of products > 5000 products we don't need cross_Validation randomize and take 80%
categories_with_large_products=["Industrial & Scientific", "Jewelry", "Arts, Crafts & Sewing", "Toys & Games", "Video Games","Computers & Accessories", "Software", "Cell Phones & Accessories", "Electronics"]#[ "Jewelry","Toys & Games", "Video Games" , "Cell Phones & Accessories", "Electronics"]
for category_name in categories_with_large_products:
################################################################################################################################################################
#'''
#This part of the code to randomize all products within one group and then pick 80% randomly for training and 20% for testing keeping the indices of each set to be able to formulate the queries
modified_categories_with_indices= categories_source+category_name+"/"
try:
os.stat(modified_categories_with_indices)
except:
os.mkdir(modified_categories_with_indices)
training_ratio = 0.9
source_category_path = category_source + category_name + ".txt"
source_feature_vector_path = source_features_path + category_name + ".txt"
cat_train_test_desination_directory_stage_1 = train_test_destination_stage1 + category_name + "/"
train_test_destination_for_cat = train_test_destination + category_name + "/"
Randomize_Product_List_and_Picktraining(source_category_path,category_name, training_ratio,modified_categories_with_indices,product_base_directory,drive,query_size,source_feature_vector_path)
Retreive_Train_Test_Per_Category(source_feature_vector_path,category_name,modified_categories_with_indices,cat_train_test_desination_directory_stage_1)
#'''
################################################################################################################################################################
################################################################################################################################################################
#This part was just to prepare to get the sales rank and the TQ rank for all products in all categories to be utilized in forming the training and testing sets
'''sourceCategory="C:\Yassien_RMIT PhD\Datasets\TruthDiscovery_Datasets\Web data Amazon reviews/Unique_Products_Stanford_three\Experiment 2\All_Categories_Data_25_Basic_Features_With_10_Time_Interval_TQ_Target_For_Ranking/"+category_name+".txt"
destinationCategory="F:\Yassien_PhD\Experiment_5\Categories_Ranked_by_TQ_Rank/"+category_name+".txt"
PrepareCategoriesWithSalesRankRanking(sourceCategory,destinationCategory,category_name)'''
################################################################################################################################################################
#This part converts the randomized training and testing sets into queries with the given size
cat_train_test_desination_directory_stage_1 = train_test_destination_stage1 + category_name + "/"
train_test_destination_for_cat = train_test_destination + category_name + "/"
try:
os.stat(train_test_destination_for_cat)
except:
os.mkdir(train_test_destination_for_cat)
train_test_destination_for_cat+= "/Cutoff_10/"
modified_categories_with_indices = categories_source + category_name + "/"
validation_ratio = 0.2
new_q_index = DivideTrainingSetIntoQueries(cat_train_test_desination_directory_stage_1,category_name,train_test_destination_for_cat,query_size,validation_ratio)
sales_rank_original_ranking_path = drive+"Yassien_PhD\Experiment_5\Categories_Ranked_by_Sales_Rank/"+category_name+".txt"
modified_categories_with_indices = categories_source + category_name + "/"
DivideTestingSetIntoQueries(cat_train_test_desination_directory_stage_1,category_name,train_test_destination_for_cat,modified_categories_with_indices,sales_rank_original_ranking_path,query_size,new_q_index)
return
def Randomize_Queries_List_training_Testing_FromSubCats(queries_directory,randomized_queries_dest,training_ratio):
print("This proedure randomizes queries for training and testing")
queries_list = []
for file_name in os.listdir(queries_directory):
queries_list.append(int(file_name.split('.')[0]))
print("Total # Queries "+ str(len(queries_list)))
training_queries = []
testing_queries = []
num_training = int(len(queries_list)*training_ratio)
num_testing = len(queries_list)-num_training
while len(queries_list) > num_testing:
# Code randomization
choice = random.choice(list(queries_list))
training_queries.append(choice)
queries_list.remove(choice)
testing_queries = queries_list
#Sorting testing and training query ids
training_queries.sort()
testing_queries.sort()
print("Training # Queries " + str(len(training_queries)))
print("Training Queries ")
print(training_queries)
training_file_path = randomized_queries_dest+"training.txt"
filehandle = open(training_file_path,'w')
for query in training_queries:
filehandle.write(str(query)+"\n")
filehandle.close
print("Testing # Queries " + str(len(testing_queries)))
print("testing_queries Queries ")
print(testing_queries)
testing_file_path = randomized_queries_dest + "testing.txt"
filehandle = open(testing_file_path, 'w')
for query in testing_queries:
filehandle.write(str(query) + "\n")
filehandle.close
return training_queries,testing_queries
def Relabel_Products_Within_query(query,new_query_index):
num_products_assigned_to_q = len(query)
index = 0
query_pair = []
for product_line in query:
rank = str(product_line).split(' ')[0]
query_pair.append((index, int(rank)))
index += 1
#print("query_pair before ")
#print(query_pair)
given_query_pair = []
for p in query_pair:
given_query_pair.append(p)
mergeSort(query_pair)
query_pair.reverse()
#print("query pair after")
#print(query_pair)
#print("given_query_pair")
#print(given_query_pair)
rank = len(query_pair) - 1
correct_rank = 0
relabled_queries = []
for qpair in given_query_pair:
new_query = ""
correct_rank = 0
for p in query_pair:
if p[0] == qpair[0]:
break
else:
correct_rank += 1
correct_rank = len(query_pair) - 1 - correct_rank
# print("qpair[0] "+str(qpair[0]))
#print("correct_rank is "+str(correct_rank))
old_query = query[qpair[0]]
new_query = str(correct_rank) + ' '
old_query_iter = str(old_query).split(' ')
for i in range(1, len(old_query_iter)):
if i == 1:
temp = old_query_iter[i].split(':')
new_q_index = temp[0] + ":" + str(new_query_index)
new_query += new_q_index + ' '
else:
if i == len(old_query_iter) - 1:
new_query += old_query_iter[i]
else:
new_query += old_query_iter[i] + ' '
relabled_queries.append(new_query)
rank -= 1
return relabled_queries
def Pick_Feature_Data_For_Given_Queries(randomized_queries_path,queries_directory,all_product_indices,query_map_code_file_path,training_testing_dest_directory,features_base_directory,train,categories_with_large_products):
query_ids = []
with open(randomized_queries_path, 'r') as fp:
for line in fp:
query_ids.append(line.split('\n')[0])
print(query_ids)
query_id_map_list = dict()
ignore_first_line = 0
with open(query_map_code_file_path, 'r') as fp:
for line in fp:
if ignore_first_line == 0:
ignore_first_line = 1
continue
row = line.split('\t')
query_id_map_list[row[0]]=(row[1],row[2].split('\n')[0]) #add query id mapped to main Category and then sub_category
#print(query_id_map_list)
all_product_cat_indices = dict()
all_category_feature_vect_dict = dict()
for category_name in categories_with_large_products:
indicies_file_path = all_product_indices + category_name + ".txt"
all_product_indices_dict = dict()
ignore_first_line = 0
print(indicies_file_path)
with open(indicies_file_path, 'r') as fp:
for line in fp:
if ignore_first_line == 0:
ignore_first_line = 1
continue
row = line.split('\t')
all_product_indices_dict[row[0]]=int(row[1])
all_product_cat_indices[category_name]=all_product_indices_dict
query_feat_file_path = features_base_directory + category_name + ".txt"
all_vectors = []
with open(query_feat_file_path, 'r') as fp:
for feature_row in fp:
all_vectors.append(feature_row.split('\n')[0])
all_category_feature_vect_dict[category_name]=all_vectors
#print(len(all_product_indices_dict))
#print(all_product_indices_dict)
train_file_path =""
test_file_path=""
if train == 1:
train_file_path=training_testing_dest_directory+"train.txt"
filehandle = open(train_file_path,'w')
else:
test_file_path = training_testing_dest_directory+"test.txt"
filehandle = open(test_file_path, 'w')
for query_row in query_ids:
queryid = query_row
print(queryid)
query_ind = queryid#queryid.split('_')[1]
query_cat_sub_cat = query_id_map_list[queryid]
query_category = query_cat_sub_cat[0]
query_sub_category = query_cat_sub_cat[1]
query_file_path = queries_directory+queryid+".txt"
query_data = []
with open(query_file_path, 'r') as fp:
for line in fp:
product_id = line.split('\n')[0]
all_product_indices_dict=all_product_cat_indices[query_category]
product_index = int(all_product_indices_dict[product_id])
#print(product_id+" "+str(product_index))
'''query_feat_file_path = features_base_directory+query_category+".txt"
all_vectors = []
with open(query_feat_file_path, 'r') as fp:
for feature_row in fp:
all_vectors.append(feature_row.split('\n')[0])
#print(all_vectors[product_index])
'''
#print(query_category)
#print(product_index)
#print(len(all_category_feature_vect_dict[query_category]))
#print(all_category_feature_vect_dict)
query_data.append(all_category_feature_vect_dict[query_category][product_index])
#print(len(query_data))
#print(query_data)
#for pro_feat in query_data:
# print(pro_feat)
relabled_query = Relabel_Products_Within_query(query_data,query_ind)
#print("Relabled ")
for pro_feat in relabled_query:
#print(pro_feat)
filehandle.write(pro_feat+"\n")
filehandle.close()
if train == 1:
shutil.copy2(train_file_path, training_testing_dest_directory+"valid.txt")
return
def Pick_Training_Testing_Features(randomized_queries_dest,queries_directory,all_product_indices,query_map_code_file_path,training_testing_dest_directory,features_base_directory,categories_with_large_products):
print("This procedure fetches the feature vector for each product associated to each query")
#First we start fetching training data
randomized_training_path = randomized_queries_dest + "training.txt"
Pick_Feature_Data_For_Given_Queries(randomized_training_path,queries_directory,all_product_indices,query_map_code_file_path,training_testing_dest_directory,features_base_directory,1,categories_with_large_products)
#Next we start fetching testing data
randomized_testing_path = randomized_queries_dest + "testing.txt"
Pick_Feature_Data_For_Given_Queries(randomized_testing_path, queries_directory, all_product_indices,
query_map_code_file_path, training_testing_dest_directory,
features_base_directory, 0,categories_with_large_products)
return
from RankingHelper import Collect_Queries_With_Given_Num_Products
def Adjust_Training_Testing_Data_Per_Cat_From_All_Cat_Data_Regression_Amazon(drive):
base_training_directory =drive+"\Yassien_PhD\Experiment_6\Train_Test_Category_TQ_Target_Sub_Cat_Setup/"
training_query_cat_path = drive+"\Yassien_PhD\Experiment_5\Randomized_Queryset/query_cat_map_train.txt"
test_query_cat_path = drive + "\Yassien_PhD\Experiment_5\Randomized_Queryset/query_cat_map_test.txt"
all_cat_traning_file_path = base_training_directory+"train.txt"
all_cat_testing_file_path = base_training_directory + "test.txt"
orig_catNames = ["Industrial & Scientific", "Jewelry", "Arts, Crafts & Sewing", "Toys & Games", "Video Games",
"Computers & Accessories", "Software", "Cell Phones & Accessories", "Electronics"]
Adjust_Training_Testing_Data_Per_Cat_From_All_Cat_Data_Regression(drive, base_training_directory, training_query_cat_path,
test_query_cat_path, all_cat_traning_file_path,
all_cat_testing_file_path, orig_catNames)
return
def Adjust_Training_Testing_Data_Per_Cat_From_All_Cat_Data_Yelp(drive):
base_training_directory =drive+"\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Experiment_3\Train_Test_Category_With_Time_Interval_TQ_Target_Sub_Cat_Setup/"
training_query_cat_path = drive+"\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Experiment_3\Randomized_Queryset/query_cat_map_train.txt"
test_query_cat_path = drive + "\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Experiment_3\Randomized_Queryset/query_cat_map_test.txt"
all_cat_traning_file_path = base_training_directory+"train.txt"
all_cat_testing_file_path = base_training_directory + "test.txt"
orig_catNames = ["Cafes", "Chinese","Mexican" , "Italian","American (Traditional)", "Thai", "Bars", "Japanese", "American (New)"]
Adjust_Training_Testing_Data_Per_Cat_From_All_Cat_Data(drive, base_training_directory, training_query_cat_path,
test_query_cat_path, all_cat_traning_file_path,
all_cat_testing_file_path, orig_catNames)
return
def Adjust_Training_Testing_Data_Per_Cat_From_All_Cat_Data_Amazon(drive):
base_training_directory =drive+"\Yassien_PhD\Experiment_5\Train_Test_Category_With_Time_Interval_TQ_Target_Sub_Cat_Setup/"
training_query_cat_path = drive+"\Yassien_PhD\Experiment_5\Randomized_Queryset/query_cat_map_train.txt"
test_query_cat_path = drive + "\Yassien_PhD\Experiment_5\Randomized_Queryset/query_cat_map_test.txt"
all_cat_traning_file_path = base_training_directory+"train.txt"
all_cat_testing_file_path = base_training_directory + "test.txt"
orig_catNames = ["Industrial & Scientific", "Jewelry", "Arts, Crafts & Sewing", "Toys & Games", "Video Games",
"Computers & Accessories", "Software", "Cell Phones & Accessories", "Electronics"]
Adjust_Training_Testing_Data_Per_Cat_From_All_Cat_Data(drive, base_training_directory, training_query_cat_path,
test_query_cat_path, all_cat_traning_file_path,
all_cat_testing_file_path, orig_catNames)
return
def Adjust_Training_Testing_Data_Per_Cat_From_All_Cat_Data(drive,base_training_directory,training_query_cat_path,test_query_cat_path,all_cat_traning_file_path,
all_cat_testing_file_path,orig_catNames):
training_query_cat_dict = dict()
testing_query_cat_dict = dict()
with open(training_query_cat_path, 'r') as fp:
for line in fp:
row = line.split('\t')
queries = []
for i in range(1,len(row)-1):
queries.append(row[i])
queries.append(row[len(row)-1].split('\n')[0])
training_query_cat_dict[row[0]]=queries
#print("training")
#print(training_query_cat_dict)
with open(test_query_cat_path, 'r') as fp:
for line in fp:
row = line.split('\t')
queries = []
for i in range(1,len(row)-1):
queries.append(row[i])
queries.append(row[len(row)-1].split('\n')[0])
testing_query_cat_dict[row[0]]=queries
#print("testing")
#print(testing_query_cat_dict)
for category_name in orig_catNames:
print("Considering "+category_name)
dest_directory=base_training_directory+category_name+"/"
try:
os.stat(dest_directory)
except:
os.mkdir(dest_directory)
training_cat_queries = training_query_cat_dict[category_name]
new_training_file_path = dest_directory+"train.txt"
filhandle = open(new_training_file_path,'w')
num_quries_found = 0
prev_id = -1
with open(all_cat_traning_file_path, 'r') as fp:
for line in fp:
row = line.split(' ')
qid = row[1].split(':')[1]
if qid in training_cat_queries:
filhandle.write(line)
if qid !=prev_id:
num_quries_found+=1
prev_id = qid
filhandle.close()
print("Original Num Training queries "+str(len(training_cat_queries))+" Written num queries "+str(num_quries_found))
shutil.copy2(new_training_file_path,dest_directory+"valid.txt")
testing_cat_queries = testing_query_cat_dict[category_name]
new_test_file_path = dest_directory + "test.txt"
filhandle = open(new_test_file_path, 'w')
num_quries_found=0
prev_id = -1
with open(all_cat_testing_file_path, 'r') as fp:
for line in fp:
row = line.split(' ')
qid = row[1].split(':')[1]
if qid in testing_cat_queries:
filhandle.write(line)
if qid !=prev_id:
num_quries_found+=1
prev_id = qid
filhandle.close()
print("Original Num Testing queries " + str(len(testing_cat_queries)) + " Written num queries " + str(num_quries_found))
print("*****************************************************************************************************************")
return
def Adjust_Training_Testing_Data_Per_Cat_From_All_Cat_Data_Regression(drive,base_training_directory,training_query_cat_path,test_query_cat_path,all_cat_traning_file_path,
all_cat_testing_file_path,orig_catNames):
query_main_directory = drive+"\Yassien_PhD\Experiment_6\Queries_Per_Product_Category_Num_Pro_GT_8/"
query_map_file_path = query_main_directory+"query_code_map.txt"
tq_rank_main_directory=drive+"\Yassien_PhD\Experiment_5\Categories_Ranked_by_TQ_Rank/"
query_id_cat_sub_cat_dict = dict()
ignore_first_line = 0
with open(query_map_file_path, 'r') as fp:
for line in fp:
if ignore_first_line == 0:
ignore_first_line = 1
continue
row = line.split('\t')
query_id_cat_sub_cat_dict[row[0]]=(row[1],row[2])
training_query_cat_dict = dict()
testing_query_cat_dict = dict()
with open(training_query_cat_path, 'r') as fp:
for line in fp:
row = line.split('\t')
queries = []
for i in range(1,len(row)-1):
queries.append(row[i])
queries.append(row[len(row)-1].split('\n')[0])
training_query_cat_dict[row[0]]=queries
#print("training")
#print(training_query_cat_dict)
with open(test_query_cat_path, 'r') as fp:
for line in fp:
row = line.split('\t')
queries = []
for i in range(1,len(row)-1):
queries.append(row[i])
queries.append(row[len(row)-1].split('\n')[0])
testing_query_cat_dict[row[0]]=queries
#print("testing")
#print(testing_query_cat_dict)
for category_name in orig_catNames:
print("Considering "+category_name)
dest_directory=base_training_directory+category_name+"/"
try:
os.stat(dest_directory)
except:
os.mkdir(dest_directory)
cat_tq_product_dict= dict()
cat_tq_file_path = tq_rank_main_directory+category_name+".txt"
with open(cat_tq_file_path, 'r') as fp:
for line in fp:
row = line.split('\t')
cat_tq_product_dict[row[0]]=row[1].split('\n')[0]
continue_train = 1
try:
training_cat_queries = training_query_cat_dict[category_name]
except KeyError:
print(category_name+" not found in training")
continue_train = 0
if continue_train == 1:
new_training_file_path = dest_directory+"train.txt"
filhandle = open(new_training_file_path,'w')
num_quries_found = 1
prev_id = -1
index = 0
product_list = []
with open(all_cat_traning_file_path, 'r') as fp:
for line in fp:
row = line.split(' ')
qid = row[1].split(':')[1]
if prev_id == -1:
prev_id = qid
index=0
product_cat = query_id_cat_sub_cat_dict[qid][0]
product_sub_cat = query_id_cat_sub_cat_dict[qid][1].split('\n')[0]
#print(product_cat+" "+product_sub_cat)
if qid in training_cat_queries:
if qid !=prev_id:
num_quries_found+=1
prev_id = qid
index=0
query_file_path = query_main_directory + product_cat + "/" + product_sub_cat + ".txt"
#if qid !=prev_id or len(product_list)==0:
product_list=[]
with open(query_file_path, 'r') as fp:
for newline in fp:
newline = newline.split('\n')
product_list.append(newline[0])
if index<len(product_list):
#print("qid is "+str(qid)+" index "+str(index))
tq_rank = cat_tq_product_dict[product_list[index]]
index += 1
filhandle.write(str(tq_rank)+" ")
for i in range(2,len(row)):
if i == len(row)-1:
filhandle.write(row[i])
else:
filhandle.write(row[i] + " ")
else:
print("didn't find index "+str(index))
print("len prod list "+str(len(product_list)))
print(qid)
print(query_file_path)
filhandle.close()
print("Original Num Training queries "+str(len(training_cat_queries))+" Written num queries "+str(num_quries_found))
shutil.copy2(new_training_file_path,dest_directory+"valid.txt")
try:
testing_cat_queries = testing_query_cat_dict[category_name]
except KeyError:
print(category_name+" not found in testing")
continue
new_test_file_path = dest_directory + "test.txt"
filhandle = open(new_test_file_path, 'w')
num_quries_found = 1
prev_id = -1
index = 0
product_list = []
with open(all_cat_testing_file_path, 'r') as fp:
for line in fp:
row = line.split(' ')
qid = row[1].split(':')[1]
if prev_id == -1:
prev_id = qid
index=0
product_cat = query_id_cat_sub_cat_dict[qid][0]
product_sub_cat = query_id_cat_sub_cat_dict[qid][1].split('\n')[0]
if qid in testing_cat_queries:
if qid !=prev_id:
num_quries_found+=1
prev_id = qid
index=0
query_file_path = query_main_directory + product_cat + "/" + product_sub_cat + ".txt"
#if qid !=prev_id or len(product_list)==0:
product_list=[]
with open(query_file_path, 'r') as fp:
for newline in fp:
newline = newline.split('\n')
product_list.append(newline[0])
if index<len(product_list):
#print("qid is "+str(qid)+" index "+str(index))
tq_rank = cat_tq_product_dict[product_list[index]]
index += 1
filhandle.write(str(tq_rank)+" ")
for i in range(2,len(row)):
if i == len(row)-1:
filhandle.write(row[i])
else:
filhandle.write(row[i] + " ")
else:
print("didn't find index "+str(index))
print("len prod list "+str(len(product_list)))
print(qid)
print(query_file_path)
filhandle.close()
print("Original Num Testing queries " + str(len(testing_cat_queries)) + " Written num queries " + str(num_quries_found))
print("*****************************************************************************************************************")
return
def Prepare_Training_Testing_Data_Sub_Cat_Experiment_Setup_Amazon(drive):
print("This procedure performs the preparation of the training and testing from queries based on sub categories for all cagegories at once Amazon Dataset")
orig_catNames = ["Industrial & Scientific", "Jewelry", "Arts, Crafts & Sewing", "Toys & Games", "Video Games",
"Computers & Accessories", "Software", "Cell Phones & Accessories", "Electronics"]
query_cat_folder = "D:\Yassien_PhD\Experiment_5\Queries_Per_Product_Category/"
for cat in orig_catNames:
print("Considering "+cat)
category_folder=query_cat_folder+cat+"/"
desired_num_products=8
dest_directory="D:\Yassien_PhD\Experiment_5\Queries_Per_Product_Category_Num_Pro_GT_8/"+cat+"/"
'''try:
os.stat(dest_directory)
except:
os.mkdir(dest_directory)
Collect_Queries_With_Given_Num_Products(category_folder,desired_num_products,dest_directory)'''
#Extract_and_Code_Queries_From_Sub_Categories()
queries_directory = drive+"\Yassien_PhD\Experiment_5\Queries_Per_Product_Category_Num_Pro_GT_8_Coded/"
all_product_indices = drive+"\Yassien_PhD\Experiment_5\All_Products_Per_Cat_Indices/"
query_map_code_file_path = drive+"\Yassien_PhD\Experiment_5\Queries_Per_Product_Category_Num_Pro_GT_8/query_code_map.txt"
training_testing_dest_directory = drive+"\Yassien_PhD\Experiment_5\Train_Test_Category_With_Time_Interval_TQ_Target_Sub_Cat_Setup/"
randomized_queries_dest = drive+"\Yassien_PhD\Experiment_5\Randomized_Queryset/"
features_base_directory = drive+"\Yassien_PhD\Experiment_4\All_Categories_Data_25_Basic_Features_With_10_Time_Intervals/"
categories_with_large_products = orig_catNames
training_ratio = 0.8
validation_ratio = 0.2
#Randomize_Queries_List_training_Testing_FromSubCats(queries_directory,randomized_queries_dest,training_ratio)
Pick_Training_Testing_Features(randomized_queries_dest,queries_directory,all_product_indices,query_map_code_file_path,training_testing_dest_directory,features_base_directory,categories_with_large_products)
return
def Prepare_Training_Testing_Data_Sub_Cat_Experiment_Setup_Yelp(drive):
print("This procedure performs the preparation of the training and testing from queries based on sub categories for all cagegories at once Yelp Dataset")
orig_catNames = ["Cafes", "Chinese","Mexican" , "Italian","American (Traditional)", "Thai", "Bars", "Japanese", "American (New)"]
query_cat_folder = "D:\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Queries_Per_Product_Category/"
for cat in orig_catNames:
print("Considering "+cat)
category_folder=query_cat_folder+cat+"/"
desired_num_products=8
dest_directory=drive+"\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Experiment_3\Queries_Per_Product_Category_Num_Pro_GT_8/"+cat+"/"
'''try:
os.stat(dest_directory)
except:
os.mkdir(dest_directory)
Collect_Queries_With_Given_Num_Products(category_folder,desired_num_products,dest_directory)'''
extracted_queries_base = drive+"\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Experiment_3\Queries_Per_Product_Category_Num_Pro_GT_8/"
base_dest = drive+"\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Experiment_3\Queries_Per_Product_Category_Num_Pro_GT_8_Coded/"
coded_query_map_path = extracted_queries_base + "query_code_map.txt"
#Extract_and_Code_Queries_From_Sub_Categories(extracted_queries_base,base_dest,coded_query_map_path,orig_catNames)
queries_directory = drive+"\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Experiment_3\Queries_Per_Product_Category_Num_Pro_GT_8_Coded/"
all_product_indices = drive+"\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Experiment_3\All_Products_Per_Cat_Indices/"
query_map_code_file_path = drive+"\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Experiment_3\Queries_Per_Product_Category_Num_Pro_GT_8/query_code_map.txt"
training_testing_dest_directory = drive+"\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Experiment_3\Train_Test_Category_With_Time_Interval_TQ_Target_Sub_Cat_Setup/"
randomized_queries_dest = drive+"\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Experiment_3\Randomized_Queryset/"
features_base_directory = drive+"\Yassien_PhD\yelp_dataset_challenge_academic_dataset\Different_time_feature_vectors\All_Categories_Data_25_Basic_Features_With_1_Time_Intervals/"
training_ratio = 0.8
validation_ratio = 0.2
#Randomize_Queries_List_training_Testing_FromSubCats(queries_directory,randomized_queries_dest,training_ratio)
Pick_Training_Testing_Features(randomized_queries_dest,queries_directory,all_product_indices,query_map_code_file_path,training_testing_dest_directory,features_base_directory,orig_catNames)
return
def Old_Kendall_Tau_Measuring_On_All_Predictions_Level(products_to_test,predictions,base_predictions_directory,category_name,R_path,correlationFileHandle):
# This is the old measuring of kendall tau metric
if len(products_to_test) != len(predictions):
print("Error Un even lists")
print("Num products from sales " + str(len(products_to_test)) + " From predictions " + str(len(predictions)))
#print("#####################################")
#print("After Sorting")
mergeSort(products_to_test)
mergeSort(predictions)
products_to_test.reverse() # reverse as it is ordered in ascending order and in our notation the higher the value the better the rank
predictions.reverse()
#print(products_to_test)
#print(predictions)
# Create sorted sales rank file
print("Writing sorted sales rank file ")
file_path = base_predictions_directory + category_name + "/Cutoff_10/" + "Sorted_Sales_Rank.txt"
filehandle = open(file_path, 'w')
filehandle.write("Index\tRank\n")
for sales_rank in products_to_test:
filehandle.write(str(sales_rank[0]) + "\t" + str(sales_rank[1]) + "\n")
filehandle.close()
# Create sorted predictions
print("Writing sorted predictions file ")
file_path = base_predictions_directory + category_name + "/Cutoff_10/" + "Sorted_Predictions.txt"
filehandle = open(file_path, 'w')
filehandle.write("Index\tValue\n")
for pred in predictions:
filehandle.write(str(pred[0]) + "\t" + str(pred[1]) + "\n")
filehandle.close()
# Create R_Difference file where you put the two
print("Writing R_Difference File for kendall Calculation")
r_difference_path = base_predictions_directory + category_name + "/Cutoff_10/" + "R_Difference.txt"
filehandle = open(r_difference_path, 'w')
for i in range(len(products_to_test)):
filehandle.write(str(i + 1) + "\t")
initial_index = products_to_test[i][0]
temp_index = 0
for pred in predictions:
if initial_index == pred[0]:
break
temp_index += 1
filehandle.write(str(temp_index + 1) + "\n")
filehandle.close()
from Testing import runKenallExtractScript, writeCorrelationRScriptOneFile
# Creating the R Script to run Kendall tau
rScriptFilePath = writeCorrelationRScriptOneFile(r_difference_path, 1,
base_predictions_directory + category_name + "/Cutoff_10/")
print(rScriptFilePath)
print("Kendall value is ")
kendall = runKenallExtractScript(rScriptFilePath, R_path)
print("The avg Kendall ")
print(kendall)
correlationFileHandle.write(category_name + "\t\t" + str(kendall) + "\n")
return
def New_Kendall_Tau_Measuring_On_Query_Level(base_predictions_directory,category_name,products_to_test,R_path,correlationFileHandle):
# Here is the new kendall tau metric measurment which will be on the query base and then we average the results from Sales Rank way
# '''
r_difference_folder_path = base_predictions_directory + category_name + "/Cutoff_10/R_Difference/"
try:
os.stat(r_difference_folder_path)
except:
os.mkdir(r_difference_folder_path)
testing_path = base_predictions_directory + category_name + "/Cutoff_10/" + "test.txt"
query_ranks = []
prods_per_query = []
num_products = 0
query = []
queryid = ""
nums = 0
with open(testing_path, 'r') as filep:
for line in filep:
row = line.split(' ')
if queryid == "":
queryid = row[1]
nums += 1
num_products += 1
continue
if queryid != row[1]:
queryid = row[1]
prods_per_query.append(nums)
nums = 1
num_products += 1
else:
nums += 1
num_products += 1
if nums > 0:
prods_per_query.append(nums)
# print("Nums per query")
# print(prods_per_query)
# print("****************************************************")
index = 0
for k in range(len(prods_per_query)):
nums = prods_per_query[k]
inner = 0
query = []
for z in range(index, index + nums):
pair = products_to_test[z]
pair = (inner, pair[1])
products_to_test[z] = pair
query.append(products_to_test[z])
# print(products_to_test[z])
inner += 1
query_ranks.append(query)
index += nums
# print("****************************************************")
print("Num Products " + str(num_products))
predictions = []
predictions_path = base_predictions_directory + category_name + "/Cutoff_10/" + "predictions.txt"
with open(predictions_path, 'r') as filep:
for line in filep:
predictions.append(float(line))
query_predictions = []
index = 0
for i in range(len(query_ranks)):
query = query_ranks[i]
#print("query Before ")
#print(query)
given_salesrank_query = []
for p in query: