From 7391cc77c6044d79700aed6b0e8e019736abfdcb Mon Sep 17 00:00:00 2001
From: liberty <2700472871@qq.com>
Date: Thu, 30 May 2024 14:44:30 +0800
Subject: [PATCH 1/8] [model] add dhn

---
 examples/dhn/MA.txt         | 7071 +++++++++++++++++++++++++++++++++++
 examples/dhn/dhn_trainer.py |  361 ++
 gammagl/models/dhn.py       |   82 +
 3 files changed, 7514 insertions(+)
 create mode 100644 examples/dhn/MA.txt
 create mode 100644 examples/dhn/dhn_trainer.py
 create mode 100644 gammagl/models/dhn.py

diff --git a/examples/dhn/MA.txt b/examples/dhn/MA.txt
new file mode 100644
index 00000000..3d275250
--- /dev/null
+++ b/examples/dhn/MA.txt
@@ -0,0 +1,7071 @@
+M0 A1177
+M1 A742
+M2 A478
+M3 A713
+M5 A955
+M6 A698
+M8 A99
+M9 A996
+M10 A580
+M11 A1235
+M12 A292
+M13 A742
+M14 A742
+M15 A580
+M16 A418
+M17 A99
+M18 A742
+M19 A1009
+M20 A188
+M21 A562
+M22 A753
+M23 A188
+M24 A962
+M25 A1043
+M26 A219
+M27 A1040
+M28 A419
+M29 A851
+M30 A594
+M31 A698
+M32 A1040
+M33 A742
+M34 A1054
+M36 A148
+M37 A887
+M39 A562
+M40 A340
+M41 A778
+M42 A1117
+M44 A826
+M45 A816
+M46 A1212
+M47 A287
+M48 A856
+M49 A794
+M50 A219
+M51 A641
+M52 A687
+M53 A148
+M54 A566
+M58 A53
+M59 A187
+M60 A284
+M61 A964
+M62 A349
+M63 A478
+M64 A1213
+M65 A287
+M66 A826
+M67 A53
+M68 A423
+M69 A1040
+M70 A1018
+M71 A1009
+M72 A19
+M73 A1009
+M74 A842
+M75 A1050
+M76 A142
+M77 A338
+M78 A423
+M79 A974
+M80 A1040
+M83 A70
+M84 A1341
+M86 A974
+M87 A1350
+M88 A1310
+M91 A224
+M92 A865
+M93 A908
+M94 A650
+M95 A123
+M96 A550
+M97 A219
+M99 A188
+M100 A1024
+M101 A178
+M102 A287
+M103 A287
+M104 A903
+M105 A842
+M106 A742
+M107 A411
+M108 A28
+M109 A698
+M110 A6
+M111 A645
+M112 A707
+M113 A808
+M114 A39
+M115 A39
+M116 A1009
+M117 A1009
+M118 A742
+M119 A937
+M120 A826
+M121 A1007
+M122 A591
+M123 A1054
+M126 A3
+M127 A99
+M128 A713
+M129 A224
+M130 A99
+M134 A742
+M135 A1050
+M136 A808
+M137 A478
+M141 A831
+M142 A349
+M144 A1054
+M145 A1054
+M146 A147
+M147 A178
+M148 A445
+M149 A702
+M150 A1381
+M151 A245
+M152 A1009
+M153 A964
+M155 A1050
+M157 A1371
+M158 A1050
+M159 A826
+M160 A99
+M161 A698
+M162 A908
+M163 A908
+M164 A176
+M165 A735
+M166 A19
+M168 A349
+M169 A550
+M171 A1035
+M172 A272
+M173 A248
+M174 A39
+M175 A591
+M176 A147
+M178 A1003
+M179 A219
+M181 A742
+M182 A856
+M183 A351
+M184 A133
+M185 A1310
+M186 A287
+M187 A39
+M188 A274
+M189 A1050
+M190 A1009
+M191 A1382
+M192 A133
+M193 A808
+M194 A519
+M195 A1073
+M197 A1009
+M200 A591
+M202 A1073
+M203 A1117
+M205 A742
+M207 A287
+M208 A224
+M209 A1178
+M210 A1054
+M211 A1024
+M212 A1040
+M213 A419
+M214 A1282
+M215 A1263
+M216 A317
+M217 A671
+M218 A21
+M219 A641
+M220 A1050
+M221 A829
+M222 A1035
+M224 A903
+M225 A351
+M226 A1199
+M227 A539
+M228 A1234
+M229 A351
+M230 A1373
+M231 A70
+M232 A592
+M234 A1050
+M235 A9
+M236 A1039
+M237 A1039
+M238 A1345
+M239 A1054
+M240 A1039
+M241 A1117
+M243 A903
+M245 A773
+M249 A1154
+M250 A591
+M251 A36
+M252 A984
+M253 A389
+M254 A178
+M255 A178
+M256 A156
+M257 A219
+M258 A555
+M259 A478
+M263 A1345
+M264 A1043
+M265 A1054
+M267 A1145
+M268 A566
+M269 A245
+M270 A962
+M271 A1213
+M272 A1009
+M274 A1009
+M275 A419
+M276 A643
+M277 A1340
+M278 A351
+M280 A974
+M282 A903
+M283 A318
+M284 A1050
+M285 A1073
+M286 A1125
+M287 A1046
+M288 A572
+M289 A742
+M290 A1179
+M291 A1089
+M292 A1179
+M293 A577
+M294 A36
+M295 A464
+M296 A219
+M297 A1199
+M298 A507
+M299 A816
+M300 A903
+M302 A1339
+M303 A1054
+M304 A490
+M305 A742
+M306 A1177
+M307 A219
+M308 A219
+M309 A671
+M310 A1027
+M311 A224
+M312 A1281
+M314 A36
+M315 A338
+M316 A29
+M318 A947
+M319 A280
+M320 A99
+M321 A861
+M322 A351
+M324 A543
+M325 A36
+M326 A219
+M327 A826
+M329 A1367
+M330 A466
+M331 A29
+M332 A1141
+M334 A916
+M335 A418
+M336 A783
+M337 A1265
+M338 A1374
+M339 A263
+M340 A962
+M341 A340
+M342 A133
+M344 A899
+M345 A910
+M346 A1326
+M347 A245
+M348 A546
+M349 A1110
+M351 A441
+M352 A1009
+M353 A224
+M354 A1179
+M356 A566
+M357 A886
+M358 A1314
+M359 A482
+M360 A1048
+M361 A219
+M363 A1040
+M364 A964
+M365 A245
+M366 A1040
+M368 A425
+M369 A473
+M370 A1050
+M371 A982
+M372 A418
+M373 A1163
+M374 A953
+M375 A550
+M376 A58
+M377 A434
+M378 A441
+M379 A908
+M380 A1073
+M381 A528
+M382 A178
+M383 A961
+M384 A243
+M385 A591
+M386 A961
+M387 A566
+M388 A419
+M389 A1154
+M390 A300
+M391 A919
+M392 A291
+M393 A340
+M394 A831
+M395 A1024
+M396 A224
+M397 A591
+M398 A351
+M399 A566
+M400 A178
+M401 A1141
+M402 A1345
+M405 A1179
+M406 A466
+M407 A1131
+M408 A156
+M409 A1009
+M410 A714
+M411 A197
+M412 A1393
+M413 A156
+M414 A506
+M415 A36
+M416 A335
+M417 A954
+M420 A808
+M421 A155
+M422 A870
+M423 A246
+M424 A589
+M425 A148
+M426 A32
+M427 A414
+M428 A95
+M429 A982
+M430 A259
+M431 A565
+M432 A885
+M433 A547
+M434 A1040
+M435 A935
+M436 A9
+M437 A816
+M438 A1050
+M439 A287
+M440 A123
+M441 A671
+M442 A245
+M443 A950
+M444 A1345
+M445 A9
+M446 A245
+M447 A982
+M448 A423
+M449 A1345
+M450 A1048
+M451 A1127
+M452 A219
+M453 A1393
+M454 A1213
+M455 A903
+M456 A982
+M457 A224
+M458 A982
+M460 A1345
+M461 A861
+M462 A831
+M463 A1179
+M464 A591
+M465 A419
+M467 A1305
+M468 A1127
+M469 A246
+M470 A178
+M471 A566
+M472 A411
+M474 A861
+M475 A70
+M476 A1179
+M478 A1345
+M481 A753
+M482 A242
+M483 A1054
+M484 A1214
+M486 A982
+M487 A351
+M488 A1039
+M490 A808
+M491 A245
+M492 A993
+M493 A363
+M496 A691
+M498 A1024
+M499 A1000
+M500 A593
+M501 A1373
+M503 A667
+M505 A1047
+M508 A593
+M509 A1131
+M510 A490
+M512 A831
+M513 A698
+M514 A197
+M515 A1024
+M516 A340
+M517 A1305
+M520 A448
+M521 A590
+M522 A1009
+M523 A1291
+M524 A445
+M525 A687
+M526 A1054
+M527 A133
+M528 A903
+M529 A1314
+M530 A982
+M531 A988
+M532 A248
+M534 A9
+M535 A1345
+M536 A224
+M538 A961
+M540 A831
+M541 A577
+M542 A391
+M544 A36
+M545 A1188
+M546 A245
+M547 A328
+M548 A580
+M549 A982
+M550 A1345
+M551 A245
+M552 A458
+M553 A233
+M554 A580
+M556 A351
+M557 A566
+M558 A1290
+M560 A691
+M561 A546
+M562 A964
+M564 A1350
+M565 A903
+M566 A139
+M568 A550
+M569 A982
+M570 A645
+M571 A29
+M572 A936
+M574 A641
+M576 A968
+M577 A363
+M579 A851
+M580 A284
+M581 A904
+M582 A970
+M583 A53
+M584 A246
+M585 A1373
+M587 A701
+M588 A539
+M589 A591
+M590 A36
+M591 A245
+M592 A317
+M594 A1050
+M595 A851
+M596 A963
+M597 A423
+M598 A1350
+M599 A908
+M600 A443
+M601 A773
+M602 A1345
+M603 A1103
+M604 A286
+M605 A1048
+M606 A66
+M607 A90
+M609 A1179
+M610 A580
+M611 A178
+M612 A349
+M613 A808
+M614 A36
+M615 A571
+M618 A245
+M620 A1272
+M625 A99
+M627 A1352
+M628 A363
+M631 A39
+M632 A577
+M633 A590
+M635 A419
+M638 A803
+M639 A197
+M641 A219
+M642 A590
+M643 A754
+M644 A907
+M646 A968
+M648 A224
+M650 A1310
+M651 A567
+M652 A9
+M653 A338
+M654 A1127
+M655 A224
+M656 A1282
+M657 A1052
+M658 A1054
+M659 A224
+M660 A29
+M661 A808
+M662 A1040
+M663 A336
+M664 A977
+M665 A363
+M666 A352
+M667 A245
+M668 A903
+M669 A246
+M671 A1050
+M672 A229
+M673 A1009
+M674 A713
+M675 A1117
+M676 A1365
+M677 A1215
+M678 A1141
+M679 A816
+M682 A885
+M683 A178
+M684 A550
+M686 A245
+M687 A313
+M690 A645
+M691 A338
+M692 A566
+M693 A928
+M694 A710
+M695 A1380
+M696 A903
+M697 A585
+M698 A546
+M699 A566
+M701 A643
+M702 A550
+M707 A29
+M708 A571
+M709 A908
+M710 A577
+M711 A1121
+M712 A224
+M714 A1339
+M715 A304
+M716 A351
+M718 A122
+M719 A681
+M721 A661
+M722 A1179
+M724 A816
+M725 A216
+M726 A363
+M728 A1050
+M729 A1120
+M731 A910
+M732 A903
+M733 A245
+M734 A197
+M735 A974
+M736 A1050
+M737 A36
+M738 A308
+M739 A1050
+M740 A1272
+M742 A351
+M743 A706
+M745 A588
+M746 A1179
+M747 A1040
+M748 A821
+M749 A817
+M751 A935
+M753 A698
+M754 A253
+M755 A778
+M760 A118
+M761 A287
+M763 A631
+M764 A1108
+M766 A567
+M767 A9
+M768 A1256
+M769 A826
+M770 A383
+M771 A1122
+M772 A351
+M773 A964
+M774 A246
+M778 A713
+M779 A1135
+M781 A103
+M783 A577
+M784 A577
+M785 A245
+M786 A1040
+M787 A351
+M788 A831
+M789 A1048
+M793 A582
+M794 A99
+M795 A572
+M796 A974
+M798 A958
+M799 A326
+M801 A188
+M802 A603
+M803 A279
+M804 A550
+M805 A691
+M806 A974
+M807 A742
+M808 A816
+M809 A133
+M811 A910
+M813 A772
+M814 A123
+M815 A1072
+M817 A260
+M818 A1206
+M821 A1120
+M822 A928
+M823 A54
+M824 A1141
+M827 A1127
+M828 A982
+M829 A546
+M830 A1382
+M831 A903
+M832 A826
+M834 A1373
+M835 A36
+M836 A224
+M837 A98
+M838 A577
+M840 A119
+M841 A445
+M842 A1009
+M843 A691
+M844 A1245
+M845 A224
+M846 A1157
+M847 A141
+M848 A1179
+M849 A506
+M850 A1179
+M851 A1009
+M852 A421
+M853 A1199
+M854 A903
+M855 A151
+M856 A1040
+M858 A865
+M859 A363
+M860 A245
+M861 A75
+M862 A284
+M863 A399
+M864 A19
+M865 A423
+M866 A177
+M868 A577
+M869 A1111
+M871 A886
+M872 A1290
+M873 A808
+M874 A351
+M876 A1382
+M877 A133
+M878 A1127
+M879 A577
+M880 A908
+M881 A214
+M882 A242
+M886 A147
+M887 A1038
+M889 A605
+M890 A573
+M891 A363
+M892 A844
+M893 A1009
+M894 A363
+M895 A197
+M896 A263
+M898 A1395
+M900 A156
+M901 A826
+M902 A762
+M903 A1213
+M904 A1345
+M905 A742
+M906 A1024
+M907 A908
+M908 A855
+M909 A1340
+M910 A1040
+M911 A219
+M912 A1376
+M913 A1393
+M914 A51
+M915 A123
+M916 A1142
+M917 A650
+M918 A1117
+M920 A577
+M923 A326
+M924 A1166
+M925 A1131
+M926 A39
+M929 A996
+M930 A1050
+M931 A1157
+M932 A10
+M933 A591
+M934 A566
+M935 A1345
+M937 A246
+M938 A550
+M939 A349
+M940 A178
+M941 A123
+M942 A36
+M943 A246
+M944 A29
+M945 A742
+M946 A1179
+M947 A739
+M948 A371
+M949 A974
+M950 A903
+M951 A123
+M953 A383
+M954 A1048
+M955 A1089
+M956 A831
+M957 A1235
+M958 A419
+M959 A36
+M960 A1039
+M964 A193
+M965 A284
+M966 A317
+M967 A1104
+M968 A196
+M969 A351
+M970 A702
+M972 A842
+M973 A977
+M974 A250
+M975 A656
+M977 A1127
+M978 A1245
+M979 A1235
+M980 A595
+M981 A222
+M982 A962
+M983 A1040
+M984 A1316
+M987 A1179
+M989 A1120
+M990 A219
+M991 A1192
+M993 A698
+M995 A1199
+M996 A562
+M997 A351
+M998 A383
+M999 A628
+M1000 A954
+M1001 A245
+M1003 A1290
+M1004 A977
+M1005 A1043
+M1006 A1345
+M1008 A1373
+M1009 A1048
+M1010 A550
+M1011 A1235
+M1012 A337
+M1013 A419
+M1014 A250
+M1015 A1024
+M1017 A1039
+M1018 A1043
+M1019 A1287
+M1020 A778
+M1021 A70
+M1022 A1043
+M1025 A66
+M1026 A908
+M1027 A1032
+M1028 A591
+M1029 A1141
+M1030 A9
+M1031 A1310
+M1032 A903
+M1033 A520
+M1034 A816
+M1036 A619
+M1037 A831
+M1038 A542
+M1039 A566
+M1040 A375
+M1041 A1054
+M1042 A1179
+M1043 A885
+M1045 A1310
+M1046 A713
+M1048 A1235
+M1049 A1054
+M1050 A691
+M1051 A903
+M1052 A826
+M1053 A419
+M1054 A961
+M1055 A1131
+M1058 A296
+M1059 A57
+M1060 A363
+M1063 A246
+M1065 A246
+M1066 A197
+M1067 A443
+M1068 A539
+M1069 A1179
+M1070 A1215
+M1072 A1139
+M1073 A697
+M1074 A419
+M1075 A586
+M1076 A1076
+M1077 A490
+M1078 A243
+M1079 A1179
+M1080 A670
+M1081 A9
+M1082 A566
+M1083 A482
+M1084 A1054
+M1085 A1179
+M1086 A996
+M1087 A987
+M1088 A14
+M1089 A963
+M1090 A349
+M1091 A320
+M1092 A916
+M1093 A578
+M1094 A51
+M1095 A1235
+M1096 A742
+M1097 A363
+M1098 A223
+M1100 A490
+M1102 A236
+M1103 A1141
+M1104 A473
+M1105 A1272
+M1106 A714
+M1107 A208
+M1108 A1384
+M1109 A306
+M1110 A1353
+M1111 A903
+M1112 A561
+M1113 A81
+M1114 A219
+M1115 A245
+M1116 A1373
+M1118 A864
+M1120 A86
+M1122 A95
+M1123 A1046
+M1124 A351
+M1125 A808
+M1126 A817
+M1127 A318
+M1128 A338
+M1129 A954
+M1130 A903
+M1131 A588
+M1134 A194
+M1137 A740
+M1138 A566
+M1139 A908
+M1140 A577
+M1141 A363
+M1142 A338
+M1144 A340
+M1145 A1038
+M1146 A1073
+M1147 A335
+M1148 A1345
+M1149 A9
+M1151 A1050
+M1153 A816
+M1154 A349
+M1155 A246
+M1156 A894
+M1157 A831
+M1158 A920
+M1159 A313
+M1161 A1382
+M1162 A9
+M1163 A340
+M1164 A1065
+M1166 A501
+M1167 A70
+M1168 A577
+M1170 A903
+M1171 A19
+M1172 A948
+M1173 A910
+M1176 A679
+M1177 A1146
+M1179 A389
+M1181 A245
+M1183 A1300
+M1184 A209
+M1185 A287
+M1186 A1117
+M1187 A993
+M1188 A1365
+M1189 A974
+M1190 A1376
+M1191 A363
+M1192 A1235
+M1195 A224
+M1196 A925
+M1201 A1365
+M1203 A1127
+M1204 A714
+M1205 A566
+M1206 A831
+M1207 A1048
+M1208 A1024
+M1209 A577
+M1210 A722
+M1211 A1151
+M1212 A1023
+M1213 A861
+M1214 A546
+M1215 A698
+M1216 A1097
+M1217 A1275
+M1218 A1018
+M1219 A698
+M1220 A336
+M1221 A111
+M1222 A223
+M1224 A224
+M1225 A559
+M1227 A768
+M1228 A762
+M1230 A580
+M1231 A550
+M1232 A419
+M1233 A826
+M1234 A1235
+M1237 A271
+M1238 A695
+M1239 A450
+M1240 A742
+M1242 A607
+M1243 A645
+M1246 A351
+M1247 A1040
+M1248 A808
+M1250 A303
+M1251 A876
+M1253 A1329
+M1255 A390
+M1256 A1106
+M1257 A903
+M1258 A471
+M1263 A1235
+M1264 A143
+M1266 A768
+M1267 A419
+M1268 A697
+M1269 A224
+M1271 A997
+M1272 A66
+M1273 A861
+M1274 A754
+M1275 A1274
+M1276 A119
+M1277 A257
+M1278 A1039
+M1280 A767
+M1281 A577
+M1282 A1381
+M1283 A1232
+M1284 A70
+M1285 A577
+M1286 A46
+M1287 A556
+M1288 A197
+M1290 A762
+M1291 A903
+M1292 A363
+M1293 A1043
+M1295 A1048
+M1296 A133
+M1298 A92
+M1299 A217
+M1301 A259
+M1303 A681
+M1304 A490
+M1305 A1073
+M1306 A652
+M1307 A1007
+M1308 A704
+M1310 A844
+M1311 A631
+M1312 A754
+M1313 A521
+M1314 A1035
+M1317 A947
+M1318 A196
+M1319 A104
+M1320 A133
+M1321 A392
+M1322 A117
+M1325 A1035
+M1326 A593
+M1328 A471
+M1332 A1024
+M1333 A968
+M1334 A709
+M1336 A1395
+M1337 A577
+M1338 A850
+M1339 A1
+M1340 A908
+M1341 A65
+M1342 A1345
+M1344 A78
+M1345 A39
+M1348 A1179
+M1349 A14
+M1353 A817
+M1354 A742
+M1355 A698
+M1356 A55
+M1357 A648
+M1358 A340
+M1359 A444
+M1360 A1373
+M1362 A107
+M1363 A1048
+M1365 A617
+M1366 A546
+M1368 A363
+M1369 A317
+M1370 A589
+M1371 A698
+M1373 A99
+M1374 A1025
+M1375 A245
+M1377 A9
+M1378 A885
+M1379 A1179
+M1380 A956
+M1383 A750
+M1386 A9
+M1387 A242
+M1388 A1373
+M1389 A562
+M1390 A741
+M1391 A275
+M1392 A808
+M1393 A312
+M1394 A327
+M1395 A678
+M1396 A1007
+M1397 A123
+M1399 A628
+M1400 A671
+M1401 A1232
+M1402 A252
+M1405 A741
+M1406 A1235
+M1407 A351
+M1408 A714
+M1409 A425
+M1410 A1393
+M1411 A577
+M1412 A1235
+M1414 A1007
+M1415 A51
+M1416 A1235
+M1417 A705
+M1418 A328
+M1419 A654
+M1420 A742
+M1421 A847
+M1422 A219
+M1423 A880
+M1424 A1035
+M1425 A400
+M1426 A1157
+M1427 A1018
+M1428 A224
+M1429 A1179
+M1431 A591
+M1433 A591
+M1434 A1367
+M1436 A742
+M1437 A1141
+M1438 A1207
+M1439 A698
+M1440 A1157
+M1441 A379
+M1443 A1124
+M1445 A1111
+M1449 A742
+M1450 A38
+M1451 A815
+M1452 A1039
+M1453 A219
+M1454 A448
+M1456 A709
+M1457 A308
+M1458 A691
+M1460 A1010
+M1461 A1312
+M1462 A1235
+M1463 A577
+M1464 A102
+M1467 A243
+M1471 A1040
+M1472 A630
+M1474 A1149
+M1475 A338
+M1476 A87
+M1477 A318
+M1478 A816
+M1479 A762
+M1480 A691
+M1483 A714
+M1484 A618
+M1485 A180
+M1486 A1235
+M1487 A201
+M1488 A109
+M1489 A107
+M1490 A178
+M1492 A211
+M1493 A389
+M1495 A760
+M1496 A338
+M1497 A1056
+M1498 A716
+M1500 A709
+M1501 A352
+M1502 A1213
+M1503 A139
+M1505 A1104
+M1508 A1341
+M1510 A14
+M1512 A1111
+M1513 A202
+M1514 A1054
+M1516 A903
+M1519 A742
+M1520 A281
+M1521 A908
+M1522 A858
+M1523 A1014
+M1524 A1050
+M1525 A636
+M1526 A1022
+M1530 A724
+M1533 A582
+M1534 A816
+M1535 A133
+M1538 A156
+M1539 A652
+M1540 A418
+M1541 A935
+M1542 A622
+M1543 A1048
+M1545 A995
+M1546 A349
+M1548 A577
+M1549 A1024
+M1551 A729
+M1552 A641
+M1553 A59
+M1554 A441
+M1555 A432
+M1556 A962
+M1558 A250
+M1559 A816
+M1560 A219
+M1561 A1008
+M1562 A903
+M1564 A326
+M1565 A1163
+M1567 A367
+M1568 A1048
+M1569 A1263
+M1571 A566
+M1573 A857
+M1574 A70
+M1575 A714
+M1576 A1179
+M1577 A1055
+M1578 A1048
+M1579 A808
+M1580 A64
+M1581 A546
+M1582 A1381
+M1584 A352
+M1585 A118
+M1586 A1345
+M1587 A1245
+M1588 A224
+M1589 A1127
+M1591 A566
+M1592 A383
+M1593 A880
+M1594 A196
+M1595 A571
+M1596 A661
+M1597 A6
+M1598 A243
+M1599 A327
+M1600 A9
+M1601 A1348
+M1603 A1308
+M1604 A977
+M1605 A577
+M1606 A383
+M1607 A1310
+M1610 A931
+M1611 A808
+M1612 A698
+M1613 A237
+M1615 A338
+M1616 A643
+M1617 A390
+M1618 A714
+M1619 A1157
+M1621 A863
+M1624 A243
+M1626 A645
+M1627 A249
+M1630 A964
+M1633 A367
+M1636 A908
+M1637 A993
+M1638 A856
+M1639 A929
+M1640 A1010
+M1642 A742
+M1645 A641
+M1646 A103
+M1647 A566
+M1648 A1163
+M1649 A277
+M1650 A18
+M1651 A29
+M1652 A1165
+M1654 A1319
+M1655 A577
+M1657 A1048
+M1660 A1390
+M1661 A224
+M1662 A1018
+M1665 A36
+M1666 A1007
+M1668 A349
+M1672 A1377
+M1673 A564
+M1674 A714
+M1675 A1179
+M1676 A560
+M1677 A722
+M1680 A1106
+M1681 A1007
+M1682 A1122
+M1683 A1141
+M1685 A46
+M1688 A641
+M1689 A197
+M1690 A317
+M1691 A245
+M1692 A1379
+M1694 A1376
+M1696 A112
+M1699 A826
+M1701 A1040
+M1703 A708
+M1705 A918
+M1707 A9
+M1708 A359
+M1711 A411
+M1712 A89
+M1713 A1179
+M1714 A577
+M1715 A28
+M1716 A197
+M1718 A349
+M1719 A376
+M1722 A178
+M1725 A816
+M1726 A908
+M1728 A858
+M1734 A589
+M1735 A843
+M1738 A1213
+M1739 A519
+M1741 A111
+M1742 A418
+M1743 A1179
+M1744 A588
+M1747 A591
+M1749 A566
+M1751 A9
+M1752 A714
+M1753 A742
+M1754 A996
+M1755 A1175
+M1756 A383
+M1757 A894
+M1759 A147
+M1760 A550
+M1762 A1048
+M1763 A740
+M1764 A577
+M1765 A530
+M1768 A1141
+M1770 A410
+M1771 A154
+M1772 A1171
+M1773 A985
+M1774 A908
+M1775 A473
+M1777 A566
+M1778 A419
+M1779 A1290
+M1780 A935
+M1783 A1051
+M1785 A399
+M1786 A452
+M1787 A816
+M1789 A321
+M1792 A1018
+M1794 A1024
+M1797 A1145
+M1798 A648
+M1801 A698
+M1803 A425
+M1804 A44
+M1805 A974
+M1806 A9
+M1807 A62
+M1808 A422
+M1811 A284
+M1812 A245
+M1813 A546
+M1814 A929
+M1815 A173
+M1816 A754
+M1820 A550
+M1821 A123
+M1822 A692
+M1824 A741
+M1825 A246
+M1826 A301
+M1827 A577
+M1828 A1272
+M1829 A577
+M1830 A1352
+M1832 A893
+M1833 A1089
+M1834 A590
+M1835 A1111
+M1837 A758
+M1838 A910
+M1839 A661
+M1840 A935
+M1841 A246
+M1844 A72
+M1845 A408
+M1846 A1110
+M1847 A546
+M1848 A36
+M1849 A974
+M1850 A66
+M1851 A880
+M1852 A1039
+M1853 A789
+M1854 A577
+M1857 A602
+M1859 A1317
+M1861 A881
+M1862 A641
+M1864 A1117
+M1865 A716
+M1867 A591
+M1868 A1050
+M1869 A977
+M1870 A519
+M1871 A419
+M1872 A311
+M1873 A14
+M1874 A419
+M1875 A562
+M1876 A856
+M1877 A414
+M1879 A266
+M1881 A1326
+M1882 A1066
+M1884 A577
+M1885 A375
+M1886 A934
+M1888 A756
+M1889 A584
+M1890 A748
+M1891 A1039
+M1892 A66
+M1893 A19
+M1895 A874
+M1896 A747
+M1898 A742
+M1899 A857
+M1900 A419
+M1901 A577
+M1903 A577
+M1904 A789
+M1905 A51
+M1906 A197
+M1907 A772
+M1908 A1046
+M1909 A698
+M1910 A240
+M1914 A149
+M1915 A1066
+M1917 A562
+M1918 A400
+M1919 A1350
+M1920 A340
+M1921 A51
+M1924 A19
+M1925 A714
+M1926 A349
+M1928 A267
+M1929 A1157
+M1933 A97
+M1935 A273
+M1937 A9
+M1939 A1048
+M1940 A367
+M1941 A856
+M1943 A634
+M1944 A1054
+M1945 A1179
+M1948 A349
+M1949 A1035
+M1952 A982
+M1958 A49
+M1959 A90
+M1961 A467
+M1962 A965
+M1965 A683
+M1966 A713
+M1967 A1155
+M1969 A376
+M1970 A645
+M1971 A56
+M1973 A964
+M1974 A389
+M1976 A688
+M1978 A713
+M1979 A996
+M1980 A777
+M1981 A338
+M1982 A687
+M1984 A93
+M1985 A691
+M1986 A329
+M1987 A237
+M1989 A250
+M1992 A841
+M1993 A808
+M1994 A724
+M1995 A1043
+M1998 A301
+M1999 A156
+M2000 A51
+M2003 A405
+M2005 A903
+M2007 A360
+M2008 A1186
+M2011 A539
+M2015 A209
+M2016 A1127
+M2017 A885
+M2018 A425
+M2019 A363
+M2020 A947
+M2021 A468
+M2022 A1039
+M2023 A382
+M2024 A245
+M2025 A513
+M2026 A29
+M2027 A340
+M2028 A628
+M2031 A9
+M2033 A789
+M2034 A133
+M2035 A695
+M2036 A863
+M2038 A284
+M2039 A1116
+M2040 A722
+M2041 A482
+M2042 A1235
+M2043 A1024
+M2046 A36
+M2047 A826
+M2048 A562
+M2049 A1043
+M2050 A982
+M2051 A566
+M2052 A1009
+M2053 A367
+M2054 A789
+M2055 A383
+M2056 A1373
+M2057 A1054
+M2058 A974
+M2059 A1054
+M2060 A1024
+M2062 A1039
+M2063 A918
+M2065 A349
+M2069 A767
+M2070 A464
+M2071 A910
+M2072 A924
+M2076 A973
+M2081 A1199
+M2082 A946
+M2083 A1052
+M2084 A546
+M2085 A36
+M2086 A977
+M2087 A222
+M2088 A1076
+M2090 A1112
+M2091 A513
+M2093 A1038
+M2094 A373
+M2095 A351
+M2096 A808
+M2099 A1073
+M2101 A350
+M2102 A743
+M2103 A46
+M2104 A29
+M2105 A974
+M2107 A974
+M2108 A861
+M2109 A816
+M2110 A202
+M2111 A1043
+M2113 A66
+M2115 A691
+M2118 A109
+M2121 A1010
+M2122 A51
+M2123 A870
+M2124 A577
+M2126 A826
+M2127 A1111
+M2128 A444
+M2129 A340
+M2132 A977
+M2133 A677
+M2134 A418
+M2136 A1127
+M2137 A287
+M2138 A43
+M2139 A1318
+M2140 A816
+M2141 A318
+M2144 A1339
+M2146 A641
+M2149 A287
+M2151 A117
+M2152 A566
+M2153 A876
+M2154 A651
+M2156 A655
+M2157 A1019
+M2158 A808
+M2160 A867
+M2161 A870
+M2162 A1089
+M2163 A428
+M2164 A1022
+M2165 A571
+M2166 A104
+M2167 A17
+M2168 A775
+M2170 A785
+M2171 A1172
+M2172 A1154
+M2173 A772
+M2174 A229
+M2175 A1109
+M2176 A636
+M2177 A742
+M2179 A1168
+M2180 A870
+M2181 A1157
+M2184 A1047
+M2189 A842
+M2190 A1068
+M2192 A151
+M2193 A300
+M2194 A191
+M2195 A1068
+M2196 A414
+M2198 A1127
+M2199 A591
+M2203 A324
+M2204 A178
+M2205 A531
+M2208 A1090
+M2209 A61
+M2212 A910
+M2214 A167
+M2217 A782
+M2218 A1111
+M2219 A630
+M2220 A1241
+M2221 A778
+M2222 A816
+M2223 A156
+M2224 A1089
+M2225 A1127
+M2226 A710
+M2230 A833
+M2231 A593
+M2232 A996
+M2234 A322
+M2236 A842
+M2237 A523
+M2239 A250
+M2242 A1336
+M2243 A556
+M2244 A196
+M2245 A80
+M2246 A690
+M2247 A1131
+M2248 A284
+M2249 A982
+M2250 A885
+M2251 A260
+M2253 A855
+M2255 A1040
+M2260 A1043
+M2262 A885
+M2263 A730
+M2267 A246
+M2268 A1319
+M2269 A453
+M2270 A1048
+M2273 A423
+M2274 A816
+M2275 A1213
+M2277 A691
+M2278 A903
+M2279 A948
+M2280 A802
+M2282 A1018
+M2284 A129
+M2285 A506
+M2286 A156
+M2287 A881
+M2289 A123
+M2290 A577
+M2291 A987
+M2292 A418
+M2293 A749
+M2294 A577
+M2296 A1178
+M2297 A432
+M2298 A818
+M2305 A741
+M2306 A1346
+M2308 A1213
+M2310 A539
+M2311 A1040
+M2312 A908
+M2314 A592
+M2315 A769
+M2316 A1047
+M2317 A1304
+M2319 A826
+M2321 A743
+M2323 A677
+M2324 A1237
+M2326 A947
+M2329 A826
+M2330 A94
+M2331 A242
+M2333 A178
+M2334 A1006
+M2337 A165
+M2343 A318
+M2344 A2
+M2346 A640
+M2352 A1141
+M2354 A519
+M2355 A699
+M2356 A31
+M2357 A411
+M2358 A197
+M2359 A12
+M2360 A35
+M2361 A1149
+M2363 A1365
+M2364 A153
+M2367 A827
+M2368 A366
+M2372 A94
+M2375 A419
+M2377 A426
+M2378 A482
+M2380 A695
+M2381 A1110
+M2383 A1345
+M2386 A903
+M2387 A310
+M2389 A742
+M2391 A529
+M2392 A956
+M2393 A224
+M2395 A718
+M2396 A425
+M2397 A1111
+M2398 A423
+M2399 A903
+M2400 A703
+M2401 A692
+M2402 A1007
+M2403 A29
+M2404 A13
+M2405 A743
+M2407 A1021
+M2408 A346
+M2409 A531
+M2411 A795
+M2413 A903
+M2414 A562
+M2415 A1259
+M2418 A1179
+M2420 A1089
+M2421 A546
+M2424 A690
+M2425 A577
+M2428 A548
+M2429 A910
+M2430 A881
+M2431 A295
+M2432 A1122
+M2433 A885
+M2434 A991
+M2435 A156
+M2436 A1254
+M2437 A1345
+M2439 A1073
+M2442 A1153
+M2443 A1354
+M2446 A727
+M2449 A742
+M2450 A698
+M2452 A713
+M2454 A685
+M2455 A993
+M2456 A762
+M2457 A881
+M2463 A593
+M2465 A1079
+M2469 A700
+M2470 A303
+M2473 A1059
+M2474 A577
+M2475 A562
+M2476 A527
+M2481 A1075
+M2482 A624
+M2485 A1176
+M2486 A66
+M2488 A628
+M2489 A227
+M2490 A561
+M2492 A1089
+M2493 A19
+M2495 A1317
+M2496 A1117
+M2497 A86
+M2498 A1012
+M2499 A340
+M2501 A196
+M2502 A12
+M2503 A158
+M2508 A603
+M2509 A571
+M2510 A1374
+M2513 A961
+M2516 A13
+M2517 A1310
+M2520 A1024
+M2521 A20
+M2523 A754
+M2527 A937
+M2531 A876
+M2532 A1040
+M2533 A754
+M2534 A1327
+M2535 A996
+M2536 A243
+M2538 A1008
+M2539 A591
+M2542 A1352
+M2543 A1092
+M2544 A1024
+M2546 A70
+M2547 A245
+M2548 A379
+M2553 A533
+M2554 A648
+M2556 A415
+M2560 A974
+M2562 A1157
+M2563 A1245
+M2565 A698
+M2567 A79
+M2568 A903
+M2570 A847
+M2571 A1367
+M2572 A1179
+M2577 A881
+M2578 A987
+M2579 A775
+M2582 A317
+M2583 A929
+M2584 A861
+M2589 A1206
+M2590 A1046
+M2592 A1091
+M2599 A1239
+M2600 A1052
+M2601 A876
+M2602 A328
+M2603 A1048
+M2604 A1050
+M2606 A1235
+M2607 A51
+M2608 A772
+M2611 A1340
+M2613 A9
+M2614 A856
+M2615 A1256
+M2616 A1155
+M2617 A1018
+M2619 A1089
+M2620 A1123
+M2622 A900
+M2623 A1339
+M2624 A636
+M2625 A826
+M2626 A1073
+M2627 A399
+M2628 A1227
+M2629 A577
+M2630 A577
+M2632 A36
+M2634 A1355
+M2636 A577
+M2638 A641
+M2639 A562
+M2640 A1024
+M2641 A214
+M2642 A58
+M2644 A1182
+M2645 A1186
+M2646 A109
+M2647 A807
+M2648 A800
+M2650 A419
+M2651 A964
+M2652 A1179
+M2653 A383
+M2655 A281
+M2658 A1312
+M2659 A400
+M2660 A677
+M2664 A974
+M2666 A808
+M2667 A961
+M2668 A477
+M2669 A549
+M2672 A1129
+M2675 A833
+M2677 A383
+M2678 A311
+M2679 A254
+M2680 A317
+M2681 A240
+M2682 A178
+M2683 A831
+M2687 A775
+M2689 A243
+M2691 A1373
+M2692 A222
+M2694 A1353
+M2695 A881
+M2698 A1339
+M2700 A577
+M2703 A831
+M2704 A411
+M2705 A927
+M2706 A133
+M2707 A910
+M2709 A51
+M2710 A117
+M2714 A562
+M2716 A351
+M2717 A186
+M2719 A1307
+M2721 A1040
+M2723 A1043
+M2725 A808
+M2726 A359
+M2728 A882
+M2731 A1043
+M2733 A506
+M2736 A1345
+M2738 A655
+M2741 A539
+M2742 A161
+M2746 A396
+M2755 A1048
+M2757 A219
+M2760 A977
+M2761 A851
+M2763 A1222
+M2764 A246
+M2768 A1050
+M2771 A93
+M2772 A363
+M2773 A368
+M2776 A857
+M2777 A1127
+M2778 A841
+M2779 A492
+M2780 A29
+M2781 A714
+M2784 A1303
+M2785 A920
+M2787 A1155
+M2788 A1319
+M2789 A99
+M2790 A422
+M2793 A778
+M2794 A577
+M2795 A400
+M2796 A1345
+M2798 A1227
+M2800 A412
+M2803 A156
+M2804 A844
+M2805 A250
+M2806 A502
+M2808 A499
+M2809 A51
+M2811 A1202
+M2813 A120
+M2815 A1081
+M2816 A351
+M2818 A399
+M2819 A641
+M2820 A773
+M2821 A1004
+M2822 A546
+M2823 A482
+M2825 A156
+M2826 A1281
+M2827 A118
+M2829 A882
+M2831 A974
+M2832 A591
+M2833 A1193
+M2835 A1039
+M2836 A70
+M2837 A577
+M2838 A386
+M2840 A645
+M2841 A645
+M2842 A1190
+M2844 A1175
+M2845 A338
+M2848 A653
+M2850 A949
+M2851 A1146
+M2852 A889
+M2856 A533
+M2857 A519
+M2859 A1116
+M2860 A648
+M2862 A762
+M2864 A698
+M2865 A1272
+M2867 A256
+M2868 A889
+M2869 A1117
+M2876 A1345
+M2877 A692
+M2878 A531
+M2880 A419
+M2882 A1039
+M2883 A418
+M2886 A519
+M2887 A1141
+M2888 A318
+M2891 A1265
+M2895 A270
+M2896 A1179
+M2898 A178
+M2900 A1235
+M2901 A281
+M2902 A1381
+M2905 A383
+M2907 A257
+M2908 A1111
+M2910 A803
+M2912 A1235
+M2915 A958
+M2916 A44
+M2917 A870
+M2918 A1245
+M2920 A688
+M2921 A368
+M2922 A386
+M2923 A425
+M2926 A566
+M2928 A349
+M2929 A245
+M2930 A577
+M2931 A1252
+M2932 A928
+M2933 A390
+M2934 A1184
+M2937 A1357
+M2938 A591
+M2939 A453
+M2942 A80
+M2944 A1125
+M2945 A1382
+M2948 A351
+M2949 A861
+M2951 A59
+M2953 A208
+M2954 A857
+M2955 A1139
+M2957 A162
+M2959 A1033
+M2961 A773
+M2962 A242
+M2963 A816
+M2965 A1110
+M2966 A1064
+M2967 A243
+M2968 A661
+M2969 A974
+M2970 A957
+M2971 A546
+M2974 A850
+M2979 A775
+M2980 A1371
+M2981 A717
+M2983 A1187
+M2984 A1265
+M2985 A117
+M2986 A5
+M2989 A935
+M2994 A305
+M2995 A954
+M2997 A398
+M2998 A1263
+M3000 A899
+M3001 A93
+M3002 A48
+M3003 A513
+M3004 A707
+M3007 A515
+M3008 A485
+M3010 A274
+M3013 A868
+M3016 A849
+M3017 A816
+M3019 A1321
+M3023 A1170
+M3024 A566
+M3026 A685
+M3027 A438
+M3028 A334
+M3029 A826
+M3030 A867
+M3031 A1222
+M3035 A394
+M3036 A94
+M3037 A712
+M3038 A765
+M3039 A414
+M3040 A86
+M3041 A265
+M3042 A625
+M3045 A363
+M3046 A1069
+M3048 A945
+M3050 A29
+M3053 A550
+M3054 A1117
+M3055 A419
+M3058 A742
+M3060 A794
+M3061 A971
+M3062 A661
+M3063 A411
+M3064 A1146
+M3066 A56
+M3067 A156
+M3069 A1207
+M3070 A726
+M3071 A820
+M3075 A525
+M3076 A677
+M3078 A756
+M3080 A37
+M3082 A1129
+M3086 A813
+M3089 A591
+M3091 A86
+M3092 A1048
+M3093 A762
+M3094 A965
+M3095 A1024
+M3096 A648
+M3097 A86
+M3098 A656
+M3099 A199
+M3100 A1050
+M3102 A755
+M3103 A414
+M3104 A982
+M3105 A1256
+M3106 A1127
+M3107 A581
+M3108 A1111
+M3111 A739
+M3112 A771
+M3113 A103
+M3114 A920
+M3116 A79
+M3117 A816
+M3119 A273
+M3120 A351
+M3122 A1315
+M3123 A543
+M3124 A445
+M3125 A421
+M3127 A1
+M3128 A1050
+M3130 A107
+M3132 A1318
+M3133 A1056
+M3135 A714
+M3138 A228
+M3139 A1048
+M3140 A980
+M3143 A1086
+M3144 A1179
+M3148 A546
+M3150 A550
+M3155 A214
+M3156 A482
+M3158 A785
+M3160 A1157
+M3161 A571
+M3163 A478
+M3164 A1393
+M3165 A590
+M3166 A717
+M3168 A1293
+M3170 A742
+M3173 A523
+M3175 A1092
+M3179 A1155
+M3182 A475
+M3184 A1376
+M3185 A988
+M3186 A826
+M3188 A1310
+M3192 A101
+M3193 A961
+M3196 A42
+M3200 A90
+M3201 A1043
+M3202 A1384
+M3203 A942
+M3205 A687
+M3206 A1035
+M3209 A857
+M3210 A550
+M3212 A831
+M3213 A803
+M3214 A1039
+M3215 A1123
+M3216 A993
+M3219 A723
+M3220 A1294
+M3227 A1330
+M3228 A778
+M3231 A1180
+M3233 A282
+M3240 A1303
+M3253 A1260
+M3255 A1057
+M3256 A199
+M3258 A885
+M3260 A920
+M3262 A1083
+M3264 A1294
+M3267 A697
+M3271 A1040
+M3272 A880
+M3274 A329
+M3276 A525
+M3277 A107
+M3278 A1090
+M3281 A1055
+M3282 A490
+M3284 A865
+M3285 A1317
+M3287 A649
+M3288 A342
+M3290 A1
+M3291 A340
+M3294 A1274
+M3296 A1364
+M3297 A831
+M3300 A477
+M3302 A14
+M3304 A661
+M3306 A191
+M3307 A208
+M3308 A826
+M3314 A1202
+M3317 A577
+M3320 A359
+M3331 A513
+M3332 A340
+M3337 A1131
+M3338 A524
+M3343 A1281
+M3344 A525
+M3345 A513
+M3347 A404
+M3351 A209
+M3353 A741
+M3355 A245
+M3358 A415
+M3360 A224
+M3362 A1238
+M3366 A199
+M3368 A338
+M3370 A259
+M3374 A1127
+M3377 A1321
+M3378 A1114
+M3380 A149
+M3382 A351
+M3384 A641
+M3386 A844
+M3387 A100
+M3389 A562
+M3390 A445
+M3391 A396
+M3393 A296
+M3394 A1091
+M3396 A808
+M3399 A1213
+M3400 A257
+M3401 A1046
+M3402 A937
+M3404 A506
+M3405 A51
+M3406 A523
+M3409 A555
+M3414 A1162
+M3417 A220
+M3421 A705
+M3424 A1024
+M3428 A137
+M3430 A1352
+M3434 A9
+M3435 A464
+M3438 A698
+M3439 A1068
+M3440 A28
+M3441 A40
+M3442 A1156
+M3443 A1145
+M3447 A698
+M3448 A340
+M3450 A614
+M3452 A961
+M3453 A305
+M3457 A689
+M3459 A1127
+M3461 A698
+M3463 A698
+M3465 A1055
+M3466 A197
+M3467 A369
+M3468 A338
+M3469 A1110
+M3470 A349
+M3472 A76
+M3473 A21
+M3475 A631
+M3476 A219
+M3477 A1040
+M3478 A831
+M3479 A855
+M3480 A1141
+M3483 A962
+M3484 A739
+M3485 A715
+M3488 A884
+M3489 A737
+M3491 A1351
+M3492 A1272
+M3493 A594
+M3494 A737
+M3496 A1036
+M3498 A64
+M3499 A677
+M3503 A355
+M3505 A1021
+M3508 A243
+M3510 A525
+M3511 A657
+M3512 A595
+M3514 A1381
+M3515 A617
+M3516 A1147
+M3519 A1050
+M3520 A1043
+M3527 A202
+M3530 A590
+M3531 A1118
+M3535 A844
+M3537 A1032
+M3540 A977
+M3542 A1141
+M3543 A334
+M3553 A438
+M3554 A400
+M3559 A417
+M3560 A914
+M3561 A1379
+M3562 A505
+M3566 A539
+M3567 A713
+M3569 A329
+M3570 A964
+M3574 A726
+M3575 A66
+M3577 A1115
+M3578 A612
+M3580 A1092
+M3582 A742
+M3586 A709
+M3587 A1213
+M3589 A1048
+M3591 A468
+M3592 A1235
+M3594 A1117
+M3595 A296
+M3596 A97
+M3597 A1371
+M3598 A1083
+M3599 A544
+M3602 A971
+M3603 A246
+M3605 A142
+M3607 A1047
+M3609 A550
+M3611 A1081
+M3618 A762
+M3619 A1213
+M3621 A274
+M3627 A617
+M3629 A327
+M3630 A952
+M3631 A217
+M3633 A614
+M3637 A1376
+M3638 A887
+M3640 A604
+M3641 A714
+M3647 A687
+M3648 A1290
+M3649 A814
+M3654 A383
+M3662 A716
+M3663 A1178
+M3664 A203
+M3668 A1256
+M3669 A406
+M3671 A621
+M3673 A977
+M3674 A974
+M3675 A156
+M3679 A1198
+M3680 A461
+M3681 A414
+M3688 A1350
+M3689 A506
+M3696 A724
+M3697 A1012
+M3699 A867
+M3704 A785
+M3709 A742
+M3711 A573
+M3713 A808
+M3714 A550
+M3716 A432
+M3718 A1092
+M3719 A1044
+M3726 A588
+M3727 A1129
+M3729 A1146
+M3730 A1086
+M3731 A199
+M3732 A228
+M3733 A885
+M3737 A13
+M3743 A550
+M3744 A1083
+M3746 A319
+M3753 A483
+M3760 A439
+M3762 A566
+M3765 A987
+M3767 A1393
+M3768 A1235
+M3769 A17
+M3771 A993
+M3773 A996
+M3775 A156
+M3776 A617
+M3778 A933
+M3779 A826
+M3781 A317
+M3782 A513
+M3783 A876
+M3784 A1078
+M3788 A250
+M3795 A1092
+M3803 A1053
+M3804 A1075
+M3805 A1245
+M3811 A1185
+M3812 A903
+M3817 A742
+M3821 A935
+M3823 A1068
+M3824 A208
+M3825 A659
+M3830 A622
+M3831 A383
+M3841 A319
+M3842 A598
+M3843 A795
+M3844 A117
+M3845 A1249
+M3846 A831
+M3848 A517
+M3849 A44
+M3852 A817
+M3853 A385
+M3854 A641
+M3857 A304
+M3859 A644
+M3869 A1088
+M3873 A783
+M3874 A947
+M3876 A38
+M3879 A1054
+M3880 A334
+M3882 A254
+M3883 A172
+M3884 A1213
+M3886 A627
+M3888 A974
+M3889 A881
+M3891 A826
+M3892 A389
+M3894 A903
+M3896 A612
+M3897 A65
+M3898 A127
+M3907 A974
+M3908 A1046
+M3910 A648
+M3911 A706
+M3912 A787
+M3913 A591
+M3915 A965
+M3916 A611
+M3918 A991
+M3919 A959
+M3920 A847
+M3921 A1382
+M3922 A1183
+M3924 A471
+M3925 A1339
+M3927 A150
+M3928 A83
+M3936 A375
+M3938 A903
+M3940 A250
+M3942 A895
+M3944 A620
+M3945 A1162
+M3947 A964
+M3952 A1127
+M3956 A485
+M3961 A1065
+M3967 A9
+M3968 A439
+M3969 A935
+M3972 A622
+M3980 A28
+M3981 A419
+M3984 A1111
+M3987 A1022
+M3988 A445
+M3991 A615
+M3994 A1336
+M3999 A691
+M4001 A700
+M4003 A365
+M4009 A886
+M4011 A1355
+M4019 A348
+M4023 A265
+M4025 A1235
+M4029 A657
+M4032 A910
+M4034 A707
+M4035 A553
+M4036 A1286
+M4041 A875
+M4044 A567
+M4046 A62
+M4050 A375
+M4051 A1007
+M4053 A604
+M4054 A649
+M4055 A954
+M4057 A64
+M4058 A356
+M4060 A951
+M4063 A760
+M4066 A952
+M4067 A1291
+M4068 A462
+M4072 A1064
+M4074 A312
+M4077 A261
+M4078 A1004
+M4085 A391
+M4086 A422
+M4091 A355
+M4092 A284
+M4096 A176
+M4097 A550
+M4098 A820
+M4099 A70
+M4100 A166
+M4101 A858
+M4102 A831
+M4103 A944
+M4104 A742
+M4105 A644
+M4110 A960
+M4112 A166
+M4118 A367
+M4120 A972
+M4124 A66
+M4128 A742
+M4132 A517
+M4135 A875
+M4137 A342
+M4141 A783
+M4145 A512
+M4151 A686
+M4158 A265
+M4159 A1158
+M4161 A101
+M4162 A830
+M4165 A875
+M4167 A24
+M4168 A265
+M4171 A922
+M4173 A742
+M4180 A601
+M4182 A1186
+M4183 A224
+M4188 A1127
+M4189 A553
+M4195 A90
+M4198 A1129
+M4202 A571
+M4208 A718
+M4212 A1007
+M4216 A1115
+M4220 A1
+M4222 A107
+M4223 A695
+M4226 A870
+M4230 A37
+M4231 A741
+M4239 A83
+M4247 A439
+M4250 A881
+M4252 A881
+M4254 A138
+M4256 A573
+M4259 A120
+M4265 A490
+M4266 A694
+M4268 A1110
+M4269 A655
+M4274 A351
+M4277 A698
+M4278 A1047
+M4279 A246
+M4282 A685
+M4286 A1194
+M4290 A993
+M4293 A1298
+M4295 A775
+M4302 A506
+M4303 A286
+M4308 A589
+M4313 A683
+M4315 A539
+M4316 A1008
+M4324 A881
+M4330 A641
+M4331 A342
+M4338 A649
+M4344 A139
+M4345 A801
+M4346 A1117
+M4350 A1018
+M4352 A742
+M4357 A795
+M4358 A119
+M4359 A1357
+M4362 A1286
+M4366 A855
+M4369 A977
+M4375 A699
+M4379 A422
+M4380 A1286
+M4381 A1111
+M4384 A919
+M4388 A787
+M4389 A200
+M4396 A566
+M4401 A147
+M4404 A1182
+M4405 A861
+M4408 A515
+M4411 A993
+M4412 A661
+M4413 A724
+M4419 A770
+M4421 A705
+M4423 A437
+M4432 A602
+M4437 A737
+M4446 A1050
+M4458 A816
+M4461 A332
+M4462 A1345
+M4466 A245
+M4472 A1143
+M4474 A1139
+M4481 A1286
+M4483 A255
+M4484 A1359
+M4485 A1214
+M4486 A1366
+M4489 A1127
+M4491 A1131
+M4493 A245
+M4494 A987
+M4496 A855
+M4498 A977
+M4511 A1298
+M4516 A919
+M4530 A363
+M4532 A1089
+M4533 A594
+M4535 A577
+M4541 A848
+M4543 A383
+M4551 A955
+M4556 A831
+M4557 A698
+M4563 A1040
+M4564 A672
+M4565 A1111
+M4568 A789
+M4569 A1072
+M4573 A265
+M4574 A746
+M4575 A588
+M4577 A1008
+M4584 A1265
+M4589 A46
+M4594 A514
+M4596 A871
+M4599 A1310
+M4606 A1333
+M4614 A581
+M4623 A61
+M4631 A861
+M4636 A464
+M4638 A119
+M4643 A659
+M4649 A36
+M4650 A338
+M4654 A273
+M4657 A944
+M4661 A1071
+M4667 A164
+M4674 A566
+M4677 A584
+M4679 A715
+M4680 A514
+M4681 A70
+M4685 A731
+M4690 A418
+M4691 A1087
+M4694 A1043
+M4695 A815
+M4710 A1218
+M4730 A418
+M4734 A793
+M4738 A793
+M4741 A463
+M4744 A672
+M4745 A1122
+M4752 A274
+M4761 A94
+M4762 A993
+M4765 A286
+M4773 A1014
+M4779 A999
+M4781 A604
+M4791 A260
+M4792 A1033
+M4795 A141
+M4804 A1093
+M4813 A365
+M4821 A1089
+M4822 A837
+M4830 A284
+M4839 A105
+M4842 A81
+M4852 A351
+M4853 A455
+M4857 A10
+M4860 A485
+M4866 A489
+M4868 A1191
+M4882 A117
+M4888 A713
+M4894 A573
+M4896 A1393
+M4897 A977
+M4905 A349
+M4909 A166
+M4923 A811
+M4927 A695
+M4936 A1090
+M4941 A649
+M4942 A1381
+M4949 A751
+M4950 A751
+M4955 A385
+M4960 A477
+M4971 A807
+M4973 A296
+M4979 A589
+M4997 A554
+M5007 A710
+M5008 A1193
+M5012 A211
+M5014 A735
+M5021 A840
+M5026 A345
+M0 A752
+M1 A263
+M2 A1105
+M3 A826
+M5 A614
+M6 A910
+M8 A1040
+M9 A1073
+M10 A771
+M11 A119
+M12 A60
+M13 A263
+M14 A639
+M15 A517
+M17 A1040
+M18 A1049
+M19 A1036
+M20 A140
+M21 A359
+M22 A1129
+M23 A140
+M24 A1125
+M25 A882
+M26 A156
+M27 A974
+M28 A329
+M29 A1310
+M30 A235
+M31 A910
+M32 A1393
+M33 A996
+M34 A548
+M36 A433
+M37 A1038
+M38 A1157
+M39 A359
+M40 A1122
+M41 A882
+M42 A15
+M44 A851
+M45 A1024
+M46 A178
+M47 A418
+M48 A993
+M49 A832
+M50 A1210
+M51 A868
+M52 A958
+M54 A947
+M58 A483
+M59 A545
+M60 A1216
+M61 A51
+M62 A1157
+M63 A329
+M65 A539
+M66 A389
+M67 A352
+M68 A560
+M69 A340
+M70 A962
+M71 A1382
+M72 A691
+M73 A241
+M74 A9
+M77 A546
+M78 A1269
+M79 A1048
+M80 A974
+M83 A1310
+M86 A1035
+M88 A780
+M92 A782
+M93 A422
+M94 A579
+M95 A1373
+M96 A243
+M97 A713
+M99 A140
+M100 A1373
+M101 A448
+M102 A539
+M103 A831
+M104 A788
+M105 A1066
+M106 A996
+M107 A141
+M108 A432
+M111 A234
+M112 A542
+M113 A133
+M114 A1073
+M115 A1073
+M116 A242
+M117 A657
+M118 A962
+M119 A53
+M120 A419
+M121 A445
+M122 A255
+M123 A1117
+M127 A1039
+M128 A242
+M129 A1188
+M130 A1039
+M134 A1018
+M135 A317
+M137 A329
+M141 A1050
+M142 A381
+M143 A736
+M144 A156
+M145 A633
+M147 A263
+M148 A1326
+M149 A272
+M151 A1345
+M152 A861
+M153 A808
+M155 A317
+M157 A306
+M158 A807
+M159 A1302
+M160 A425
+M161 A910
+M162 A318
+M164 A572
+M165 A191
+M166 A252
+M168 A36
+M169 A560
+M171 A28
+M174 A885
+M175 A178
+M176 A131
+M178 A860
+M179 A713
+M181 A947
+M183 A1392
+M184 A585
+M185 A783
+M186 A490
+M187 A885
+M188 A401
+M189 A768
+M190 A431
+M192 A190
+M193 A2
+M194 A377
+M195 A70
+M197 A154
+M200 A255
+M202 A205
+M203 A340
+M205 A263
+M207 A831
+M210 A1028
+M211 A1373
+M212 A794
+M213 A448
+M216 A880
+M217 A865
+M218 A1173
+M219 A546
+M220 A279
+M221 A404
+M222 A466
+M223 A757
+M226 A1007
+M227 A242
+M228 A1329
+M229 A1348
+M230 A82
+M231 A1281
+M234 A603
+M235 A1050
+M236 A962
+M237 A962
+M238 A53
+M240 A419
+M241 A562
+M245 A1029
+M248 A773
+M249 A1144
+M250 A255
+M251 A466
+M252 A272
+M253 A709
+M254 A246
+M255 A133
+M256 A498
+M257 A482
+M259 A1111
+M264 A498
+M266 A951
+M267 A15
+M268 A597
+M270 A263
+M271 A208
+M272 A445
+M274 A993
+M275 A263
+M276 A1234
+M277 A1196
+M278 A1331
+M280 A1345
+M281 A1377
+M282 A133
+M283 A1290
+M284 A965
+M285 A220
+M286 A144
+M287 A706
+M288 A568
+M289 A826
+M290 A461
+M291 A950
+M292 A1221
+M293 A747
+M294 A29
+M295 A622
+M296 A478
+M297 A1097
+M298 A509
+M299 A363
+M300 A553
+M303 A633
+M305 A133
+M306 A296
+M307 A318
+M308 A550
+M311 A81
+M312 A1150
+M314 A754
+M315 A1326
+M316 A1162
+M318 A1392
+M321 A902
+M324 A846
+M325 A691
+M326 A419
+M329 A1131
+M330 A176
+M332 A245
+M334 A57
+M335 A54
+M336 A1322
+M337 A721
+M338 A1010
+M339 A859
+M340 A263
+M341 A318
+M342 A621
+M344 A469
+M345 A1047
+M347 A981
+M348 A603
+M349 A703
+M350 A90
+M351 A861
+M352 A602
+M353 A53
+M354 A861
+M356 A1192
+M357 A532
+M359 A609
+M360 A1272
+M361 A351
+M363 A208
+M364 A567
+M366 A794
+M369 A1107
+M370 A1111
+M372 A982
+M373 A1129
+M374 A247
+M375 A107
+M376 A933
+M378 A232
+M379 A133
+M380 A9
+M382 A49
+M383 A1290
+M384 A315
+M385 A284
+M386 A314
+M387 A178
+M389 A1144
+M390 A690
+M391 A934
+M392 A1234
+M393 A705
+M394 A1254
+M395 A1373
+M397 A1054
+M398 A1153
+M399 A70
+M400 A1007
+M401 A441
+M402 A982
+M405 A1117
+M406 A141
+M407 A570
+M408 A754
+M409 A1027
+M410 A43
+M411 A351
+M412 A1065
+M413 A498
+M415 A861
+M416 A943
+M417 A690
+M420 A178
+M423 A1234
+M424 A477
+M425 A221
+M427 A718
+M428 A333
+M430 A1327
+M431 A734
+M432 A617
+M434 A641
+M435 A313
+M436 A590
+M437 A363
+M438 A795
+M439 A490
+M441 A984
+M442 A590
+M443 A761
+M444 A246
+M445 A1393
+M446 A808
+M448 A98
+M449 A982
+M450 A36
+M451 A1393
+M453 A935
+M455 A1177
+M456 A1319
+M457 A117
+M458 A748
+M460 A412
+M461 A1
+M462 A1039
+M463 A1157
+M465 A701
+M467 A1349
+M468 A1350
+M470 A441
+M471 A314
+M474 A533
+M475 A1129
+M476 A216
+M478 A982
+M482 A1278
+M483 A961
+M484 A347
+M486 A1367
+M487 A389
+M488 A1326
+M490 A1317
+M491 A1033
+M493 A250
+M498 A908
+M500 A288
+M501 A890
+M503 A1087
+M505 A684
+M508 A65
+M509 A1234
+M510 A36
+M512 A808
+M513 A250
+M514 A982
+M515 A981
+M516 A910
+M517 A825
+M520 A1230
+M521 A17
+M522 A709
+M523 A759
+M524 A1135
+M525 A1165
+M526 A576
+M527 A9
+M528 A66
+M534 A1064
+M535 A982
+M536 A538
+M538 A480
+M539 A880
+M540 A224
+M542 A624
+M544 A337
+M545 A699
+M546 A107
+M547 A768
+M548 A1210
+M549 A429
+M550 A260
+M551 A1219
+M556 A937
+M557 A993
+M558 A582
+M560 A1224
+M561 A1263
+M562 A245
+M564 A1281
+M565 A79
+M568 A947
+M569 A1280
+M570 A1008
+M571 A533
+M572 A1021
+M574 A872
+M576 A1327
+M577 A706
+M579 A1310
+M580 A936
+M581 A1002
+M583 A732
+M584 A517
+M587 A352
+M588 A178
+M589 A246
+M590 A754
+M591 A1168
+M592 A863
+M594 A253
+M595 A238
+M596 A272
+M598 A603
+M599 A9
+M600 A378
+M602 A549
+M603 A680
+M604 A856
+M605 A351
+M607 A65
+M609 A245
+M610 A1383
+M611 A690
+M612 A21
+M613 A606
+M618 A981
+M620 A520
+M624 A284
+M627 A355
+M628 A997
+M629 A979
+M631 A885
+M632 A119
+M633 A17
+M635 A9
+M639 A602
+M640 A278
+M641 A169
+M642 A168
+M643 A28
+M644 A1353
+M646 A1080
+M648 A1373
+M650 A384
+M651 A1010
+M652 A542
+M653 A1327
+M655 A1321
+M657 A749
+M658 A214
+M660 A482
+M662 A1280
+M663 A774
+M664 A932
+M665 A16
+M666 A836
+M667 A884
+M668 A1068
+M669 A762
+M670 A802
+M671 A410
+M672 A216
+M673 A609
+M674 A822
+M675 A987
+M678 A199
+M679 A1230
+M681 A528
+M682 A490
+M683 A301
+M686 A622
+M687 A649
+M690 A1046
+M691 A245
+M692 A795
+M693 A294
+M696 A242
+M697 A1291
+M698 A1339
+M701 A415
+M702 A882
+M703 A836
+M706 A117
+M707 A423
+M708 A1137
+M709 A539
+M710 A747
+M711 A889
+M712 A298
+M714 A483
+M715 A782
+M717 A271
+M718 A479
+M719 A620
+M721 A384
+M722 A1133
+M723 A1137
+M724 A242
+M725 A857
+M726 A881
+M728 A935
+M730 A701
+M731 A246
+M732 A961
+M733 A9
+M734 A546
+M735 A795
+M736 A445
+M737 A1305
+M738 A1068
+M739 A935
+M740 A313
+M743 A1361
+M745 A1254
+M746 A94
+M747 A752
+M748 A1319
+M749 A537
+M750 A967
+M751 A750
+M753 A245
+M754 A572
+M755 A327
+M760 A258
+M761 A577
+M763 A471
+M764 A1197
+M766 A513
+M767 A29
+M768 A324
+M769 A908
+M771 A284
+M772 A9
+M773 A692
+M774 A1111
+M775 A488
+M778 A443
+M781 A947
+M783 A1098
+M784 A197
+M785 A1321
+M786 A1010
+M787 A1009
+M788 A588
+M789 A1305
+M793 A1156
+M794 A1040
+M796 A1287
+M798 A352
+M801 A760
+M803 A1332
+M804 A593
+M805 A1115
+M806 A234
+M807 A722
+M808 A1016
+M809 A1185
+M814 A1109
+M815 A966
+M817 A253
+M818 A753
+M821 A982
+M822 A124
+M823 A516
+M824 A199
+M827 A197
+M828 A833
+M829 A1390
+M830 A961
+M831 A11
+M834 A1329
+M835 A892
+M836 A1096
+M838 A747
+M840 A1395
+M841 A622
+M842 A313
+M843 A278
+M844 A1204
+M845 A362
+M846 A910
+M848 A1018
+M849 A40
+M850 A250
+M851 A249
+M853 A847
+M854 A190
+M855 A865
+M859 A1190
+M860 A714
+M862 A572
+M864 A1369
+M865 A796
+M866 A569
+M868 A681
+M869 A484
+M871 A879
+M873 A1215
+M876 A552
+M877 A928
+M878 A9
+M879 A249
+M880 A677
+M881 A1281
+M882 A1293
+M886 A531
+M887 A73
+M890 A786
+M891 A777
+M892 A882
+M893 A861
+M895 A778
+M896 A762
+M898 A1197
+M900 A808
+M901 A464
+M902 A548
+M903 A1257
+M904 A1185
+M905 A856
+M906 A780
+M907 A533
+M908 A861
+M909 A1351
+M910 A935
+M911 A224
+M912 A1317
+M913 A1371
+M914 A714
+M915 A1120
+M917 A990
+M918 A1122
+M920 A256
+M923 A239
+M924 A315
+M926 A885
+M929 A9
+M930 A772
+M931 A547
+M933 A831
+M934 A29
+M935 A982
+M936 A721
+M937 A594
+M938 A593
+M939 A529
+M940 A1050
+M941 A588
+M942 A882
+M943 A281
+M944 A718
+M945 A714
+M946 A1343
+M947 A656
+M948 A271
+M949 A351
+M950 A402
+M951 A550
+M953 A562
+M955 A732
+M956 A964
+M957 A9
+M958 A625
+M960 A794
+M964 A730
+M965 A1370
+M966 A1132
+M969 A210
+M970 A1347
+M972 A566
+M973 A296
+M974 A1199
+M975 A1316
+M977 A1393
+M978 A236
+M979 A181
+M981 A1283
+M982 A316
+M983 A935
+M984 A569
+M987 A963
+M989 A923
+M991 A846
+M993 A772
+M996 A1310
+M997 A608
+M998 A511
+M999 A363
+M1001 A1070
+M1002 A1209
+M1003 A1229
+M1004 A661
+M1005 A1219
+M1006 A1041
+M1008 A1229
+M1009 A661
+M1011 A156
+M1012 A527
+M1013 A1064
+M1014 A1043
+M1015 A674
+M1017 A588
+M1018 A201
+M1019 A747
+M1020 A610
+M1021 A935
+M1022 A147
+M1024 A648
+M1025 A643
+M1026 A441
+M1028 A984
+M1029 A29
+M1031 A1335
+M1032 A308
+M1034 A19
+M1037 A474
+M1038 A849
+M1039 A530
+M1041 A1348
+M1042 A133
+M1043 A183
+M1045 A280
+M1048 A340
+M1049 A987
+M1050 A853
+M1051 A506
+M1052 A441
+M1053 A279
+M1054 A1315
+M1055 A937
+M1057 A1377
+M1058 A1084
+M1059 A578
+M1060 A450
+M1065 A560
+M1066 A1215
+M1067 A400
+M1069 A351
+M1070 A393
+M1072 A543
+M1073 A1137
+M1074 A731
+M1077 A691
+M1078 A936
+M1079 A246
+M1080 A1167
+M1081 A342
+M1083 A1226
+M1084 A641
+M1085 A821
+M1086 A852
+M1087 A825
+M1089 A799
+M1090 A898
+M1092 A770
+M1094 A419
+M1095 A265
+M1096 A448
+M1097 A174
+M1098 A732
+M1100 A29
+M1103 A1303
+M1104 A982
+M1105 A857
+M1106 A148
+M1108 A341
+M1111 A714
+M1112 A957
+M1114 A156
+M1115 A980
+M1116 A1352
+M1118 A1152
+M1122 A333
+M1123 A1093
+M1124 A37
+M1125 A506
+M1126 A1000
+M1127 A448
+M1128 A1318
+M1131 A937
+M1134 A543
+M1135 A1323
+M1137 A193
+M1138 A789
+M1139 A216
+M1140 A178
+M1142 A1199
+M1144 A194
+M1145 A1263
+M1146 A11
+M1148 A250
+M1149 A19
+M1151 A1110
+M1153 A807
+M1156 A942
+M1157 A108
+M1158 A507
+M1161 A1276
+M1162 A482
+M1163 A65
+M1164 A907
+M1165 A1246
+M1166 A68
+M1167 A308
+M1168 A242
+M1171 A78
+M1172 A818
+M1173 A583
+M1176 A1160
+M1177 A788
+M1179 A754
+M1181 A1163
+M1183 A164
+M1185 A826
+M1186 A373
+M1189 A9
+M1190 A731
+M1191 A1382
+M1192 A550
+M1193 A277
+M1194 A1375
+M1196 A853
+M1199 A1319
+M1203 A808
+M1204 A584
+M1205 A517
+M1206 A351
+M1207 A889
+M1208 A993
+M1209 A243
+M1210 A1270
+M1211 A599
+M1212 A914
+M1213 A218
+M1214 A1392
+M1215 A1393
+M1216 A934
+M1218 A1022
+M1219 A178
+M1220 A774
+M1221 A43
+M1223 A453
+M1224 A1003
+M1225 A1333
+M1227 A587
+M1228 A1314
+M1230 A674
+M1232 A90
+M1233 A1054
+M1234 A478
+M1237 A1055
+M1238 A657
+M1240 A177
+M1243 A1072
+M1246 A655
+M1247 A177
+M1248 A9
+M1250 A1241
+M1253 A592
+M1256 A354
+M1257 A780
+M1258 A480
+M1259 A583
+M1263 A1163
+M1267 A242
+M1268 A1181
+M1271 A739
+M1272 A1046
+M1276 A15
+M1278 A974
+M1281 A151
+M1282 A420
+M1285 A910
+M1286 A1234
+M1287 A1068
+M1288 A968
+M1289 A1067
+M1290 A291
+M1291 A313
+M1292 A942
+M1293 A611
+M1295 A714
+M1296 A327
+M1297 A971
+M1301 A1154
+M1303 A153
+M1304 A1382
+M1305 A11
+M1306 A1192
+M1307 A734
+M1308 A897
+M1310 A546
+M1311 A1146
+M1312 A521
+M1314 A913
+M1317 A1003
+M1318 A1293
+M1319 A115
+M1320 A948
+M1322 A14
+M1325 A832
+M1326 A651
+M1328 A233
+M1331 A1114
+M1332 A1373
+M1333 A1339
+M1337 A363
+M1338 A78
+M1339 A584
+M1340 A1371
+M1342 A741
+M1345 A478
+M1348 A1174
+M1349 A1161
+M1351 A1236
+M1353 A1042
+M1354 A90
+M1355 A437
+M1356 A47
+M1357 A570
+M1360 A1191
+M1363 A519
+M1366 A476
+M1369 A1350
+M1370 A546
+M1371 A178
+M1373 A1122
+M1374 A126
+M1375 A828
+M1378 A1120
+M1379 A1346
+M1380 A149
+M1383 A579
+M1385 A1267
+M1387 A674
+M1388 A1161
+M1389 A123
+M1390 A561
+M1393 A75
+M1394 A1234
+M1395 A481
+M1396 A847
+M1397 A227
+M1399 A1380
+M1400 A1190
+M1401 A279
+M1402 A347
+M1403 A129
+M1405 A732
+M1406 A1243
+M1407 A340
+M1408 A243
+M1409 A1244
+M1412 A81
+M1413 A583
+M1415 A795
+M1417 A772
+M1418 A444
+M1420 A724
+M1421 A381
+M1423 A728
+M1424 A414
+M1425 A460
+M1426 A411
+M1427 A1301
+M1428 A639
+M1429 A352
+M1431 A1007
+M1434 A260
+M1436 A197
+M1438 A812
+M1439 A224
+M1440 A838
+M1441 A1122
+M1443 A456
+M1445 A953
+M1448 A56
+M1449 A448
+M1451 A370
+M1452 A246
+M1453 A1043
+M1455 A756
+M1456 A501
+M1458 A1155
+M1460 A79
+M1461 A664
+M1462 A276
+M1463 A1107
+M1465 A1311
+M1467 A208
+M1471 A602
+M1472 A828
+M1475 A236
+M1476 A41
+M1478 A958
+M1479 A1393
+M1480 A598
+M1483 A1050
+M1486 A880
+M1487 A1282
+M1488 A1213
+M1489 A348
+M1490 A226
+M1492 A339
+M1493 A753
+M1495 A602
+M1498 A594
+M1500 A1204
+M1502 A1111
+M1503 A845
+M1505 A360
+M1510 A749
+M1513 A611
+M1514 A44
+M1516 A467
+M1518 A663
+M1519 A242
+M1521 A546
+M1523 A1377
+M1524 A1199
+M1525 A739
+M1526 A1135
+M1530 A481
+M1534 A1276
+M1535 A249
+M1537 A333
+M1539 A883
+M1540 A588
+M1541 A93
+M1542 A1244
+M1543 A400
+M1544 A597
+M1545 A809
+M1546 A910
+M1548 A242
+M1549 A617
+M1551 A1263
+M1552 A1100
+M1554 A561
+M1555 A627
+M1556 A62
+M1558 A1337
+M1559 A448
+M1564 A634
+M1567 A916
+M1568 A929
+M1569 A838
+M1571 A119
+M1574 A1232
+M1575 A671
+M1576 A993
+M1578 A1111
+M1579 A1278
+M1581 A190
+M1584 A252
+M1586 A982
+M1587 A1149
+M1588 A53
+M1589 A1254
+M1591 A691
+M1592 A1179
+M1593 A1021
+M1594 A1047
+M1595 A1010
+M1597 A276
+M1598 A80
+M1599 A1067
+M1600 A178
+M1601 A289
+M1603 A878
+M1604 A9
+M1605 A643
+M1606 A728
+M1608 A464
+M1610 A747
+M1611 A178
+M1612 A589
+M1615 A245
+M1616 A753
+M1618 A590
+M1619 A565
+M1621 A310
+M1624 A456
+M1626 A1310
+M1630 A1011
+M1632 A326
+M1633 A1309
+M1636 A1229
+M1637 A1227
+M1641 A186
+M1642 A962
+M1645 A997
+M1646 A549
+M1647 A109
+M1648 A935
+M1650 A1278
+M1654 A324
+M1655 A76
+M1657 A817
+M1660 A68
+M1661 A227
+M1662 A1310
+M1665 A228
+M1666 A32
+M1668 A388
+M1674 A1129
+M1675 A86
+M1676 A396
+M1677 A216
+M1681 A282
+M1682 A170
+M1683 A769
+M1685 A958
+M1688 A243
+M1689 A857
+M1690 A242
+M1691 A111
+M1692 A485
+M1694 A898
+M1699 A873
+M1701 A20
+M1707 A1332
+M1708 A394
+M1711 A1081
+M1713 A9
+M1714 A915
+M1715 A1294
+M1716 A377
+M1718 A1066
+M1722 A317
+M1725 A705
+M1726 A399
+M1735 A541
+M1739 A21
+M1741 A988
+M1742 A89
+M1743 A1353
+M1744 A400
+M1747 A916
+M1751 A308
+M1753 A1382
+M1754 A245
+M1755 A1042
+M1756 A826
+M1757 A357
+M1758 A620
+M1759 A760
+M1760 A1310
+M1762 A661
+M1763 A380
+M1764 A9
+M1767 A934
+M1768 A329
+M1775 A590
+M1776 A810
+M1778 A448
+M1783 A796
+M1784 A776
+M1785 A1192
+M1787 A740
+M1788 A405
+M1789 A574
+M1792 A1004
+M1796 A683
+M1798 A1303
+M1799 A497
+M1801 A641
+M1803 A523
+M1804 A1176
+M1805 A1048
+M1806 A807
+M1808 A747
+M1811 A1224
+M1812 A9
+M1813 A880
+M1814 A1156
+M1816 A1129
+M1820 A926
+M1821 A432
+M1822 A357
+M1824 A1300
+M1825 A774
+M1827 A831
+M1828 A958
+M1829 A964
+M1830 A272
+M1834 A382
+M1837 A1098
+M1838 A571
+M1839 A279
+M1840 A899
+M1841 A284
+M1846 A1154
+M1849 A147
+M1850 A1344
+M1852 A641
+M1853 A1227
+M1854 A1064
+M1856 A1053
+M1859 A79
+M1861 A1067
+M1862 A34
+M1864 A502
+M1865 A1165
+M1867 A452
+M1868 A1041
+M1870 A880
+M1872 A173
+M1873 A79
+M1874 A1299
+M1875 A851
+M1876 A1212
+M1877 A718
+M1879 A819
+M1881 A1166
+M1882 A1086
+M1885 A1010
+M1886 A564
+M1889 A934
+M1890 A1138
+M1891 A166
+M1893 A352
+M1895 A269
+M1898 A156
+M1899 A564
+M1900 A965
+M1901 A847
+M1902 A29
+M1903 A79
+M1904 A1227
+M1905 A375
+M1906 A335
+M1907 A971
+M1908 A230
+M1910 A1014
+M1914 A944
+M1915 A422
+M1918 A285
+M1920 A714
+M1924 A1128
+M1925 A49
+M1928 A1017
+M1929 A701
+M1931 A1220
+M1935 A1203
+M1937 A298
+M1938 A626
+M1939 A724
+M1941 A714
+M1945 A1141
+M1948 A37
+M1949 A1145
+M1952 A376
+M1955 A227
+M1959 A976
+M1961 A970
+M1962 A666
+M1965 A1297
+M1968 A752
+M1970 A538
+M1973 A1111
+M1974 A705
+M1975 A392
+M1976 A1289
+M1978 A965
+M1979 A419
+M1980 A960
+M1981 A349
+M1982 A588
+M1985 A1268
+M1986 A302
+M1987 A204
+M1989 A242
+M1992 A520
+M1994 A481
+M1998 A1396
+M1999 A478
+M2000 A549
+M2005 A341
+M2008 A233
+M2011 A178
+M2015 A325
+M2016 A1212
+M2017 A250
+M2018 A1244
+M2019 A915
+M2020 A1392
+M2021 A445
+M2022 A620
+M2023 A586
+M2024 A111
+M2025 A511
+M2027 A1103
+M2028 A1146
+M2031 A1148
+M2033 A944
+M2034 A740
+M2035 A571
+M2036 A335
+M2039 A344
+M2040 A785
+M2042 A506
+M2043 A579
+M2046 A777
+M2047 A747
+M2048 A1048
+M2049 A882
+M2050 A21
+M2051 A1262
+M2052 A221
+M2054 A1360
+M2055 A831
+M2056 A94
+M2057 A361
+M2062 A48
+M2065 A539
+M2070 A950
+M2071 A242
+M2074 A658
+M2079 A1134
+M2081 A807
+M2082 A724
+M2083 A1242
+M2084 A1068
+M2085 A93
+M2086 A714
+M2090 A941
+M2091 A391
+M2093 A158
+M2094 A1280
+M2095 A1300
+M2096 A674
+M2099 A11
+M2102 A247
+M2103 A1370
+M2104 A1149
+M2105 A1345
+M2107 A349
+M2108 A835
+M2109 A910
+M2110 A1248
+M2111 A884
+M2113 A1396
+M2114 A694
+M2115 A1130
+M2118 A120
+M2121 A259
+M2124 A1141
+M2125 A413
+M2126 A176
+M2127 A373
+M2128 A670
+M2129 A878
+M2131 A497
+M2133 A141
+M2134 A832
+M2136 A1063
+M2137 A577
+M2138 A880
+M2139 A391
+M2144 A87
+M2146 A1110
+M2151 A1379
+M2152 A829
+M2153 A477
+M2154 A830
+M2156 A1052
+M2158 A880
+M2160 A457
+M2161 A940
+M2163 A701
+M2164 A828
+M2165 A840
+M2166 A115
+M2167 A946
+M2168 A1046
+M2170 A234
+M2174 A193
+M2175 A587
+M2176 A1312
+M2177 A513
+M2179 A1128
+M2181 A1068
+M2184 A1111
+M2189 A1177
+M2192 A1319
+M2193 A690
+M2196 A104
+M2198 A931
+M2199 A531
+M2205 A822
+M2207 A909
+M2209 A1110
+M2210 A821
+M2212 A373
+M2214 A806
+M2217 A1251
+M2219 A701
+M2221 A1123
+M2222 A299
+M2223 A762
+M2224 A1101
+M2225 A1035
+M2227 A1346
+M2230 A311
+M2231 A1227
+M2232 A328
+M2234 A1337
+M2237 A585
+M2238 A675
+M2239 A116
+M2240 A1046
+M2243 A366
+M2244 A993
+M2245 A743
+M2246 A637
+M2249 A961
+M2250 A1129
+M2251 A946
+M2253 A1382
+M2255 A1110
+M2260 A794
+M2262 A1120
+M2263 A212
+M2265 A649
+M2267 A928
+M2268 A239
+M2269 A1309
+M2270 A1359
+M2274 A958
+M2275 A643
+M2277 A187
+M2278 A237
+M2279 A1151
+M2281 A892
+M2282 A966
+M2284 A667
+M2285 A465
+M2286 A844
+M2287 A1118
+M2289 A679
+M2290 A1199
+M2291 A50
+M2292 A832
+M2293 A772
+M2294 A674
+M2297 A907
+M2298 A13
+M2304 A1366
+M2305 A1384
+M2307 A892
+M2308 A500
+M2310 A250
+M2311 A1127
+M2312 A822
+M2315 A314
+M2316 A1124
+M2317 A873
+M2319 A389
+M2321 A261
+M2323 A445
+M2324 A139
+M2326 A621
+M2328 A1268
+M2329 A376
+M2331 A44
+M2333 A133
+M2334 A241
+M2337 A761
+M2352 A1305
+M2354 A1259
+M2355 A100
+M2357 A1371
+M2358 A542
+M2359 A724
+M2360 A1062
+M2361 A1299
+M2362 A1184
+M2363 A518
+M2364 A1359
+M2367 A1375
+M2369 A28
+M2372 A148
+M2375 A808
+M2377 A181
+M2380 A66
+M2381 A57
+M2383 A117
+M2386 A714
+M2387 A869
+M2389 A899
+M2393 A1367
+M2396 A1244
+M2397 A118
+M2398 A944
+M2399 A1365
+M2400 A554
+M2401 A386
+M2402 A239
+M2403 A63
+M2405 A247
+M2406 A739
+M2407 A335
+M2408 A592
+M2409 A423
+M2411 A651
+M2413 A889
+M2414 A1043
+M2417 A1146
+M2418 A9
+M2420 A1327
+M2422 A651
+M2425 A1098
+M2429 A1093
+M2430 A1367
+M2432 A978
+M2433 A1211
+M2434 A617
+M2435 A539
+M2437 A273
+M2439 A1089
+M2441 A88
+M2442 A1289
+M2449 A1048
+M2450 A156
+M2452 A908
+M2454 A1364
+M2456 A176
+M2457 A327
+M2463 A301
+M2465 A262
+M2470 A380
+M2474 A519
+M2475 A1175
+M2476 A285
+M2481 A279
+M2488 A414
+M2489 A442
+M2492 A1083
+M2493 A78
+M2495 A963
+M2496 A1064
+M2497 A269
+M2498 A410
+M2499 A1122
+M2501 A217
+M2502 A700
+M2508 A450
+M2509 A1224
+M2510 A1
+M2513 A701
+M2516 A906
+M2517 A546
+M2520 A643
+M2521 A941
+M2523 A794
+M2527 A914
+M2531 A1089
+M2532 A147
+M2533 A794
+M2535 A156
+M2536 A1044
+M2538 A933
+M2539 A863
+M2542 A573
+M2543 A508
+M2546 A395
+M2547 A1048
+M2548 A1355
+M2553 A1019
+M2554 A821
+M2560 A433
+M2562 A1068
+M2563 A350
+M2565 A1363
+M2568 A1365
+M2571 A1245
+M2572 A224
+M2577 A1065
+M2578 A724
+M2579 A1296
+M2582 A47
+M2584 A11
+M2589 A451
+M2590 A104
+M2592 A1154
+M2595 A798
+M2599 A845
+M2600 A1282
+M2601 A1033
+M2602 A259
+M2603 A320
+M2604 A107
+M2606 A1359
+M2607 A1376
+M2608 A1031
+M2611 A270
+M2613 A1188
+M2614 A1105
+M2615 A1315
+M2616 A1123
+M2617 A743
+M2619 A1083
+M2620 A902
+M2623 A1021
+M2624 A171
+M2625 A762
+M2626 A900
+M2628 A1004
+M2629 A714
+M2630 A762
+M2632 A1317
+M2634 A1382
+M2636 A363
+M2638 A589
+M2639 A123
+M2640 A533
+M2641 A376
+M2642 A586
+M2644 A493
+M2645 A576
+M2647 A412
+M2648 A1065
+M2650 A448
+M2651 A854
+M2652 A228
+M2653 A1315
+M2655 A779
+M2659 A963
+M2660 A181
+M2664 A1013
+M2668 A618
+M2669 A646
+M2672 A1155
+M2677 A402
+M2681 A1357
+M2682 A176
+M2689 A81
+M2691 A943
+M2692 A323
+M2695 A186
+M2698 A1027
+M2700 A506
+M2701 A380
+M2703 A340
+M2705 A1119
+M2706 A716
+M2707 A130
+M2709 A1111
+M2710 A1194
+M2714 A123
+M2716 A1327
+M2717 A971
+M2719 A183
+M2721 A558
+M2723 A326
+M2725 A1043
+M2726 A474
+M2731 A1096
+M2736 A14
+M2738 A880
+M2746 A1396
+M2755 A949
+M2757 A1317
+M2760 A9
+M2761 A1183
+M2763 A765
+M2764 A80
+M2768 A777
+M2769 A16
+M2770 A234
+M2772 A450
+M2773 A302
+M2774 A1055
+M2776 A391
+M2777 A1212
+M2780 A430
+M2781 A1213
+M2783 A927
+M2785 A1062
+M2786 A454
+M2787 A248
+M2788 A89
+M2789 A1141
+M2794 A23
+M2795 A352
+M2798 A1004
+M2800 A253
+M2802 A60
+M2803 A620
+M2805 A96
+M2806 A858
+M2809 A795
+M2813 A92
+M2816 A477
+M2818 A380
+M2819 A389
+M2820 A1029
+M2821 A136
+M2822 A660
+M2823 A976
+M2826 A570
+M2832 A1311
+M2835 A1157
+M2836 A1281
+M2837 A197
+M2838 A599
+M2840 A307
+M2841 A584
+M2844 A586
+M2845 A389
+M2848 A616
+M2850 A1285
+M2851 A368
+M2852 A1231
+M2856 A792
+M2857 A82
+M2860 A865
+M2862 A1068
+M2864 A561
+M2865 A445
+M2867 A837
+M2868 A194
+M2869 A29
+M2876 A942
+M2882 A641
+M2883 A1373
+M2886 A5
+M2887 A1379
+M2888 A549
+M2895 A1085
+M2896 A728
+M2897 A1299
+M2898 A70
+M2900 A197
+M2901 A1197
+M2902 A461
+M2905 A83
+M2907 A1396
+M2908 A1150
+M2910 A388
+M2915 A732
+M2917 A439
+M2918 A1149
+M2921 A382
+M2923 A34
+M2924 A1396
+M2926 A250
+M2928 A277
+M2929 A1199
+M2930 A737
+M2932 A766
+M2933 A921
+M2934 A1201
+M2937 A1277
+M2938 A571
+M2939 A1068
+M2942 A483
+M2944 A144
+M2945 A1305
+M2948 A968
+M2949 A382
+M2951 A1164
+M2954 A421
+M2957 A633
+M2959 A689
+M2961 A1192
+M2962 A1025
+M2963 A424
+M2965 A841
+M2966 A302
+M2967 A784
+M2968 A1200
+M2969 A51
+M2971 A603
+M2973 A682
+M2974 A640
+M2979 A523
+M2980 A1339
+M2984 A930
+M2986 A410
+M2988 A1057
+M2989 A1093
+M2994 A1032
+M2995 A78
+M2997 A237
+M2998 A767
+M3000 A167
+M3001 A52
+M3002 A1118
+M3003 A840
+M3004 A792
+M3007 A998
+M3008 A941
+M3010 A401
+M3017 A178
+M3019 A1222
+M3023 A200
+M3030 A457
+M3031 A934
+M3035 A970
+M3037 A139
+M3038 A660
+M3039 A386
+M3040 A720
+M3041 A1088
+M3045 A1181
+M3050 A920
+M3053 A218
+M3054 A1008
+M3055 A284
+M3056 A1224
+M3058 A219
+M3060 A1259
+M3061 A1119
+M3062 A643
+M3064 A1388
+M3066 A244
+M3067 A1345
+M3069 A812
+M3070 A1295
+M3071 A86
+M3076 A970
+M3078 A292
+M3080 A1255
+M3089 A351
+M3091 A720
+M3093 A1243
+M3095 A740
+M3096 A570
+M3097 A720
+M3099 A1312
+M3100 A14
+M3103 A740
+M3104 A647
+M3106 A931
+M3108 A860
+M3110 A6
+M3111 A138
+M3112 A967
+M3113 A1023
+M3114 A1303
+M3117 A178
+M3119 A1198
+M3120 A937
+M3123 A744
+M3124 A1017
+M3127 A883
+M3128 A300
+M3132 A914
+M3135 A561
+M3138 A729
+M3139 A1290
+M3140 A1004
+M3141 A137
+M3143 A282
+M3144 A146
+M3146 A1291
+M3148 A151
+M3150 A226
+M3155 A519
+M3156 A150
+M3158 A1358
+M3160 A762
+M3161 A1010
+M3163 A1165
+M3164 A170
+M3165 A741
+M3168 A913
+M3170 A133
+M3173 A1010
+M3175 A1123
+M3184 A976
+M3185 A1253
+M3193 A391
+M3198 A313
+M3200 A1006
+M3205 A243
+M3206 A910
+M3209 A717
+M3212 A677
+M3213 A790
+M3215 A543
+M3216 A142
+M3220 A360
+M3227 A702
+M3231 A625
+M3240 A1253
+M3241 A814
+M3249 A1188
+M3251 A753
+M3256 A426
+M3258 A683
+M3260 A1303
+M3267 A1088
+M3270 A705
+M3271 A1213
+M3274 A762
+M3277 A76
+M3284 A445
+M3285 A963
+M3288 A708
+M3290 A483
+M3291 A70
+M3293 A763
+M3294 A1061
+M3296 A1322
+M3297 A980
+M3300 A904
+M3304 A1189
+M3305 A1308
+M3306 A1208
+M3307 A423
+M3308 A396
+M3317 A1107
+M3331 A372
+M3332 A55
+M3337 A520
+M3338 A606
+M3343 A554
+M3344 A504
+M3345 A511
+M3353 A370
+M3355 A770
+M3366 A963
+M3368 A589
+M3370 A1327
+M3374 A1179
+M3378 A1285
+M3380 A906
+M3382 A210
+M3384 A588
+M3386 A820
+M3388 A446
+M3389 A12
+M3390 A572
+M3391 A1396
+M3392 A1369
+M3393 A313
+M3394 A1022
+M3396 A929
+M3399 A894
+M3400 A1294
+M3401 A950
+M3405 A1213
+M3414 A684
+M3417 A970
+M3424 A1122
+M3430 A413
+M3435 A1146
+M3438 A769
+M3439 A712
+M3440 A1000
+M3441 A521
+M3442 A838
+M3443 A538
+M3447 A1157
+M3448 A732
+M3450 A33
+M3452 A391
+M3453 A77
+M3457 A1185
+M3459 A574
+M3461 A910
+M3463 A645
+M3465 A668
+M3466 A119
+M3468 A1018
+M3470 A196
+M3471 A0
+M3473 A1227
+M3475 A391
+M3476 A1210
+M3477 A1108
+M3479 A1345
+M3480 A1379
+M3492 A468
+M3493 A235
+M3498 A202
+M3499 A625
+M3501 A613
+M3503 A294
+M3505 A18
+M3512 A1034
+M3514 A971
+M3515 A143
+M3516 A1096
+M3523 A128
+M3530 A863
+M3531 A486
+M3537 A689
+M3540 A511
+M3542 A593
+M3543 A1211
+M3559 A727
+M3560 A40
+M3561 A608
+M3566 A476
+M3567 A1259
+M3569 A458
+M3570 A735
+M3574 A614
+M3575 A1390
+M3578 A1385
+M3580 A1193
+M3581 A327
+M3582 A777
+M3586 A1171
+M3587 A208
+M3591 A1253
+M3592 A643
+M3594 A914
+M3597 A1339
+M3598 A1362
+M3599 A805
+M3602 A465
+M3603 A701
+M3605 A1353
+M3607 A169
+M3618 A889
+M3621 A54
+M3627 A861
+M3631 A115
+M3637 A154
+M3638 A797
+M3640 A1000
+M3641 A1226
+M3647 A1122
+M3648 A20
+M3652 A686
+M3654 A511
+M3663 A463
+M3668 A439
+M3671 A1316
+M3674 A1345
+M3675 A1213
+M3679 A222
+M3680 A1011
+M3681 A770
+M3688 A1042
+M3689 A919
+M3696 A613
+M3697 A310
+M3699 A457
+M3704 A1358
+M3709 A1128
+M3713 A52
+M3714 A645
+M3716 A730
+M3717 A1338
+M3718 A1193
+M3719 A1306
+M3722 A719
+M3726 A482
+M3727 A1263
+M3729 A1388
+M3732 A510
+M3733 A43
+M3734 A1279
+M3741 A475
+M3746 A170
+M3751 A1168
+M3753 A1303
+M3760 A763
+M3762 A1386
+M3765 A672
+M3766 A1313
+M3767 A18
+M3769 A372
+M3771 A1379
+M3773 A238
+M3775 A289
+M3778 A945
+M3779 A117
+M3780 A725
+M3781 A429
+M3782 A1283
+M3783 A1287
+M3784 A1285
+M3788 A778
+M3795 A1081
+M3811 A165
+M3814 A194
+M3817 A1163
+M3821 A183
+M3824 A291
+M3825 A341
+M3831 A795
+M3844 A1052
+M3846 A861
+M3849 A296
+M3852 A1011
+M3854 A31
+M3857 A267
+M3858 A291
+M3867 A69
+M3869 A1231
+M3873 A947
+M3879 A633
+M3880 A1211
+M3882 A472
+M3884 A942
+M3886 A869
+M3888 A1048
+M3889 A854
+M3891 A1302
+M3892 A1181
+M3894 A1041
+M3896 A1385
+M3898 A1284
+M3907 A1353
+M3908 A706
+M3910 A1303
+M3911 A1014
+M3912 A358
+M3913 A572
+M3915 A666
+M3917 A601
+M3918 A1084
+M3920 A538
+M3921 A857
+M3922 A983
+M3924 A839
+M3925 A483
+M3928 A1391
+M3929 A705
+M3936 A299
+M3938 A226
+M3940 A755
+M3942 A1008
+M3945 A1219
+M3947 A1043
+M3955 A398
+M3961 A784
+M3967 A1188
+M3969 A1084
+M3972 A615
+M3980 A552
+M3981 A90
+M3984 A753
+M3987 A920
+M3988 A522
+M3991 A897
+M3994 A344
+M3998 A629
+M3999 A1077
+M4009 A879
+M4011 A662
+M4023 A537
+M4025 A1199
+M4032 A1028
+M4035 A1221
+M4040 A667
+M4041 A1313
+M4044 A1391
+M4046 A829
+M4049 A738
+M4050 A291
+M4051 A1135
+M4053 A1383
+M4057 A922
+M4063 A14
+M4067 A1249
+M4068 A689
+M4069 A488
+M4072 A149
+M4076 A459
+M4077 A787
+M4085 A1351
+M4086 A10
+M4092 A1204
+M4097 A293
+M4098 A1310
+M4099 A471
+M4102 A1367
+M4104 A1163
+M4110 A635
+M4112 A917
+M4124 A712
+M4128 A996
+M4132 A423
+M4135 A417
+M4141 A1322
+M4158 A1181
+M4159 A869
+M4171 A763
+M4173 A242
+M4180 A1298
+M4182 A576
+M4183 A242
+M4188 A1396
+M4189 A1389
+M4198 A195
+M4202 A1010
+M4208 A560
+M4212 A692
+M4214 A1133
+M4216 A890
+M4220 A542
+M4223 A44
+M4230 A1069
+M4231 A352
+M4239 A650
+M4252 A1189
+M4254 A6
+M4256 A786
+M4257 A179
+M4260 A385
+M4265 A494
+M4274 A132
+M4277 A826
+M4278 A1037
+M4279 A852
+M4282 A1364
+M4290 A405
+M4293 A651
+M4295 A1237
+M4300 A596
+M4303 A121
+M4313 A1297
+M4315 A774
+M4330 A579
+M4331 A299
+M4336 A218
+M4344 A159
+M4346 A836
+M4350 A1310
+M4352 A724
+M4358 A1277
+M4359 A615
+M4369 A689
+M4381 A1155
+M4396 A830
+M4401 A760
+M4408 A998
+M4411 A1287
+M4412 A723
+M4413 A613
+M4423 A179
+M4437 A1255
+M4446 A935
+M4458 A448
+M4462 A930
+M4466 A1321
+M4472 A1292
+M4485 A412
+M4486 A122
+M4489 A1350
+M4491 A904
+M4493 A9
+M4494 A740
+M4496 A1345
+M4508 A676
+M4511 A849
+M4526 A1182
+M4532 A1327
+M4535 A1315
+M4536 A316
+M4541 A571
+M4543 A1066
+M4551 A306
+M4557 A645
+M4564 A314
+M4565 A753
+M4568 A805
+M4569 A966
+M4573 A1088
+M4575 A554
+M4577 A992
+M4584 A721
+M4589 A948
+M4594 A166
+M4596 A1200
+M4604 A280
+M4623 A777
+M4631 A11
+M4636 A600
+M4638 A1277
+M4645 A1342
+M4649 A1317
+M4650 A293
+M4663 A1342
+M4677 A840
+M4678 A623
+M4681 A395
+M4690 A771
+M4694 A882
+M4695 A370
+M4730 A1175
+M4741 A1071
+M4744 A492
+M4745 A589
+M4752 A401
+M4761 A1254
+M4762 A154
+M4765 A1000
+M4773 A1325
+M4778 A1157
+M4791 A324
+M4795 A1201
+M4821 A1083
+M4830 A609
+M4852 A1092
+M4853 A810
+M4860 A627
+M4863 A1356
+M4866 A618
+M4882 A1194
+M4895 A491
+M4905 A388
+M4914 A1169
+M4927 A44
+M4930 A781
+M4942 A461
+M4944 A494
+M4949 A485
+M4950 A485
+M4960 A195
+M4971 A412
+M4979 A514
+M5001 A1307
+M5012 A339
+M5014 A266
+M5021 A268
+M0 A1151
+M1 A800
+M3 A338
+M6 A762
+M8 A974
+M9 A220
+M11 A90
+M12 A1105
+M13 A800
+M14 A1111
+M15 A692
+M17 A974
+M18 A1115
+M20 A604
+M23 A604
+M24 A795
+M25 A901
+M27 A1035
+M29 A679
+M30 A1105
+M31 A762
+M32 A961
+M33 A243
+M36 A1221
+M37 A548
+M38 A910
+M39 A551
+M40 A674
+M41 A396
+M44 A1064
+M45 A1373
+M46 A211
+M47 A1054
+M50 A491
+M53 A433
+M54 A1213
+M58 A84
+M60 A891
+M61 A70
+M62 A214
+M65 A226
+M66 A9
+M69 A1393
+M70 A947
+M71 A233
+M72 A797
+M73 A1301
+M74 A590
+M75 A621
+M79 A1208
+M80 A1393
+M83 A632
+M84 A804
+M86 A783
+M87 A1273
+M92 A1324
+M93 A895
+M95 A318
+M96 A428
+M97 A338
+M99 A604
+M100 A1355
+M101 A458
+M102 A284
+M103 A490
+M105 A190
+M106 A243
+M107 A970
+M109 A321
+M111 A690
+M112 A433
+M113 A1364
+M114 A205
+M115 A220
+M116 A794
+M118 A1383
+M119 A450
+M120 A9
+M121 A1135
+M122 A208
+M127 A808
+M128 A154
+M129 A1316
+M130 A808
+M134 A962
+M135 A1295
+M136 A638
+M140 A802
+M141 A794
+M142 A387
+M143 A912
+M145 A258
+M147 A530
+M148 A1135
+M149 A275
+M150 A733
+M151 A907
+M152 A602
+M155 A427
+M158 A664
+M159 A402
+M160 A993
+M161 A762
+M162 A422
+M164 A1150
+M165 A983
+M167 A660
+M169 A352
+M171 A783
+M172 A1055
+M174 A1120
+M175 A1064
+M176 A1129
+M178 A757
+M179 A862
+M181 A1010
+M182 A503
+M183 A594
+M184 A736
+M185 A85
+M186 A843
+M187 A1120
+M188 A824
+M189 A898
+M190 A1232
+M191 A1363
+M193 A441
+M194 A1263
+M195 A220
+M197 A148
+M200 A208
+M202 A1380
+M203 A168
+M205 A800
+M207 A490
+M208 A957
+M210 A781
+M211 A36
+M213 A82
+M214 A1238
+M215 A1258
+M216 A1218
+M218 A563
+M219 A782
+M220 A1150
+M221 A276
+M222 A190
+M224 A101
+M225 A251
+M226 A652
+M227 A1295
+M228 A717
+M229 A657
+M231 A1376
+M234 A1232
+M235 A106
+M236 A617
+M237 A617
+M238 A622
+M240 A866
+M241 A903
+M243 A620
+M247 A187
+M248 A1329
+M250 A208
+M251 A716
+M252 A271
+M253 A1111
+M256 A1146
+M257 A40
+M264 A154
+M270 A859
+M271 A686
+M272 A572
+M274 A538
+M276 A1021
+M277 A1187
+M278 A913
+M280 A318
+M281 A1208
+M282 A935
+M283 A581
+M285 A2
+M286 A253
+M287 A893
+M288 A811
+M289 A1115
+M290 A1280
+M291 A1128
+M292 A575
+M293 A1185
+M295 A30
+M296 A251
+M298 A206
+M299 A245
+M303 A258
+M304 A110
+M305 A754
+M306 A1396
+M307 A1114
+M308 A1393
+M309 A930
+M310 A499
+M311 A1213
+M312 A1167
+M314 A888
+M315 A471
+M316 A532
+M318 A352
+M321 A915
+M324 A1194
+M326 A1213
+M330 A832
+M332 A70
+M334 A1246
+M336 A182
+M337 A891
+M338 A1329
+M339 A484
+M340 A859
+M341 A1000
+M344 A924
+M345 A89
+M346 A665
+M347 A917
+M348 A1232
+M351 A208
+M352 A1335
+M353 A585
+M354 A1092
+M356 A990
+M359 A540
+M360 A8
+M361 A947
+M363 A729
+M364 A1156
+M365 A357
+M368 A436
+M369 A98
+M370 A882
+M371 A647
+M372 A54
+M373 A1058
+M374 A675
+M375 A623
+M376 A540
+M378 A729
+M379 A318
+M380 A963
+M382 A684
+M385 A1371
+M386 A50
+M387 A681
+M390 A353
+M391 A784
+M392 A202
+M393 A542
+M394 A372
+M395 A1355
+M397 A288
+M399 A655
+M401 A263
+M402 A1367
+M405 A176
+M406 A1248
+M407 A1324
+M410 A405
+M411 A178
+M412 A1268
+M413 A1146
+M415 A832
+M416 A665
+M420 A381
+M423 A1029
+M426 A949
+M429 A96
+M430 A1362
+M434 A159
+M435 A807
+M436 A768
+M437 A19
+M438 A929
+M440 A994
+M442 A1339
+M443 A1372
+M444 A1340
+M445 A935
+M449 A1367
+M450 A243
+M453 A659
+M455 A1153
+M457 A281
+M458 A647
+M460 A149
+M462 A687
+M463 A70
+M465 A561
+M471 A1363
+M474 A146
+M476 A449
+M482 A112
+M483 A176
+M486 A971
+M487 A377
+M488 A862
+M491 A199
+M496 A240
+M498 A1
+M499 A522
+M508 A1324
+M510 A923
+M512 A376
+M513 A131
+M514 A180
+M516 A428
+M517 A941
+M520 A1263
+M521 A208
+M522 A31
+M524 A503
+M526 A1028
+M527 A1064
+M528 A1023
+M534 A928
+M535 A1382
+M539 A402
+M541 A133
+M542 A1187
+M544 A930
+M545 A60
+M546 A981
+M547 A1339
+M548 A290
+M550 A1339
+M552 A1220
+M556 A1234
+M557 A756
+M558 A761
+M560 A1297
+M561 A990
+M562 A928
+M564 A880
+M565 A1081
+M568 A1198
+M570 A711
+M571 A571
+M572 A164
+M576 A298
+M579 A1129
+M580 A1181
+M583 A931
+M584 A80
+M587 A252
+M588 A478
+M589 A506
+M590 A888
+M591 A1094
+M592 A1272
+M594 A621
+M595 A651
+M597 A98
+M598 A520
+M599 A94
+M601 A1221
+M602 A594
+M603 A231
+M604 A794
+M605 A651
+M606 A404
+M609 A467
+M611 A688
+M612 A225
+M614 A500
+M618 A1338
+M623 A1103
+M624 A817
+M627 A596
+M631 A589
+M632 A1368
+M633 A565
+M635 A547
+M636 A967
+M639 A930
+M640 A185
+M642 A993
+M644 A472
+M646 A298
+M648 A351
+M650 A391
+M651 A1314
+M653 A934
+M655 A750
+M656 A1238
+M657 A1242
+M658 A243
+M659 A149
+M661 A557
+M662 A595
+M663 A766
+M664 A192
+M666 A276
+M667 A122
+M668 A553
+M669 A486
+M671 A320
+M672 A980
+M673 A173
+M675 A1371
+M676 A300
+M677 A437
+M678 A887
+M679 A1153
+M684 A872
+M686 A202
+M687 A759
+M690 A567
+M691 A1125
+M692 A14
+M695 A344
+M696 A535
+M698 A5
+M699 A155
+M701 A920
+M702 A1333
+M703 A509
+M706 A106
+M707 A260
+M709 A1001
+M710 A1185
+M711 A526
+M712 A899
+M714 A1363
+M717 A221
+M718 A1102
+M721 A84
+M722 A1309
+M723 A390
+M725 A545
+M726 A52
+M728 A902
+M729 A277
+M731 A191
+M732 A1192
+M734 A1272
+M736 A930
+M737 A484
+M738 A524
+M739 A1047
+M740 A299
+M746 A292
+M747 A923
+M751 A320
+M753 A1293
+M754 A432
+M755 A155
+M761 A123
+M763 A155
+M766 A327
+M767 A1106
+M768 A1093
+M769 A550
+M771 A75
+M773 A1389
+M774 A373
+M775 A1200
+M776 A339
+M778 A378
+M781 A373
+M784 A997
+M785 A1378
+M786 A247
+M787 A242
+M788 A705
+M789 A446
+M793 A750
+M794 A974
+M795 A387
+M796 A79
+M801 A1177
+M804 A21
+M806 A222
+M807 A332
+M809 A620
+M815 A1042
+M817 A1116
+M818 A1029
+M821 A1367
+M823 A227
+M824 A869
+M827 A242
+M828 A886
+M830 A511
+M831 A775
+M834 A1108
+M838 A1185
+M840 A1197
+M841 A149
+M842 A1058
+M843 A45
+M844 A483
+M846 A590
+M847 A1098
+M850 A1180
+M853 A30
+M854 A118
+M856 A280
+M860 A5
+M862 A1311
+M863 A150
+M864 A705
+M865 A736
+M869 A33
+M871 A784
+M874 A492
+M876 A1364
+M877 A118
+M878 A433
+M879 A361
+M880 A942
+M881 A1232
+M887 A91
+M890 A668
+M891 A177
+M892 A1263
+M893 A877
+M894 A745
+M895 A1224
+M896 A1310
+M897 A350
+M900 A433
+M901 A122
+M902 A213
+M903 A1150
+M905 A482
+M906 A891
+M907 A928
+M908 A106
+M909 A1284
+M911 A645
+M913 A512
+M914 A1044
+M915 A243
+M916 A1389
+M917 A1119
+M918 A1228
+M920 A164
+M924 A909
+M925 A1078
+M926 A1120
+M929 A471
+M930 A1223
+M931 A107
+M933 A412
+M934 A590
+M935 A1359
+M937 A27
+M938 A730
+M940 A762
+M942 A1158
+M945 A589
+M946 A680
+M947 A192
+M949 A168
+M950 A1209
+M951 A588
+M953 A590
+M955 A1027
+M956 A178
+M957 A252
+M958 A193
+M959 A912
+M960 A922
+M964 A678
+M966 A888
+M969 A1159
+M970 A388
+M972 A584
+M973 A880
+M974 A1371
+M977 A18
+M978 A987
+M979 A1096
+M982 A1170
+M983 A833
+M987 A1169
+M991 A791
+M995 A437
+M996 A208
+M997 A686
+M998 A992
+M1001 A135
+M1003 A779
+M1005 A1316
+M1008 A704
+M1011 A387
+M1012 A83
+M1013 A934
+M1015 A892
+M1017 A910
+M1018 A113
+M1021 A1199
+M1022 A606
+M1024 A1392
+M1025 A1335
+M1026 A549
+M1027 A1148
+M1028 A1029
+M1029 A652
+M1031 A1208
+M1032 A659
+M1034 A352
+M1036 A184
+M1041 A449
+M1042 A975
+M1048 A400
+M1049 A670
+M1050 A1152
+M1051 A1222
+M1052 A1082
+M1053 A813
+M1057 A492
+M1058 A1314
+M1060 A362
+M1065 A623
+M1066 A1061
+M1067 A378
+M1069 A880
+M1072 A1099
+M1073 A537
+M1074 A1063
+M1075 A764
+M1078 A80
+M1079 A807
+M1080 A262
+M1081 A734
+M1083 A1015
+M1084 A966
+M1085 A777
+M1087 A503
+M1089 A1396
+M1090 A815
+M1092 A1328
+M1094 A22
+M1095 A53
+M1096 A1045
+M1098 A252
+M1100 A935
+M1102 A1142
+M1104 A1300
+M1105 A479
+M1107 A1333
+M1111 A246
+M1112 A493
+M1113 A1195
+M1119 A180
+M1124 A176
+M1125 A1153
+M1126 A479
+M1127 A1151
+M1128 A302
+M1131 A468
+M1134 A146
+M1137 A1108
+M1138 A782
+M1139 A956
+M1140 A677
+M1141 A209
+M1142 A314
+M1144 A1294
+M1145 A1322
+M1146 A102
+M1147 A369
+M1148 A1137
+M1149 A506
+M1151 A444
+M1154 A227
+M1155 A475
+M1157 A1065
+M1159 A1384
+M1162 A1028
+M1166 A244
+M1167 A561
+M1168 A548
+M1171 A345
+M1172 A1133
+M1181 A1000
+M1184 A198
+M1185 A123
+M1186 A703
+M1188 A518
+M1189 A210
+M1190 A432
+M1191 A741
+M1192 A284
+M1193 A1223
+M1196 A241
+M1199 A586
+M1200 A951
+M1201 A518
+M1203 A70
+M1204 A527
+M1205 A476
+M1206 A800
+M1207 A227
+M1209 A252
+M1210 A934
+M1212 A692
+M1214 A1129
+M1219 A246
+M1220 A766
+M1221 A730
+M1224 A823
+M1227 A788
+M1228 A316
+M1230 A471
+M1231 A1008
+M1232 A20
+M1233 A974
+M1234 A1243
+M1237 A638
+M1238 A1132
+M1239 A959
+M1240 A479
+M1243 A1064
+M1246 A561
+M1247 A484
+M1248 A550
+M1253 A526
+M1258 A750
+M1261 A932
+M1263 A417
+M1266 A595
+M1267 A547
+M1271 A675
+M1272 A1356
+M1277 A767
+M1278 A856
+M1284 A141
+M1285 A1308
+M1287 A337
+M1288 A1224
+M1289 A501
+M1290 A602
+M1291 A150
+M1295 A235
+M1296 A116
+M1299 A460
+M1303 A1396
+M1304 A304
+M1305 A102
+M1306 A871
+M1307 A1303
+M1308 A1356
+M1311 A1034
+M1312 A922
+M1313 A440
+M1319 A449
+M1325 A1057
+M1326 A360
+M1328 A184
+M1331 A912
+M1332 A1355
+M1333 A395
+M1345 A674
+M1348 A656
+M1351 A189
+M1352 A134
+M1353 A8
+M1354 A756
+M1355 A145
+M1357 A4
+M1358 A1
+M1360 A1023
+M1362 A1037
+M1366 A177
+M1368 A185
+M1370 A730
+M1371 A1058
+M1373 A100
+M1375 A164
+M1377 A1229
+M1378 A589
+M1379 A314
+M1380 A1253
+M1381 A730
+M1383 A1311
+M1385 A1061
+M1386 A1229
+M1387 A1102
+M1388 A522
+M1389 A1048
+M1390 A703
+M1391 A122
+M1392 A519
+M1396 A828
+M1397 A1303
+M1399 A344
+M1400 A1006
+M1401 A394
+M1402 A557
+M1403 A933
+M1405 A955
+M1409 A911
+M1410 A85
+M1411 A271
+M1412 A54
+M1415 A957
+M1416 A437
+M1417 A879
+M1418 A883
+M1421 A116
+M1423 A371
+M1424 A690
+M1426 A108
+M1428 A551
+M1431 A512
+M1433 A1269
+M1434 A180
+M1436 A314
+M1437 A272
+M1438 A703
+M1439 A1010
+M1441 A411
+M1445 A1001
+M1449 A1045
+M1451 A323
+M1454 A530
+M1455 A6
+M1458 A1026
+M1460 A931
+M1463 A1063
+M1465 A437
+M1474 A1132
+M1475 A802
+M1478 A1217
+M1479 A1079
+M1483 A359
+M1488 A882
+M1490 A205
+M1495 A518
+M1498 A868
+M1500 A1081
+M1501 A155
+M1502 A458
+M1508 A804
+M1510 A1123
+M1513 A557
+M1514 A296
+M1519 A740
+M1521 A1310
+M1523 A704
+M1524 A164
+M1525 A934
+M1530 A508
+M1533 A49
+M1534 A105
+M1535 A381
+M1539 A1175
+M1540 A29
+M1542 A704
+M1544 A374
+M1546 A2
+M1548 A313
+M1549 A1068
+M1551 A343
+M1554 A462
+M1555 A497
+M1556 A1093
+M1558 A1129
+M1561 A1288
+M1564 A362
+M1565 A1261
+M1567 A1327
+M1568 A376
+M1569 A975
+M1571 A935
+M1576 A1270
+M1577 A668
+M1578 A335
+M1579 A237
+M1581 A620
+M1582 A733
+M1584 A813
+M1586 A397
+M1587 A987
+M1589 A572
+M1591 A1254
+M1592 A1092
+M1593 A932
+M1594 A929
+M1595 A752
+M1597 A1293
+M1598 A1071
+M1599 A1114
+M1600 A1144
+M1601 A333
+M1604 A600
+M1605 A563
+M1606 A607
+M1608 A597
+M1611 A458
+M1616 A315
+M1617 A959
+M1621 A1176
+M1624 A1076
+M1637 A704
+M1638 A810
+M1641 A734
+M1646 A1276
+M1647 A44
+M1648 A1012
+M1655 A150
+M1657 A544
+M1661 A583
+M1662 A343
+M1665 A433
+M1666 A759
+M1673 A455
+M1674 A912
+M1675 A799
+M1676 A315
+M1677 A1392
+M1680 A112
+M1681 A1014
+M1683 A347
+M1685 A999
+M1688 A1310
+M1689 A656
+M1690 A1155
+M1691 A654
+M1694 A965
+M1699 A1151
+M1701 A437
+M1708 A1380
+M1711 A388
+M1712 A330
+M1713 A1131
+M1714 A1093
+M1716 A22
+M1717 A221
+M1718 A456
+M1722 A176
+M1725 A1097
+M1739 A284
+M1741 A1174
+M1742 A446
+M1743 A860
+M1744 A31
+M1751 A412
+M1755 A532
+M1759 A1146
+M1760 A1130
+M1762 A492
+M1763 A1027
+M1764 A382
+M1765 A1055
+M1768 A29
+M1771 A651
+M1772 A575
+M1773 A1309
+M1775 A645
+M1777 A704
+M1778 A770
+M1783 A18
+M1784 A806
+M1786 A889
+M1787 A654
+M1788 A193
+M1789 A726
+M1794 A1280
+M1798 A1135
+M1799 A638
+M1801 A714
+M1803 A1244
+M1805 A1208
+M1806 A309
+M1812 A216
+M1813 A176
+M1814 A757
+M1815 A215
+M1816 A6
+M1820 A183
+M1824 A955
+M1825 A780
+M1827 A780
+M1829 A885
+M1833 A1116
+M1835 A74
+M1838 A57
+M1840 A627
+M1841 A237
+M1846 A122
+M1847 A177
+M1848 A575
+M1849 A546
+M1852 A748
+M1853 A1383
+M1856 A1353
+M1857 A1215
+M1859 A497
+M1861 A680
+M1862 A1123
+M1864 A843
+M1868 A55
+M1870 A934
+M1871 A75
+M1875 A1066
+M1877 A1154
+M1879 A707
+M1881 A1261
+M1885 A1091
+M1886 A1292
+M1890 A912
+M1891 A1269
+M1892 A218
+M1894 A125
+M1895 A1061
+M1898 A375
+M1900 A1347
+M1901 A105
+M1902 A590
+M1903 A1014
+M1905 A894
+M1906 A404
+M1908 A688
+M1909 A862
+M1910 A18
+M1914 A538
+M1915 A1337
+M1917 A1123
+M1920 A329
+M1921 A772
+M1925 A942
+M1926 A322
+M1929 A191
+M1937 A276
+M1938 A896
+M1939 A758
+M1941 A246
+M1944 A1323
+M1945 A861
+M1948 A592
+M1949 A653
+M1955 A551
+M1959 A660
+M1961 A59
+M1965 A511
+M1973 A1010
+M1974 A380
+M1975 A849
+M1978 A433
+M1979 A246
+M1981 A1281
+M1982 A807
+M1989 A935
+M1994 A508
+M1995 A1161
+M1998 A1250
+M2000 A508
+M2005 A673
+M2011 A567
+M2015 A849
+M2016 A393
+M2017 A762
+M2018 A911
+M2019 A994
+M2020 A352
+M2022 A175
+M2024 A654
+M2025 A487
+M2026 A1133
+M2028 A14
+M2034 A929
+M2035 A1234
+M2036 A35
+M2038 A968
+M2040 A516
+M2041 A516
+M2042 A810
+M2046 A1303
+M2047 A459
+M2049 A901
+M2055 A1376
+M2056 A981
+M2059 A1323
+M2062 A673
+M2065 A705
+M2070 A223
+M2071 A1091
+M2076 A1320
+M2079 A669
+M2082 A758
+M2083 A1181
+M2084 A1019
+M2085 A877
+M2088 A1220
+M2091 A759
+M2095 A518
+M2099 A102
+M2103 A79
+M2104 A1193
+M2105 A318
+M2107 A1032
+M2109 A965
+M2110 A406
+M2111 A1294
+M2113 A1094
+M2115 A141
+M2118 A290
+M2122 A163
+M2124 A40
+M2127 A320
+M2128 A782
+M2129 A237
+M2133 A746
+M2134 A22
+M2137 A123
+M2138 A894
+M2139 A1351
+M2151 A432
+M2152 A1006
+M2153 A1033
+M2156 A1282
+M2160 A128
+M2164 A608
+M2165 A446
+M2166 A449
+M2168 A521
+M2170 A415
+M2171 A838
+M2174 A366
+M2176 A171
+M2177 A27
+M2181 A445
+M2184 A1340
+M2192 A483
+M2193 A353
+M2195 A72
+M2199 A423
+M2205 A1226
+M2207 A746
+M2209 A1229
+M2210 A563
+M2212 A334
+M2221 A150
+M2222 A102
+M2223 A1111
+M2225 A342
+M2227 A1108
+M2230 A616
+M2231 A1051
+M2232 A402
+M2237 A45
+M2238 A666
+M2240 A261
+M2244 A356
+M2245 A1226
+M2248 A545
+M2249 A864
+M2250 A1327
+M2251 A524
+M2253 A732
+M2255 A319
+M2260 A1165
+M2262 A589
+M2263 A392
+M2268 A1301
+M2270 A283
+M2274 A151
+M2275 A513
+M2276 A7
+M2278 A575
+M2279 A745
+M2282 A632
+M2284 A1342
+M2286 A958
+M2287 A497
+M2289 A1140
+M2292 A22
+M2293 A424
+M2294 A179
+M2297 A1135
+M2298 A1119
+M2307 A84
+M2308 A1202
+M2310 A113
+M2311 A1338
+M2312 A232
+M2314 A297
+M2315 A1165
+M2316 A1315
+M2317 A827
+M2321 A989
+M2323 A191
+M2326 A1165
+M2328 A1344
+M2329 A582
+M2331 A1084
+M2337 A1394
+M2344 A627
+M2352 A326
+M2356 A472
+M2358 A361
+M2361 A335
+M2364 A521
+M2369 A413
+M2372 A440
+M2375 A1139
+M2377 A1142
+M2380 A736
+M2383 A942
+M2389 A361
+M2391 A1301
+M2392 A1144
+M2393 A1081
+M2395 A1298
+M2396 A911
+M2399 A129
+M2401 A481
+M2402 A978
+M2403 A344
+M2406 A1386
+M2409 A126
+M2411 A682
+M2413 A1153
+M2414 A1002
+M2418 A1131
+M2420 A1336
+M2421 A703
+M2425 A629
+M2428 A1297
+M2429 A1289
+M2430 A313
+M2434 A795
+M2435 A1036
+M2436 A869
+M2437 A483
+M2439 A296
+M2442 A1349
+M2448 A163
+M2449 A968
+M2450 A606
+M2451 A515
+M2452 A1111
+M2454 A416
+M2457 A948
+M2463 A1167
+M2465 A1391
+M2470 A295
+M2473 A353
+M2474 A529
+M2475 A672
+M2476 A502
+M2481 A332
+M2486 A568
+M2488 A314
+M2489 A248
+M2490 A579
+M2491 A1030
+M2492 A1020
+M2493 A345
+M2495 A841
+M2496 A1130
+M2498 A835
+M2499 A674
+M2502 A410
+M2503 A776
+M2508 A1028
+M2509 A225
+M2510 A603
+M2513 A553
+M2517 A520
+M2520 A253
+M2521 A88
+M2535 A1111
+M2538 A806
+M2539 A523
+M2542 A837
+M2543 A835
+M2547 A651
+M2548 A926
+M2553 A1014
+M2556 A1309
+M2558 A125
+M2560 A204
+M2562 A445
+M2567 A96
+M2568 A129
+M2571 A1031
+M2572 A897
+M2577 A1093
+M2579 A137
+M2583 A1077
+M2584 A534
+M2590 A1153
+M2592 A990
+M2601 A1339
+M2602 A319
+M2603 A830
+M2607 A1322
+M2611 A545
+M2613 A597
+M2615 A215
+M2616 A1307
+M2617 A247
+M2619 A1020
+M2620 A1268
+M2624 A435
+M2625 A770
+M2627 A872
+M2628 A790
+M2629 A1011
+M2630 A1371
+M2632 A1270
+M2634 A740
+M2636 A937
+M2638 A422
+M2639 A1048
+M2641 A169
+M2642 A312
+M2644 A25
+M2645 A1395
+M2646 A50
+M2648 A45
+M2650 A82
+M2652 A1081
+M2659 A468
+M2663 A693
+M2664 A860
+M2666 A696
+M2669 A711
+M2675 A711
+M2678 A907
+M2680 A872
+M2681 A1333
+M2682 A402
+M2689 A757
+M2691 A635
+M2692 A975
+M2695 A667
+M2698 A1184
+M2700 A1213
+M2701 A819
+M2703 A1345
+M2704 A811
+M2706 A1074
+M2709 A970
+M2714 A54
+M2716 A599
+M2718 A1136
+M2721 A32
+M2723 A1346
+M2725 A832
+M2726 A1352
+M2731 A91
+M2736 A1093
+M2738 A16
+M2741 A1053
+M2755 A1116
+M2757 A450
+M2760 A1008
+M2763 A1356
+M2764 A1097
+M2770 A424
+M2772 A362
+M2773 A692
+M2776 A627
+M2777 A393
+M2778 A280
+M2780 A563
+M2786 A1197
+M2787 A694
+M2788 A563
+M2789 A1056
+M2790 A10
+M2793 A1328
+M2794 A1273
+M2795 A167
+M2796 A1217
+M2798 A790
+M2803 A725
+M2809 A22
+M2818 A20
+M2819 A243
+M2823 A1267
+M2826 A1036
+M2827 A455
+M2829 A100
+M2833 A575
+M2835 A296
+M2836 A1376
+M2837 A935
+M2840 A294
+M2844 A1292
+M2845 A1388
+M2851 A391
+M2852 A461
+M2857 A298
+M2859 A1273
+M2860 A1253
+M2864 A689
+M2865 A149
+M2868 A799
+M2869 A762
+M2876 A1240
+M2877 A139
+M2878 A204
+M2882 A1092
+M2883 A1130
+M2886 A237
+M2887 A291
+M2888 A279
+M2891 A30
+M2896 A502
+M2898 A313
+M2902 A1081
+M2905 A554
+M2908 A1258
+M2910 A759
+M2915 A934
+M2917 A940
+M2918 A987
+M2921 A586
+M2922 A1135
+M2923 A1244
+M2924 A727
+M2926 A862
+M2929 A1308
+M2930 A326
+M2931 A1123
+M2934 A1262
+M2938 A1290
+M2939 A236
+M2942 A1173
+M2944 A253
+M2945 A1041
+M2948 A1082
+M2949 A1203
+M2954 A1153
+M2955 A319
+M2959 A886
+M2961 A32
+M2962 A1267
+M2963 A449
+M2965 A568
+M2966 A114
+M2967 A160
+M2969 A1111
+M2971 A1232
+M2979 A958
+M2980 A1381
+M2982 A47
+M2984 A1082
+M2995 A345
+M2997 A1194
+M2998 A972
+M3001 A1315
+M3002 A616
+M3003 A176
+M3004 A972
+M3008 A134
+M3010 A824
+M3017 A448
+M3024 A1262
+M3026 A611
+M3028 A1211
+M3030 A128
+M3031 A755
+M3035 A1341
+M3036 A41
+M3039 A67
+M3041 A1231
+M3042 A624
+M3044 A845
+M3045 A174
+M3046 A198
+M3050 A180
+M3054 A1171
+M3055 A847
+M3056 A1299
+M3058 A744
+M3060 A894
+M3061 A1260
+M3062 A723
+M3064 A1081
+M3067 A396
+M3069 A703
+M3070 A407
+M3076 A941
+M3080 A139
+M3089 A677
+M3091 A71
+M3092 A1052
+M3095 A477
+M3096 A4
+M3098 A1016
+M3099 A312
+M3100 A404
+M3102 A195
+M3103 A365
+M3108 A137
+M3111 A270
+M3113 A1366
+M3116 A285
+M3117 A448
+M3118 A110
+M3119 A115
+M3120 A931
+M3123 A8
+M3124 A177
+M3125 A819
+M3130 A244
+M3132 A1303
+M3135 A884
+M3138 A1387
+M3139 A1000
+M3140 A415
+M3143 A1271
+M3148 A902
+M3150 A176
+M3155 A458
+M3156 A237
+M3158 A1217
+M3160 A1319
+M3161 A324
+M3163 A1296
+M3164 A142
+M3165 A43
+M3166 A845
+M3170 A754
+M3173 A1078
+M3175 A1218
+M3179 A1085
+M3188 A270
+M3193 A1370
+M3196 A992
+M3198 A445
+M3200 A1220
+M3203 A1077
+M3205 A1213
+M3206 A501
+M3209 A647
+M3212 A472
+M3213 A886
+M3228 A153
+M3233 A392
+M3241 A32
+M3249 A1322
+M3256 A889
+M3258 A509
+M3262 A1055
+M3267 A24
+M3270 A374
+M3271 A795
+M3274 A1352
+M3281 A668
+M3284 A312
+M3287 A152
+M3290 A1344
+M3294 A1261
+M3300 A160
+M3302 A1228
+M3306 A47
+M3307 A35
+M3317 A1063
+M3331 A1145
+M3345 A487
+M3351 A198
+M3353 A1014
+M3355 A447
+M3358 A1030
+M3366 A1217
+M3368 A851
+M3370 A1362
+M3374 A1364
+M3380 A1102
+M3382 A1159
+M3384 A1154
+M3386 A958
+M3390 A522
+M3391 A684
+M3392 A354
+M3393 A854
+M3394 A1027
+M3404 A1259
+M3414 A938
+M3417 A407
+M3421 A1006
+M3424 A448
+M3435 A865
+M3444 A134
+M3447 A958
+M3448 A689
+M3452 A1370
+M3457 A526
+M3461 A762
+M3463 A560
+M3466 A935
+M3468 A588
+M3470 A550
+M3475 A302
+M3476 A491
+M3477 A978
+M3479 A1382
+M3480 A291
+M3483 A1055
+M3484 A1386
+M3488 A184
+M3489 A926
+M3492 A1022
+M3493 A1105
+M3498 A487
+M3502 A673
+M3503 A496
+M3510 A496
+M3511 A449
+M3512 A268
+M3514 A249
+M3515 A55
+M3530 A731
+M3531 A136
+M3537 A283
+M3542 A981
+M3560 A314
+M3566 A1347
+M3569 A819
+M3570 A1155
+M3574 A100
+M3580 A397
+M3581 A1088
+M3582 A480
+M3586 A1324
+M3587 A686
+M3589 A1152
+M3591 A41
+M3592 A164
+M3597 A971
+M3599 A145
+M3603 A27
+M3618 A880
+M3619 A810
+M3621 A1389
+M3624 A1261
+M3627 A1100
+M3641 A929
+M3647 A1155
+M3649 A981
+M3654 A992
+M3662 A496
+M3663 A479
+M3668 A488
+M3669 A1220
+M3674 A1173
+M3675 A22
+M3680 A969
+M3681 A397
+M3696 A71
+M3697 A720
+M3699 A128
+M3704 A1217
+M3709 A76
+M3713 A500
+M3714 A373
+M3726 A208
+M3729 A1081
+M3732 A26
+M3733 A499
+M3734 A459
+M3741 A757
+M3746 A361
+M3751 A602
+M3762 A1094
+M3768 A913
+M3779 A1144
+M3788 A1084
+M3805 A817
+M3811 A267
+M3814 A1327
+M3817 A987
+M3821 A1311
+M3824 A475
+M3830 A897
+M3841 A669
+M3843 A74
+M3844 A984
+M3846 A781
+M3858 A1352
+M3867 A1069
+M3869 A213
+M3879 A258
+M3884 A767
+M3889 A1184
+M3891 A402
+M3894 A64
+M3907 A407
+M3908 A893
+M3910 A18
+M3911 A673
+M3913 A904
+M3918 A958
+M3921 A701
+M3925 A32
+M3928 A951
+M3929 A1010
+M3936 A64
+M3942 A776
+M3947 A402
+M3952 A1082
+M3955 A1295
+M3956 A798
+M3967 A597
+M3969 A952
+M3970 A1005
+M3980 A555
+M3981 A20
+M3987 A239
+M3988 A927
+M3994 A786
+M4009 A784
+M4011 A193
+M4023 A1134
+M4025 A1124
+M4035 A189
+M4046 A1070
+M4047 A668
+M4049 A1005
+M4050 A832
+M4053 A1176
+M4054 A152
+M4055 A998
+M4057 A763
+M4058 A362
+M4059 A1343
+M4063 A899
+M4066 A25
+M4068 A192
+M4077 A24
+M4085 A157
+M4097 A921
+M4102 A847
+M4112 A733
+M4124 A618
+M4128 A243
+M4132 A565
+M4137 A254
+M4141 A182
+M4145 A1005
+M4150 A125
+M4158 A830
+M4165 A837
+M4168 A537
+M4171 A74
+M4173 A740
+M4180 A1203
+M4182 A1395
+M4183 A1
+M4202 A752
+M4212 A190
+M4216 A774
+M4222 A905
+M4223 A161
+M4230 A1136
+M4231 A877
+M4239 A262
+M4247 A668
+M4252 A1077
+M4256 A668
+M4268 A536
+M4277 A342
+M4279 A312
+M4282 A416
+M4290 A291
+M4293 A280
+M4303 A387
+M4313 A511
+M4331 A995
+M4344 A813
+M4350 A343
+M4357 A163
+M4379 A792
+M4381 A829
+M4387 A738
+M4401 A1146
+M4404 A1005
+M4411 A1379
+M4421 A943
+M4437 A69
+M4446 A1047
+M4458 A500
+M4462 A854
+M4466 A1378
+M4485 A720
+M4486 A1030
+M4490 A1320
+M4491 A50
+M4492 A1249
+M4493 A216
+M4494 A841
+M4496 A511
+M4530 A1386
+M4532 A1336
+M4535 A1181
+M4536 A1255
+M4537 A668
+M4543 A1308
+M4556 A45
+M4557 A560
+M4564 A1174
+M4569 A1042
+M4573 A1231
+M4575 A341
+M4577 A1136
+M4584 A891
+M4594 A1255
+M4599 A280
+M4604 A825
+M4631 A534
+M4636 A507
+M4638 A174
+M4643 A463
+M4649 A1270
+M4650 A1099
+M4667 A479
+M4677 A84
+M4679 A596
+M4690 A893
+M4694 A901
+M4695 A323
+M4730 A1344
+M4739 A813
+M4741 A1118
+M4744 A1273
+M4752 A824
+M4761 A1297
+M4765 A919
+M4778 A910
+M4791 A813
+M4821 A1020
+M4822 A206
+M4827 A765
+M4830 A502
+M4842 A1195
+M4852 A397
+M4863 A1288
+M4927 A161
+M4930 A231
+M4942 A1081
+M4949 A801
+M4950 A801
+M4960 A607
+M4979 A1285
+M5014 A1250
diff --git a/examples/dhn/dhn_trainer.py b/examples/dhn/dhn_trainer.py
new file mode 100644
index 00000000..7de50008
--- /dev/null
+++ b/examples/dhn/dhn_trainer.py
@@ -0,0 +1,361 @@
+import random
+import numpy as np
+import tensorlayerx as tlx
+from tensorlayerx.model import TrainOneStep
+import tensorflow as tf
+
+from gammagl.data import HeteroGraph
+from gammagl.models.dhn import DHNModel, NODE_TYPE, K_HOP, NUM_FEA, NUM_NEIGHBOR, type2idx
+from gammagl.utils import k_hop_subgraph
+
+# 全局忽略UserWarning
+import warnings
+
+warnings.filterwarnings("ignore", category=UserWarning)
+
+random.seed(0)  # 确保各个算法的数据，邻居等信息一致，保证公平
+
+PATH = 'MA.txt'
+
+G = HeteroGraph()
+
+
+def load_ACM(test_ratio=0.2):
+    edge_index_M = []
+    edge_index_A = []
+
+    with open(PATH, 'r') as f:
+        for line in f.readlines():
+            src, dst = line.strip().split()  # 假设每行包含两个由空格分隔的节点标识符
+            src_type, src_id = src[0], src[1:]  # 解析源节点类型和ID
+            dst_type, dst_id = dst[0], dst[1:]  # 解析目标节点类型和ID
+
+            # 将节点ID转换为整数索引,并放入列表中
+            if src[0] == 'M':
+                edge_index_M.append(int(src_id))
+            elif src[0] == 'A':
+                edge_index_A.append(-int(src_id) - 1)
+
+            if dst[0] == 'M':
+                edge_index_M.append(int(dst_id))
+            elif dst[0] == 'A':
+                edge_index_A.append(-int(dst_id) - 1)
+
+    edge_index = tlx.convert_to_tensor([edge_index_M, edge_index_A])
+    G['M', 'MA', 'A'].edge_index = edge_index
+
+    # 计算分割点
+    sp = 1 - test_ratio * 2
+    num_edge = len(edge_index_M)
+    sp1 = int(num_edge * sp)
+    sp2 = int(num_edge * test_ratio)
+
+    G_train = HeteroGraph()
+    G_val = HeteroGraph()
+    G_test = HeteroGraph()
+
+    # 划分训练集，验证集和测试集
+    G_train['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor([edge_index_M[:sp1], edge_index_A[:sp1]])
+    G_val['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor([edge_index_M[sp1:sp1 + sp2], edge_index_A[sp1:sp1 + sp2]])
+    G_test['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor([edge_index_M[sp1 + sp2:], edge_index_A[sp1 + sp2:]])
+
+    print(
+        f"all edge: {len(G['M', 'MA', 'A'].edge_index[0])}, train edge: {len(G_train['M', 'MA', 'A'].edge_index[0])}, val edge: {len(G_val['M', 'MA', 'A'].edge_index[0])}, test edge: {len(G_test['M', 'MA', 'A'].edge_index[0])}")
+
+    return G_train, G_val, G_test
+
+
+def find_all_simple_paths(edge_index, src, dest, max_length):
+    # 将edge_index转换为邻接列表表示
+    num_nodes = max(edge_index[0].max().item(),
+                    edge_index[1].max().item(),
+                    -edge_index[0].min().item(),
+                    -edge_index[1].min().item(),
+                    src.item()) + 1
+    adj_list = [[] for _ in range(num_nodes)]
+    for u, v in zip(edge_index[0].tolist(), edge_index[1].tolist()):
+        adj_list[u].append(v)
+
+    src = src.item()
+
+    paths = []
+    visited = set()
+    stack = [(src, [src])]
+
+    while stack:
+        (node, path) = stack.pop()
+
+        if node == dest:
+            paths.append(path)
+        elif len(path) < max_length:
+            for neighbor in adj_list[node]:
+                if neighbor not in path:
+                    visited.add((node, neighbor))
+                    stack.append((neighbor, path + [neighbor]))
+            for neighbor in adj_list[node]:
+                if (node, neighbor) in visited:
+                    visited.remove((node, neighbor))
+
+    return paths
+
+
+def dist_encoder(src, dest, G, K_HOP, one_hot=True):
+    if (G.size(1) == 0):
+        paths = []
+    else:
+        paths = find_all_simple_paths(G, src, dest, K_HOP + 2)
+
+    cnt = [K_HOP + 1] * NODE_TYPE  # 超过max_spd的默认截断
+    for path in paths:
+        res = [0] * NODE_TYPE
+        for i in path:
+            if i >= 0:
+                res[type2idx['M']] += 1
+            else:
+                res[type2idx['A']] += 1
+
+        for k in range(NODE_TYPE):
+            cnt[k] = min(cnt[k], res[k])
+
+    # 生成one-hot编码
+    if one_hot:
+        one_hot_list = [np.eye(K_HOP + 2, dtype=np.float64)[cnt[i]]
+                        for i in range(NODE_TYPE)]
+        return np.concatenate(one_hot_list)
+    return cnt
+
+
+def type_encoder(node):
+    res = [0] * NODE_TYPE
+    if node.item() >= 0:
+        res[type2idx['M']] = 1.0
+    else:
+        res[type2idx['A']] = 1.0
+    return res
+
+
+mini_batch = []
+fea_batch = []
+
+
+def gen_fea_batch(G, root, fea_dict, hop):
+    fea_batch = []
+    mini_batch.append(root)
+
+    a = [0] * (K_HOP + 2) * 4 + type_encoder(root)
+
+    fea_batch.append(np.asarray(a,
+                                dtype=np.float32
+                                ).reshape(-1, NUM_FEA)
+                     )
+
+    # 一阶邻居采样
+    ns_1 = []
+    src, dst = G
+    for node in mini_batch[-1]:
+        if node.item() >= 0:
+            neighbors_mask = src == node
+        else:
+            neighbors_mask = dst == node
+        neighbors = list(dst[neighbors_mask].numpy())
+        neighbors.append(node.item())
+        random_choice_list = np.random.choice(neighbors, NUM_NEIGHBOR, replace=True)
+        ns_1.append(random_choice_list.tolist())
+    ns_1 = tlx.convert_to_tensor(ns_1)
+    mini_batch.append(ns_1[0])
+
+    de_1 = [
+        np.concatenate([fea_dict[ns_1[0][i].item()], np.asarray(type_encoder(ns_1[0][i]))], axis=0)
+        for i in range(0, ns_1[0].shape[0])
+    ]
+
+    fea_batch.append(np.asarray(de_1,
+                                dtype=np.float32).reshape(1, -1)
+                     )
+
+    # 二阶邻居采样
+    ns_2 = []
+    for node in mini_batch[-1]:
+        if node.item() >= 0:
+            neighbors_mask = src == node
+        else:
+            neighbors_mask = dst == node
+        neighbors = list(dst[neighbors_mask].numpy())
+        neighbors.append(node.item())
+        random_choice_list = np.random.choice(neighbors, NUM_NEIGHBOR, replace=True)
+        ns_2.append(random_choice_list.tolist())
+    ns_2 = tlx.convert_to_tensor(ns_2)
+
+    de_2 = []
+    for i in range(len(ns_2)):
+        tmp = []
+        for j in range(len(ns_2[0])):
+            tmp.append(
+                np.concatenate([fea_dict[ns_2[i][j].item()], np.asarray(type_encoder(ns_2[i][j]))], axis=0)
+            )
+        de_2.append(tmp)
+
+    fea_batch.append(np.asarray(de_2,
+                                dtype=np.float32).reshape(1, -1)
+                     )
+
+    return np.concatenate(fea_batch, axis=1)
+
+
+def subgraph_sampling_with_DE_node_pair(G, node_pair, K_HOP=2):
+    [A, B] = node_pair
+
+    edge_index = tlx.concat([G['M', 'MA', 'A'].edge_index, reversed(G['M', 'MA', 'A'].edge_index)], axis=1)
+
+    # 求A和B的K跳子图
+    sub_G_for_AB = k_hop_subgraph([A, B], K_HOP, edge_index)
+
+    # 使用布尔索引删除边
+    # 注意：只是删除边，点仍然保留
+    edge_index_np = sub_G_for_AB[1].numpy()
+    remove_indices = tlx.convert_to_tensor([
+        ((edge_index_np[0, i] == A) & (edge_index_np[1, i] == B)) | (
+                (edge_index_np[0, i] == B) & (edge_index_np[1, i] == A))
+        for i in range(sub_G_for_AB[1].shape[1])
+    ])
+    remove_indices = remove_indices.numpy()
+    sub_G_index = sub_G_for_AB[1][:, ~remove_indices]
+
+    sub_G_nodes = set(np.unique(sub_G_for_AB[0].numpy())) | set(
+        np.unique(sub_G_for_AB[1].numpy()))  # 获取图中的点
+    sub_G_nodes = tlx.convert_to_tensor(list(sub_G_nodes))
+
+    # 子图中所有点到node pair的距离
+    SPD_based_on_node_pair = {}
+    for node in sub_G_nodes:
+        tmpA = dist_encoder(A, node, sub_G_index, K_HOP)
+        tmpB = dist_encoder(B, node, sub_G_index, K_HOP)
+
+        SPD_based_on_node_pair[node.item()] = np.concatenate([tmpA, tmpB], axis=0)
+
+    A_fea_batch = gen_fea_batch(sub_G_index, A,
+                                SPD_based_on_node_pair, K_HOP)
+    B_fea_batch = gen_fea_batch(sub_G_index, B,
+                                SPD_based_on_node_pair, K_HOP)
+
+    return A_fea_batch, B_fea_batch
+
+
+def batch_data(G,
+               batch_size=3):
+    edge_index = G['M', 'MA', 'A'].edge_index
+    nodes = set(tlx.convert_to_tensor(np.unique(edge_index[0].numpy()))) | set(
+        tlx.convert_to_tensor(np.unique(edge_index[1].numpy())))
+
+    nodes_list = []
+    for node in nodes:
+        nodes_list.append(node.item())
+
+    num_batch = int(len(edge_index[0]) / batch_size)
+
+    # 打乱边的顺序
+    edge_index_np = np.array(edge_index)
+    permutation = np.random.permutation(edge_index_np.shape[1])  # 生成一个随机排列的索引
+    edge_index_np = edge_index_np[:, permutation]  # 使用这个排列索引来打乱 edge_index
+    edge_index = tlx.convert_to_tensor(edge_index_np)
+
+    for idx in range(num_batch):
+        batch_edge = edge_index[:, idx * batch_size:(idx + 1) * batch_size]  # 取出batch_size条边
+        batch_label = [1.0] * batch_size
+
+        batch_A_fea = []
+        batch_B_fea = []
+        batch_x = []
+        batch_y = []
+
+        i = 0
+        for by in batch_label:
+            bx = batch_edge[:, i:i + 1]
+
+            # 正样本
+            posA, posB = subgraph_sampling_with_DE_node_pair(G, bx, K_HOP=K_HOP)
+            batch_A_fea.append(posA)
+            batch_B_fea.append(posB)
+            batch_y.append(np.asarray(by, dtype=np.float32))
+
+            # 负样本
+            neg_tmpB_id = random.choice(nodes_list)
+            node_pair = tlx.convert_to_tensor([[bx[0].item()], [neg_tmpB_id]])
+
+            negA, negB = subgraph_sampling_with_DE_node_pair(G, node_pair, K_HOP=K_HOP)
+            batch_A_fea.append(negA)
+            batch_B_fea.append(negB)
+            batch_y.append(np.asarray(0.0, dtype=np.float32))
+
+        yield np.asarray(np.squeeze(batch_A_fea)), np.asarray(np.squeeze(batch_B_fea)), np.asarray(
+            batch_y).reshape(batch_size * 2, 1)
+
+
+G_train, G_val, G_test = load_ACM(test_ratio=0.3)
+
+m = DHNModel()
+
+lr = 0.0002
+optim = tlx.optimizers.Adam(lr=lr, weight_decay=0.01)
+
+train_weights = m.trainable_weights
+
+
+class Loss(tlx.model.WithLoss):
+    def __init__(self, net, loss_fn):
+        super(Loss, self).__init__(backbone=net, loss_fn=loss_fn)
+
+    def forward(self, data, y):
+        logits = self.backbone_network(data['n1'], data['n2'], data['label'])
+        y = tlx.convert_to_tensor(y)
+        loss = self._loss_fn(logits, y)
+        return loss
+
+
+net_with_loss = Loss(m, loss_fn=tlx.losses.sigmoid_cross_entropy)
+net_with_train = TrainOneStep(net_with_loss, optim, train_weights)
+
+EPOCH = 100
+BATCH_SIZE = 32
+
+tra_auc_cul = tf.keras.metrics.AUC()
+val_auc_cul = tf.keras.metrics.AUC()
+test_auc_cul = tf.keras.metrics.AUC()
+
+for epoch in range(EPOCH):
+    print("-----Epoch {}/{}-----".format(epoch + 1, EPOCH))
+
+    # train
+    m.set_train()
+    tra_batch_A_fea, tra_batch_B_fea, tra_batch_y = batch_data(G_train, BATCH_SIZE).__next__()
+    tra_out = m(tra_batch_A_fea, tra_batch_B_fea, tra_batch_y)
+
+    data = {
+        "n1": tra_batch_A_fea,
+        "n2": tra_batch_B_fea,
+        "label": tra_batch_y
+    }
+
+    tra_loss = net_with_train(data, tra_batch_y)
+    tra_auc_cul.update_state(y_true=tra_batch_y, y_pred=tlx.sigmoid(tra_out).detach().numpy())
+    tra_auc = tra_auc_cul.result().numpy()
+    print('train: ', tra_loss, tra_auc)
+
+    # val
+    m.set_eval()
+    val_batch_A_fea, val_batch_B_fea, val_batch_y = batch_data(G_val, BATCH_SIZE).__next__()
+    val_out = m(val_batch_A_fea, val_batch_B_fea, val_batch_y)
+
+    val_loss = tlx.losses.sigmoid_cross_entropy(output=val_out, target=tlx.convert_to_tensor(val_batch_y))
+    val_auc_cul.update_state(y_true=val_batch_y, y_pred=tlx.sigmoid(val_out).detach().numpy())
+    val_auc = val_auc_cul.result().numpy()
+    print("val: ", val_loss.item(), val_auc)
+
+    # test
+    test_batch_A_fea, test_batch_B_fea, test_batch_y = batch_data(G_test, BATCH_SIZE).__next__()
+    test_out = m(test_batch_A_fea, test_batch_B_fea, test_batch_y)
+
+    test_loss = tlx.losses.sigmoid_cross_entropy(output=test_out, target=tlx.convert_to_tensor(test_batch_y))
+    test_auc_cul.update_state(y_true=test_batch_y, y_pred=tlx.sigmoid(test_out).detach().numpy())
+    test_auc = test_auc_cul.result().numpy()
+    print("test: ", test_loss.item(), test_auc)
diff --git a/gammagl/models/dhn.py b/gammagl/models/dhn.py
new file mode 100644
index 00000000..af8bd069
--- /dev/null
+++ b/gammagl/models/dhn.py
@@ -0,0 +1,82 @@
+import tensorlayerx as tlx
+
+type2idx = {
+    'M': 0,
+    'A': 1,
+    # 'C': 2,
+    # 'T': 3
+}
+
+NODE_TYPE = len(type2idx)
+K_HOP = 2
+
+NUM_FEA = (K_HOP + 2) * 4 + NODE_TYPE
+NUM_NEIGHBOR = 5
+
+
+class DHN(tlx.nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.lin1 = tlx.nn.Linear(in_features=36, out_features=64, act=tlx.nn.ELU(), W_init="xavier_uniform")
+        self.lin2 = tlx.nn.Linear(in_features=82, out_features=64, act=tlx.nn.ELU(), W_init="xavier_uniform")
+        self.lin3 = tlx.nn.Linear(in_features=64, out_features=64, act=tlx.nn.ELU(), W_init="xavier_uniform")
+
+    def forward(self, fea):
+        node = tlx.convert_to_tensor(fea[:, :NUM_FEA])
+
+        # 提取neigh1和neigh2
+        neigh1 = tlx.convert_to_tensor(fea[:, NUM_FEA:NUM_FEA * (NUM_NEIGHBOR + 1)])
+        neigh1 = tlx.reshape(neigh1, [-1, NUM_NEIGHBOR, NUM_FEA])
+
+        neigh2 = tlx.convert_to_tensor(fea[:, NUM_FEA * (NUM_NEIGHBOR + 1):])
+        neigh2 = tlx.reshape(neigh2, [-1, NUM_NEIGHBOR, NUM_NEIGHBOR, NUM_FEA])
+
+        # 对neigh2进行聚合
+        neigh2_agg = tlx.reduce_mean(neigh2, axis=2)
+
+        # 连接node和聚合后的neigh2
+        tmp = tlx.concat(
+            [neigh1, neigh2_agg],
+            axis=2
+        )
+
+        # 将tmp重塑为二维张量
+        flattened_tmp = tmp.view(64 * 5, 36)  
+
+        # 将二维张量传递给线性层
+        flattened_output = self.lin1(flattened_tmp)  
+
+         # 将输出张量重新整形为三维形状
+        tmp = flattened_output.view(64, 5, 64)  
+
+        # 对node和tmp进行聚合
+        emb = tlx.concat(
+            [
+                node, tlx.reduce_mean(tmp, axis=1)
+            ],
+            axis=1
+        )
+
+        # 通过其他线性层
+        emb = self.lin2(emb)
+        emb = self.lin3(emb)
+
+        return emb
+
+
+class DHNModel(tlx.nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.dhn1 = DHN()
+        self.dhn2 = DHN()
+        self.lin1 = tlx.nn.Linear(in_features=128, out_features=32, act=tlx.nn.ELU(), W_init="xavier_uniform")
+        self.lin2 = tlx.nn.Linear(in_features=32, out_features=1, act=tlx.nn.ELU(), W_init="xavier_uniform")
+
+    def forward(self, n1, n2, label):
+        n1_emb = self.dhn1(n1)
+        n2_emb = self.dhn2(n2)
+
+        pred = self.lin1(tlx.concat([n1_emb, n2_emb], axis=1))
+        pred = self.lin2(pred)
+
+        return pred

From 9058ebf6f4852581c89945cf4995d5b5d04aa0df Mon Sep 17 00:00:00 2001
From: liberty <2700472871@qq.com>
Date: Sun, 30 Jun 2024 19:51:27 +0800
Subject: [PATCH 2/8] Modify the dhn code and use MessagePassing

---
 examples/dhn/dhn_trainer.py     | 184 ++++++++++++++++++--------------
 gammagl/layers/conv/dhn_conv.py |  90 ++++++++++++++++
 gammagl/models/dhn.py           |  82 --------------
 3 files changed, 195 insertions(+), 161 deletions(-)
 create mode 100644 gammagl/layers/conv/dhn_conv.py
 delete mode 100644 gammagl/models/dhn.py

diff --git a/examples/dhn/dhn_trainer.py b/examples/dhn/dhn_trainer.py
index 7de50008..d0cb1ee2 100644
--- a/examples/dhn/dhn_trainer.py
+++ b/examples/dhn/dhn_trainer.py
@@ -1,21 +1,20 @@
+import argparse
 import random
 import numpy as np
 import tensorlayerx as tlx
 from tensorlayerx.model import TrainOneStep
-import tensorflow as tf
+from sklearn.metrics import roc_auc_score
 
 from gammagl.data import HeteroGraph
-from gammagl.models.dhn import DHNModel, NODE_TYPE, K_HOP, NUM_FEA, NUM_NEIGHBOR, type2idx
+from gammagl.layers.conv.dhn_conv import DHNModel, NODE_TYPE, K_HOP, NUM_FEA, NUM_NEIGHBOR, BATCH_SIZE, type2idx
 from gammagl.utils import k_hop_subgraph
 
-# 全局忽略UserWarning
+# The UserWarning is ignored globally
 import warnings
 
 warnings.filterwarnings("ignore", category=UserWarning)
 
-random.seed(0)  # 确保各个算法的数据，邻居等信息一致，保证公平
-
-PATH = 'MA.txt'
+random.seed(0)
 
 G = HeteroGraph()
 
@@ -24,13 +23,13 @@ def load_ACM(test_ratio=0.2):
     edge_index_M = []
     edge_index_A = []
 
-    with open(PATH, 'r') as f:
+    with open(args.dataset_path, 'r') as f:
         for line in f.readlines():
-            src, dst = line.strip().split()  # 假设每行包含两个由空格分隔的节点标识符
-            src_type, src_id = src[0], src[1:]  # 解析源节点类型和ID
-            dst_type, dst_id = dst[0], dst[1:]  # 解析目标节点类型和ID
+            src, dst = line.strip().split()
+            src_type, src_id = src[0], src[1:]  # Resolves the source node type and ID
+            dst_type, dst_id = dst[0], dst[1:]  # Resolve the target node type and ID
 
-            # 将节点ID转换为整数索引,并放入列表中
+            # Convert the node ID to an integer index and place it in a list
             if src[0] == 'M':
                 edge_index_M.append(int(src_id))
             elif src[0] == 'A':
@@ -44,7 +43,7 @@ def load_ACM(test_ratio=0.2):
     edge_index = tlx.convert_to_tensor([edge_index_M, edge_index_A])
     G['M', 'MA', 'A'].edge_index = edge_index
 
-    # 计算分割点
+    # Computed split point
     sp = 1 - test_ratio * 2
     num_edge = len(edge_index_M)
     sp1 = int(num_edge * sp)
@@ -54,7 +53,7 @@ def load_ACM(test_ratio=0.2):
     G_val = HeteroGraph()
     G_test = HeteroGraph()
 
-    # 划分训练集，验证集和测试集
+    # Divide the training set, the verification set, and the test set
     G_train['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor([edge_index_M[:sp1], edge_index_A[:sp1]])
     G_val['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor([edge_index_M[sp1:sp1 + sp2], edge_index_A[sp1:sp1 + sp2]])
     G_test['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor([edge_index_M[sp1 + sp2:], edge_index_A[sp1 + sp2:]])
@@ -66,12 +65,12 @@ def load_ACM(test_ratio=0.2):
 
 
 def find_all_simple_paths(edge_index, src, dest, max_length):
-    # 将edge_index转换为邻接列表表示
+    # Converts edge_index to an adjacency list representation
     num_nodes = max(edge_index[0].max().item(),
                     edge_index[1].max().item(),
                     -edge_index[0].min().item(),
                     -edge_index[1].min().item(),
-                    src.item()) + 1
+                    abs(src.item())) + 1
     adj_list = [[] for _ in range(num_nodes)]
     for u, v in zip(edge_index[0].tolist(), edge_index[1].tolist()):
         adj_list[u].append(v)
@@ -99,13 +98,13 @@ def find_all_simple_paths(edge_index, src, dest, max_length):
     return paths
 
 
-def dist_encoder(src, dest, G, K_HOP, one_hot=True):
+def dist_encoder(src, dest, G, K_HOP):
     if (G.size(1) == 0):
         paths = []
     else:
         paths = find_all_simple_paths(G, src, dest, K_HOP + 2)
 
-    cnt = [K_HOP + 1] * NODE_TYPE  # 超过max_spd的默认截断
+    cnt = [K_HOP + 1] * NODE_TYPE  # Default truncation for max_spd exceeded
     for path in paths:
         res = [0] * NODE_TYPE
         for i in path:
@@ -117,8 +116,8 @@ def dist_encoder(src, dest, G, K_HOP, one_hot=True):
         for k in range(NODE_TYPE):
             cnt[k] = min(cnt[k], res[k])
 
-    # 生成one-hot编码
-    if one_hot:
+    # Generate one-hot encoding
+    if args.one_hot:
         one_hot_list = [np.eye(K_HOP + 2, dtype=np.float64)[cnt[i]]
                         for i in range(NODE_TYPE)]
         return np.concatenate(one_hot_list)
@@ -149,7 +148,7 @@ def gen_fea_batch(G, root, fea_dict, hop):
                                 ).reshape(-1, NUM_FEA)
                      )
 
-    # 一阶邻居采样
+    # 1-order neighbor sampling
     ns_1 = []
     src, dst = G
     for node in mini_batch[-1]:
@@ -173,7 +172,7 @@ def gen_fea_batch(G, root, fea_dict, hop):
                                 dtype=np.float32).reshape(1, -1)
                      )
 
-    # 二阶邻居采样
+    # 2-order neighbor sampling
     ns_2 = []
     for node in mini_batch[-1]:
         if node.item() >= 0:
@@ -207,11 +206,11 @@ def subgraph_sampling_with_DE_node_pair(G, node_pair, K_HOP=2):
 
     edge_index = tlx.concat([G['M', 'MA', 'A'].edge_index, reversed(G['M', 'MA', 'A'].edge_index)], axis=1)
 
-    # 求A和B的K跳子图
+    # Find k-hop subgraphs of A and B
     sub_G_for_AB = k_hop_subgraph([A, B], K_HOP, edge_index)
 
-    # 使用布尔索引删除边
-    # 注意：只是删除边，点仍然保留
+    # Remove edges using Boolean indexes
+    # Note: Just remove the edges, the points remain
     edge_index_np = sub_G_for_AB[1].numpy()
     remove_indices = tlx.convert_to_tensor([
         ((edge_index_np[0, i] == A) & (edge_index_np[1, i] == B)) | (
@@ -222,10 +221,10 @@ def subgraph_sampling_with_DE_node_pair(G, node_pair, K_HOP=2):
     sub_G_index = sub_G_for_AB[1][:, ~remove_indices]
 
     sub_G_nodes = set(np.unique(sub_G_for_AB[0].numpy())) | set(
-        np.unique(sub_G_for_AB[1].numpy()))  # 获取图中的点
+        np.unique(sub_G_for_AB[1].numpy()))  # Gets the points in the graph
     sub_G_nodes = tlx.convert_to_tensor(list(sub_G_nodes))
 
-    # 子图中所有点到node pair的距离
+    # Distance from all points in the subgraph to the node pair
     SPD_based_on_node_pair = {}
     for node in sub_G_nodes:
         tmpA = dist_encoder(A, node, sub_G_index, K_HOP)
@@ -253,14 +252,14 @@ def batch_data(G,
 
     num_batch = int(len(edge_index[0]) / batch_size)
 
-    # 打乱边的顺序
+    # Shuffle the order of the edges
     edge_index_np = np.array(edge_index)
-    permutation = np.random.permutation(edge_index_np.shape[1])  # 生成一个随机排列的索引
-    edge_index_np = edge_index_np[:, permutation]  # 使用这个排列索引来打乱 edge_index
+    permutation = np.random.permutation(edge_index_np.shape[1])  # Generate a randomly arranged index
+    edge_index_np = edge_index_np[:, permutation]  # Use this permutation index to scramble edge_index
     edge_index = tlx.convert_to_tensor(edge_index_np)
 
     for idx in range(num_batch):
-        batch_edge = edge_index[:, idx * batch_size:(idx + 1) * batch_size]  # 取出batch_size条边
+        batch_edge = edge_index[:, idx * batch_size:(idx + 1) * batch_size]  # Take out batch_size edges
         batch_label = [1.0] * batch_size
 
         batch_A_fea = []
@@ -272,13 +271,13 @@ def batch_data(G,
         for by in batch_label:
             bx = batch_edge[:, i:i + 1]
 
-            # 正样本
+            # Positive sample
             posA, posB = subgraph_sampling_with_DE_node_pair(G, bx, K_HOP=K_HOP)
             batch_A_fea.append(posA)
             batch_B_fea.append(posB)
             batch_y.append(np.asarray(by, dtype=np.float32))
 
-            # 负样本
+            # Negative sample
             neg_tmpB_id = random.choice(nodes_list)
             node_pair = tlx.convert_to_tensor([[bx[0].item()], [neg_tmpB_id]])
 
@@ -291,16 +290,6 @@ def batch_data(G,
             batch_y).reshape(batch_size * 2, 1)
 
 
-G_train, G_val, G_test = load_ACM(test_ratio=0.3)
-
-m = DHNModel()
-
-lr = 0.0002
-optim = tlx.optimizers.Adam(lr=lr, weight_decay=0.01)
-
-train_weights = m.trainable_weights
-
-
 class Loss(tlx.model.WithLoss):
     def __init__(self, net, loss_fn):
         super(Loss, self).__init__(backbone=net, loss_fn=loss_fn)
@@ -312,50 +301,87 @@ def forward(self, data, y):
         return loss
 
 
-net_with_loss = Loss(m, loss_fn=tlx.losses.sigmoid_cross_entropy)
-net_with_train = TrainOneStep(net_with_loss, optim, train_weights)
+class AUCMetric:
+    def __init__(self):
+        self.true_labels = []
+        self.predicted_scores = []
+
+    def update_state(self, y_true, y_pred):
+        self.true_labels.extend(y_true)
+        self.predicted_scores.extend(y_pred)
+
+    def result(self):
+        auc = roc_auc_score(self.true_labels, self.predicted_scores)
+        return auc
+
+
+def main(args):
+    if str.lower(args.dataset) not in ['acm']:
+        raise ValueError('Unknown dataset: {}'.format(args.dataset))
+    if str.lower(args.dataset) == 'acm':
+        G_train, G_val, G_test = load_ACM(test_ratio=args.test_ratio)
+
+    m = DHNModel()
+
+    optim = tlx.optimizers.Adam(lr=args.lr, weight_decay=args.drop_rate)
+    train_weights = m.trainable_weights
+
+    net_with_loss = Loss(m, loss_fn=tlx.losses.sigmoid_cross_entropy)
+    net_with_train = TrainOneStep(net_with_loss, optim, train_weights)
+
+    tra_auc_metric = AUCMetric()
+    val_auc_metric = AUCMetric()
+    test_auc_metric = AUCMetric()
+
+    for epoch in range(args.n_epoch):
+        print("-----Epoch {}/{}-----".format(epoch + 1, args.n_epoch))
 
-EPOCH = 100
-BATCH_SIZE = 32
+        # train
+        m.set_train()
+        tra_batch_A_fea, tra_batch_B_fea, tra_batch_y = batch_data(G_train, BATCH_SIZE).__next__()
+        tra_out = m(tra_batch_A_fea, tra_batch_B_fea, tra_batch_y)
 
-tra_auc_cul = tf.keras.metrics.AUC()
-val_auc_cul = tf.keras.metrics.AUC()
-test_auc_cul = tf.keras.metrics.AUC()
+        data = {
+            "n1": tra_batch_A_fea,
+            "n2": tra_batch_B_fea,
+            "label": tra_batch_y
+        }
 
-for epoch in range(EPOCH):
-    print("-----Epoch {}/{}-----".format(epoch + 1, EPOCH))
+        tra_loss = net_with_train(data, tra_batch_y)
+        tra_auc_metric.update_state(y_true=tra_batch_y, y_pred=tlx.sigmoid(tra_out).detach().numpy())
+        tra_auc = tra_auc_metric.result()
+        print('train: ', tra_loss, tra_auc)
 
-    # train
-    m.set_train()
-    tra_batch_A_fea, tra_batch_B_fea, tra_batch_y = batch_data(G_train, BATCH_SIZE).__next__()
-    tra_out = m(tra_batch_A_fea, tra_batch_B_fea, tra_batch_y)
+        # val
+        m.set_eval()
+        val_batch_A_fea, val_batch_B_fea, val_batch_y = batch_data(G_val, BATCH_SIZE).__next__()
+        val_out = m(val_batch_A_fea, val_batch_B_fea, val_batch_y)
 
-    data = {
-        "n1": tra_batch_A_fea,
-        "n2": tra_batch_B_fea,
-        "label": tra_batch_y
-    }
+        val_loss = tlx.losses.sigmoid_cross_entropy(output=val_out, target=tlx.convert_to_tensor(val_batch_y))
+        val_auc_metric.update_state(y_true=val_batch_y, y_pred=tlx.sigmoid(val_out).detach().numpy())
+        val_auc = val_auc_metric.result()
+        print("val: ", val_loss.item(), val_auc)
 
-    tra_loss = net_with_train(data, tra_batch_y)
-    tra_auc_cul.update_state(y_true=tra_batch_y, y_pred=tlx.sigmoid(tra_out).detach().numpy())
-    tra_auc = tra_auc_cul.result().numpy()
-    print('train: ', tra_loss, tra_auc)
+        # test
+        test_batch_A_fea, test_batch_B_fea, test_batch_y = batch_data(G_test, BATCH_SIZE).__next__()
+        test_out = m(test_batch_A_fea, test_batch_B_fea, test_batch_y)
 
-    # val
-    m.set_eval()
-    val_batch_A_fea, val_batch_B_fea, val_batch_y = batch_data(G_val, BATCH_SIZE).__next__()
-    val_out = m(val_batch_A_fea, val_batch_B_fea, val_batch_y)
+        test_loss = tlx.losses.sigmoid_cross_entropy(output=test_out, target=tlx.convert_to_tensor(test_batch_y))
+        test_auc_metric.update_state(y_true=test_batch_y, y_pred=tlx.sigmoid(test_out).detach().numpy())
+        test_auc = test_auc_metric.result()
+        print("test: ", test_loss.item(), test_auc)
 
-    val_loss = tlx.losses.sigmoid_cross_entropy(output=val_out, target=tlx.convert_to_tensor(val_batch_y))
-    val_auc_cul.update_state(y_true=val_batch_y, y_pred=tlx.sigmoid(val_out).detach().numpy())
-    val_auc = val_auc_cul.result().numpy()
-    print("val: ", val_loss.item(), val_auc)
 
-    # test
-    test_batch_A_fea, test_batch_B_fea, test_batch_y = batch_data(G_test, BATCH_SIZE).__next__()
-    test_out = m(test_batch_A_fea, test_batch_B_fea, test_batch_y)
+if __name__ == '__main__':
+    # parameters setting
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--test_ratio", type=float, default=0.3, help="ratio of dividing the data set")
+    parser.add_argument("--one_hot", type=bool, default=True, help="use one-hot encoding")
+    parser.add_argument("--lr", type=float, default=0.001, help="learnin rate")
+    parser.add_argument("--n_epoch", type=int, default=100, help="number of epoch")
+    parser.add_argument("--drop_rate", type=float, default=0.01, help="drop_rate")
+    parser.add_argument('--dataset', type=str, default='acm', help='dataset')
+    parser.add_argument("--dataset_path", type=str, default=r"MA.txt")
 
-    test_loss = tlx.losses.sigmoid_cross_entropy(output=test_out, target=tlx.convert_to_tensor(test_batch_y))
-    test_auc_cul.update_state(y_true=test_batch_y, y_pred=tlx.sigmoid(test_out).detach().numpy())
-    test_auc = test_auc_cul.result().numpy()
-    print("test: ", test_loss.item(), test_auc)
+    args = parser.parse_args()
+    main(args)
diff --git a/gammagl/layers/conv/dhn_conv.py b/gammagl/layers/conv/dhn_conv.py
new file mode 100644
index 00000000..dc651a6a
--- /dev/null
+++ b/gammagl/layers/conv/dhn_conv.py
@@ -0,0 +1,90 @@
+import tensorlayerx as tlx
+from gammagl.layers.conv import MessagePassing
+
+type2idx = {
+    'M': 0,
+    'A': 1,
+    # 'C': 2,
+    # 'T': 3
+}
+
+NODE_TYPE = len(type2idx)
+K_HOP = 2
+
+NUM_FEA = (K_HOP + 2) * 4 + NODE_TYPE
+NUM_NEIGHBOR = 5
+BATCH_SIZE=32
+
+
+class DHNConv(MessagePassing):
+    def __init__(self):
+        super().__init__()
+        self.lin1 = tlx.nn.Linear(in_features=2*NUM_FEA, out_features=2*BATCH_SIZE, act=tlx.nn.ELU(), W_init="xavier_uniform")
+        self.lin2 = tlx.nn.Linear(in_features=2*BATCH_SIZE+NUM_FEA, out_features=2*BATCH_SIZE, act=tlx.nn.ELU(), W_init="xavier_uniform")
+        self.lin3 = tlx.nn.Linear(in_features=2*BATCH_SIZE, out_features=2*BATCH_SIZE, act=tlx.nn.ELU(), W_init="xavier_uniform")
+
+    def aggregate(self, msg, edge_index, num_nodes=None, aggr=None):
+        if len(msg.shape)==4:
+            return tlx.reduce_mean(msg, axis=2)
+        elif len(msg.shape)==3:
+            return tlx.reduce_mean(msg, axis=1)
+
+    def forward(self, fea):
+        node = tlx.convert_to_tensor(fea[:, :NUM_FEA])
+
+        # Extract neigh1 and neigh2
+        neigh1 = tlx.convert_to_tensor(fea[:, NUM_FEA:NUM_FEA * (NUM_NEIGHBOR + 1)])
+        neigh1 = tlx.reshape(neigh1, [-1, NUM_NEIGHBOR, NUM_FEA])
+
+        neigh2 = tlx.convert_to_tensor(fea[:, NUM_FEA * (NUM_NEIGHBOR + 1):])
+        neigh2 = tlx.reshape(neigh2, [-1, NUM_NEIGHBOR, NUM_NEIGHBOR, NUM_FEA])
+
+        # aggregate on neigh2
+        neigh2_agg = self.propagate(edge_index=None, x=neigh2)
+
+        # connect the node to the aggregated neigh2
+        tmp = tlx.concat(
+            [neigh1, neigh2_agg],
+            axis=2
+        )
+
+        # transform into a two-dimensional tensor
+        flattened_tmp = tmp.view(2*BATCH_SIZE * NUM_NEIGHBOR, 2*NUM_FEA)
+        flattened_output = self.lin1(flattened_tmp)
+        # Reshape the output tensor to a three-dimensional shape
+        tmp = flattened_output.view(2*BATCH_SIZE, NUM_NEIGHBOR, 2*BATCH_SIZE)
+
+        # Aggregate node and tmp
+        emb = tlx.concat(
+            [
+                node, self.propagate(edge_index=None,x=tmp)
+            ],
+            axis=1
+        )
+
+        # Through other linear layers
+        emb = self.lin2(emb)
+        emb = self.lin3(emb)
+
+        return emb
+
+    def message(self, x, edge_index, edge_weight=None):
+        return x
+
+
+class DHNModel(tlx.nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.dhn1 = DHNConv()
+        self.dhn2 = DHNConv()
+        self.lin1 = tlx.nn.Linear(in_features=4*BATCH_SIZE, out_features=BATCH_SIZE, act=tlx.nn.ELU(), W_init="xavier_uniform")
+        self.lin2 = tlx.nn.Linear(in_features=BATCH_SIZE, out_features=1, act=tlx.nn.ELU(), W_init="xavier_uniform")
+
+    def forward(self, n1, n2, label):
+        n1_emb = self.dhn1(n1)
+        n2_emb = self.dhn2(n2)
+
+        pred = self.lin1(tlx.concat([n1_emb, n2_emb], axis=1))
+        pred = self.lin2(pred)
+
+        return pred
diff --git a/gammagl/models/dhn.py b/gammagl/models/dhn.py
deleted file mode 100644
index af8bd069..00000000
--- a/gammagl/models/dhn.py
+++ /dev/null
@@ -1,82 +0,0 @@
-import tensorlayerx as tlx
-
-type2idx = {
-    'M': 0,
-    'A': 1,
-    # 'C': 2,
-    # 'T': 3
-}
-
-NODE_TYPE = len(type2idx)
-K_HOP = 2
-
-NUM_FEA = (K_HOP + 2) * 4 + NODE_TYPE
-NUM_NEIGHBOR = 5
-
-
-class DHN(tlx.nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.lin1 = tlx.nn.Linear(in_features=36, out_features=64, act=tlx.nn.ELU(), W_init="xavier_uniform")
-        self.lin2 = tlx.nn.Linear(in_features=82, out_features=64, act=tlx.nn.ELU(), W_init="xavier_uniform")
-        self.lin3 = tlx.nn.Linear(in_features=64, out_features=64, act=tlx.nn.ELU(), W_init="xavier_uniform")
-
-    def forward(self, fea):
-        node = tlx.convert_to_tensor(fea[:, :NUM_FEA])
-
-        # 提取neigh1和neigh2
-        neigh1 = tlx.convert_to_tensor(fea[:, NUM_FEA:NUM_FEA * (NUM_NEIGHBOR + 1)])
-        neigh1 = tlx.reshape(neigh1, [-1, NUM_NEIGHBOR, NUM_FEA])
-
-        neigh2 = tlx.convert_to_tensor(fea[:, NUM_FEA * (NUM_NEIGHBOR + 1):])
-        neigh2 = tlx.reshape(neigh2, [-1, NUM_NEIGHBOR, NUM_NEIGHBOR, NUM_FEA])
-
-        # 对neigh2进行聚合
-        neigh2_agg = tlx.reduce_mean(neigh2, axis=2)
-
-        # 连接node和聚合后的neigh2
-        tmp = tlx.concat(
-            [neigh1, neigh2_agg],
-            axis=2
-        )
-
-        # 将tmp重塑为二维张量
-        flattened_tmp = tmp.view(64 * 5, 36)  
-
-        # 将二维张量传递给线性层
-        flattened_output = self.lin1(flattened_tmp)  
-
-         # 将输出张量重新整形为三维形状
-        tmp = flattened_output.view(64, 5, 64)  
-
-        # 对node和tmp进行聚合
-        emb = tlx.concat(
-            [
-                node, tlx.reduce_mean(tmp, axis=1)
-            ],
-            axis=1
-        )
-
-        # 通过其他线性层
-        emb = self.lin2(emb)
-        emb = self.lin3(emb)
-
-        return emb
-
-
-class DHNModel(tlx.nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.dhn1 = DHN()
-        self.dhn2 = DHN()
-        self.lin1 = tlx.nn.Linear(in_features=128, out_features=32, act=tlx.nn.ELU(), W_init="xavier_uniform")
-        self.lin2 = tlx.nn.Linear(in_features=32, out_features=1, act=tlx.nn.ELU(), W_init="xavier_uniform")
-
-    def forward(self, n1, n2, label):
-        n1_emb = self.dhn1(n1)
-        n2_emb = self.dhn2(n2)
-
-        pred = self.lin1(tlx.concat([n1_emb, n2_emb], axis=1))
-        pred = self.lin2(pred)
-
-        return pred

From 575eec57dbdba761589f048eb62810965b1f33a5 Mon Sep 17 00:00:00 2001
From: liberty <2700472871@qq.com>
Date: Mon, 1 Jul 2024 21:05:51 +0800
Subject: [PATCH 3/8] Add dataset and test code

---
 examples/dhn/MA.txt                       | 7071 ---------------------
 gammagl/datasets/acm4dhn.py               |    0
 gammagl/models/dhn.py                     |    0
 gammagl/utils/simple_path.py              |    0
 tests/utils/test_find_all_simple_paths.py |    0
 5 files changed, 7071 deletions(-)
 delete mode 100644 examples/dhn/MA.txt
 create mode 100644 gammagl/datasets/acm4dhn.py
 create mode 100644 gammagl/models/dhn.py
 create mode 100644 gammagl/utils/simple_path.py
 create mode 100644 tests/utils/test_find_all_simple_paths.py

diff --git a/examples/dhn/MA.txt b/examples/dhn/MA.txt
deleted file mode 100644
index 3d275250..00000000
--- a/examples/dhn/MA.txt
+++ /dev/null
@@ -1,7071 +0,0 @@
-M0 A1177
-M1 A742
-M2 A478
-M3 A713
-M5 A955
-M6 A698
-M8 A99
-M9 A996
-M10 A580
-M11 A1235
-M12 A292
-M13 A742
-M14 A742
-M15 A580
-M16 A418
-M17 A99
-M18 A742
-M19 A1009
-M20 A188
-M21 A562
-M22 A753
-M23 A188
-M24 A962
-M25 A1043
-M26 A219
-M27 A1040
-M28 A419
-M29 A851
-M30 A594
-M31 A698
-M32 A1040
-M33 A742
-M34 A1054
-M36 A148
-M37 A887
-M39 A562
-M40 A340
-M41 A778
-M42 A1117
-M44 A826
-M45 A816
-M46 A1212
-M47 A287
-M48 A856
-M49 A794
-M50 A219
-M51 A641
-M52 A687
-M53 A148
-M54 A566
-M58 A53
-M59 A187
-M60 A284
-M61 A964
-M62 A349
-M63 A478
-M64 A1213
-M65 A287
-M66 A826
-M67 A53
-M68 A423
-M69 A1040
-M70 A1018
-M71 A1009
-M72 A19
-M73 A1009
-M74 A842
-M75 A1050
-M76 A142
-M77 A338
-M78 A423
-M79 A974
-M80 A1040
-M83 A70
-M84 A1341
-M86 A974
-M87 A1350
-M88 A1310
-M91 A224
-M92 A865
-M93 A908
-M94 A650
-M95 A123
-M96 A550
-M97 A219
-M99 A188
-M100 A1024
-M101 A178
-M102 A287
-M103 A287
-M104 A903
-M105 A842
-M106 A742
-M107 A411
-M108 A28
-M109 A698
-M110 A6
-M111 A645
-M112 A707
-M113 A808
-M114 A39
-M115 A39
-M116 A1009
-M117 A1009
-M118 A742
-M119 A937
-M120 A826
-M121 A1007
-M122 A591
-M123 A1054
-M126 A3
-M127 A99
-M128 A713
-M129 A224
-M130 A99
-M134 A742
-M135 A1050
-M136 A808
-M137 A478
-M141 A831
-M142 A349
-M144 A1054
-M145 A1054
-M146 A147
-M147 A178
-M148 A445
-M149 A702
-M150 A1381
-M151 A245
-M152 A1009
-M153 A964
-M155 A1050
-M157 A1371
-M158 A1050
-M159 A826
-M160 A99
-M161 A698
-M162 A908
-M163 A908
-M164 A176
-M165 A735
-M166 A19
-M168 A349
-M169 A550
-M171 A1035
-M172 A272
-M173 A248
-M174 A39
-M175 A591
-M176 A147
-M178 A1003
-M179 A219
-M181 A742
-M182 A856
-M183 A351
-M184 A133
-M185 A1310
-M186 A287
-M187 A39
-M188 A274
-M189 A1050
-M190 A1009
-M191 A1382
-M192 A133
-M193 A808
-M194 A519
-M195 A1073
-M197 A1009
-M200 A591
-M202 A1073
-M203 A1117
-M205 A742
-M207 A287
-M208 A224
-M209 A1178
-M210 A1054
-M211 A1024
-M212 A1040
-M213 A419
-M214 A1282
-M215 A1263
-M216 A317
-M217 A671
-M218 A21
-M219 A641
-M220 A1050
-M221 A829
-M222 A1035
-M224 A903
-M225 A351
-M226 A1199
-M227 A539
-M228 A1234
-M229 A351
-M230 A1373
-M231 A70
-M232 A592
-M234 A1050
-M235 A9
-M236 A1039
-M237 A1039
-M238 A1345
-M239 A1054
-M240 A1039
-M241 A1117
-M243 A903
-M245 A773
-M249 A1154
-M250 A591
-M251 A36
-M252 A984
-M253 A389
-M254 A178
-M255 A178
-M256 A156
-M257 A219
-M258 A555
-M259 A478
-M263 A1345
-M264 A1043
-M265 A1054
-M267 A1145
-M268 A566
-M269 A245
-M270 A962
-M271 A1213
-M272 A1009
-M274 A1009
-M275 A419
-M276 A643
-M277 A1340
-M278 A351
-M280 A974
-M282 A903
-M283 A318
-M284 A1050
-M285 A1073
-M286 A1125
-M287 A1046
-M288 A572
-M289 A742
-M290 A1179
-M291 A1089
-M292 A1179
-M293 A577
-M294 A36
-M295 A464
-M296 A219
-M297 A1199
-M298 A507
-M299 A816
-M300 A903
-M302 A1339
-M303 A1054
-M304 A490
-M305 A742
-M306 A1177
-M307 A219
-M308 A219
-M309 A671
-M310 A1027
-M311 A224
-M312 A1281
-M314 A36
-M315 A338
-M316 A29
-M318 A947
-M319 A280
-M320 A99
-M321 A861
-M322 A351
-M324 A543
-M325 A36
-M326 A219
-M327 A826
-M329 A1367
-M330 A466
-M331 A29
-M332 A1141
-M334 A916
-M335 A418
-M336 A783
-M337 A1265
-M338 A1374
-M339 A263
-M340 A962
-M341 A340
-M342 A133
-M344 A899
-M345 A910
-M346 A1326
-M347 A245
-M348 A546
-M349 A1110
-M351 A441
-M352 A1009
-M353 A224
-M354 A1179
-M356 A566
-M357 A886
-M358 A1314
-M359 A482
-M360 A1048
-M361 A219
-M363 A1040
-M364 A964
-M365 A245
-M366 A1040
-M368 A425
-M369 A473
-M370 A1050
-M371 A982
-M372 A418
-M373 A1163
-M374 A953
-M375 A550
-M376 A58
-M377 A434
-M378 A441
-M379 A908
-M380 A1073
-M381 A528
-M382 A178
-M383 A961
-M384 A243
-M385 A591
-M386 A961
-M387 A566
-M388 A419
-M389 A1154
-M390 A300
-M391 A919
-M392 A291
-M393 A340
-M394 A831
-M395 A1024
-M396 A224
-M397 A591
-M398 A351
-M399 A566
-M400 A178
-M401 A1141
-M402 A1345
-M405 A1179
-M406 A466
-M407 A1131
-M408 A156
-M409 A1009
-M410 A714
-M411 A197
-M412 A1393
-M413 A156
-M414 A506
-M415 A36
-M416 A335
-M417 A954
-M420 A808
-M421 A155
-M422 A870
-M423 A246
-M424 A589
-M425 A148
-M426 A32
-M427 A414
-M428 A95
-M429 A982
-M430 A259
-M431 A565
-M432 A885
-M433 A547
-M434 A1040
-M435 A935
-M436 A9
-M437 A816
-M438 A1050
-M439 A287
-M440 A123
-M441 A671
-M442 A245
-M443 A950
-M444 A1345
-M445 A9
-M446 A245
-M447 A982
-M448 A423
-M449 A1345
-M450 A1048
-M451 A1127
-M452 A219
-M453 A1393
-M454 A1213
-M455 A903
-M456 A982
-M457 A224
-M458 A982
-M460 A1345
-M461 A861
-M462 A831
-M463 A1179
-M464 A591
-M465 A419
-M467 A1305
-M468 A1127
-M469 A246
-M470 A178
-M471 A566
-M472 A411
-M474 A861
-M475 A70
-M476 A1179
-M478 A1345
-M481 A753
-M482 A242
-M483 A1054
-M484 A1214
-M486 A982
-M487 A351
-M488 A1039
-M490 A808
-M491 A245
-M492 A993
-M493 A363
-M496 A691
-M498 A1024
-M499 A1000
-M500 A593
-M501 A1373
-M503 A667
-M505 A1047
-M508 A593
-M509 A1131
-M510 A490
-M512 A831
-M513 A698
-M514 A197
-M515 A1024
-M516 A340
-M517 A1305
-M520 A448
-M521 A590
-M522 A1009
-M523 A1291
-M524 A445
-M525 A687
-M526 A1054
-M527 A133
-M528 A903
-M529 A1314
-M530 A982
-M531 A988
-M532 A248
-M534 A9
-M535 A1345
-M536 A224
-M538 A961
-M540 A831
-M541 A577
-M542 A391
-M544 A36
-M545 A1188
-M546 A245
-M547 A328
-M548 A580
-M549 A982
-M550 A1345
-M551 A245
-M552 A458
-M553 A233
-M554 A580
-M556 A351
-M557 A566
-M558 A1290
-M560 A691
-M561 A546
-M562 A964
-M564 A1350
-M565 A903
-M566 A139
-M568 A550
-M569 A982
-M570 A645
-M571 A29
-M572 A936
-M574 A641
-M576 A968
-M577 A363
-M579 A851
-M580 A284
-M581 A904
-M582 A970
-M583 A53
-M584 A246
-M585 A1373
-M587 A701
-M588 A539
-M589 A591
-M590 A36
-M591 A245
-M592 A317
-M594 A1050
-M595 A851
-M596 A963
-M597 A423
-M598 A1350
-M599 A908
-M600 A443
-M601 A773
-M602 A1345
-M603 A1103
-M604 A286
-M605 A1048
-M606 A66
-M607 A90
-M609 A1179
-M610 A580
-M611 A178
-M612 A349
-M613 A808
-M614 A36
-M615 A571
-M618 A245
-M620 A1272
-M625 A99
-M627 A1352
-M628 A363
-M631 A39
-M632 A577
-M633 A590
-M635 A419
-M638 A803
-M639 A197
-M641 A219
-M642 A590
-M643 A754
-M644 A907
-M646 A968
-M648 A224
-M650 A1310
-M651 A567
-M652 A9
-M653 A338
-M654 A1127
-M655 A224
-M656 A1282
-M657 A1052
-M658 A1054
-M659 A224
-M660 A29
-M661 A808
-M662 A1040
-M663 A336
-M664 A977
-M665 A363
-M666 A352
-M667 A245
-M668 A903
-M669 A246
-M671 A1050
-M672 A229
-M673 A1009
-M674 A713
-M675 A1117
-M676 A1365
-M677 A1215
-M678 A1141
-M679 A816
-M682 A885
-M683 A178
-M684 A550
-M686 A245
-M687 A313
-M690 A645
-M691 A338
-M692 A566
-M693 A928
-M694 A710
-M695 A1380
-M696 A903
-M697 A585
-M698 A546
-M699 A566
-M701 A643
-M702 A550
-M707 A29
-M708 A571
-M709 A908
-M710 A577
-M711 A1121
-M712 A224
-M714 A1339
-M715 A304
-M716 A351
-M718 A122
-M719 A681
-M721 A661
-M722 A1179
-M724 A816
-M725 A216
-M726 A363
-M728 A1050
-M729 A1120
-M731 A910
-M732 A903
-M733 A245
-M734 A197
-M735 A974
-M736 A1050
-M737 A36
-M738 A308
-M739 A1050
-M740 A1272
-M742 A351
-M743 A706
-M745 A588
-M746 A1179
-M747 A1040
-M748 A821
-M749 A817
-M751 A935
-M753 A698
-M754 A253
-M755 A778
-M760 A118
-M761 A287
-M763 A631
-M764 A1108
-M766 A567
-M767 A9
-M768 A1256
-M769 A826
-M770 A383
-M771 A1122
-M772 A351
-M773 A964
-M774 A246
-M778 A713
-M779 A1135
-M781 A103
-M783 A577
-M784 A577
-M785 A245
-M786 A1040
-M787 A351
-M788 A831
-M789 A1048
-M793 A582
-M794 A99
-M795 A572
-M796 A974
-M798 A958
-M799 A326
-M801 A188
-M802 A603
-M803 A279
-M804 A550
-M805 A691
-M806 A974
-M807 A742
-M808 A816
-M809 A133
-M811 A910
-M813 A772
-M814 A123
-M815 A1072
-M817 A260
-M818 A1206
-M821 A1120
-M822 A928
-M823 A54
-M824 A1141
-M827 A1127
-M828 A982
-M829 A546
-M830 A1382
-M831 A903
-M832 A826
-M834 A1373
-M835 A36
-M836 A224
-M837 A98
-M838 A577
-M840 A119
-M841 A445
-M842 A1009
-M843 A691
-M844 A1245
-M845 A224
-M846 A1157
-M847 A141
-M848 A1179
-M849 A506
-M850 A1179
-M851 A1009
-M852 A421
-M853 A1199
-M854 A903
-M855 A151
-M856 A1040
-M858 A865
-M859 A363
-M860 A245
-M861 A75
-M862 A284
-M863 A399
-M864 A19
-M865 A423
-M866 A177
-M868 A577
-M869 A1111
-M871 A886
-M872 A1290
-M873 A808
-M874 A351
-M876 A1382
-M877 A133
-M878 A1127
-M879 A577
-M880 A908
-M881 A214
-M882 A242
-M886 A147
-M887 A1038
-M889 A605
-M890 A573
-M891 A363
-M892 A844
-M893 A1009
-M894 A363
-M895 A197
-M896 A263
-M898 A1395
-M900 A156
-M901 A826
-M902 A762
-M903 A1213
-M904 A1345
-M905 A742
-M906 A1024
-M907 A908
-M908 A855
-M909 A1340
-M910 A1040
-M911 A219
-M912 A1376
-M913 A1393
-M914 A51
-M915 A123
-M916 A1142
-M917 A650
-M918 A1117
-M920 A577
-M923 A326
-M924 A1166
-M925 A1131
-M926 A39
-M929 A996
-M930 A1050
-M931 A1157
-M932 A10
-M933 A591
-M934 A566
-M935 A1345
-M937 A246
-M938 A550
-M939 A349
-M940 A178
-M941 A123
-M942 A36
-M943 A246
-M944 A29
-M945 A742
-M946 A1179
-M947 A739
-M948 A371
-M949 A974
-M950 A903
-M951 A123
-M953 A383
-M954 A1048
-M955 A1089
-M956 A831
-M957 A1235
-M958 A419
-M959 A36
-M960 A1039
-M964 A193
-M965 A284
-M966 A317
-M967 A1104
-M968 A196
-M969 A351
-M970 A702
-M972 A842
-M973 A977
-M974 A250
-M975 A656
-M977 A1127
-M978 A1245
-M979 A1235
-M980 A595
-M981 A222
-M982 A962
-M983 A1040
-M984 A1316
-M987 A1179
-M989 A1120
-M990 A219
-M991 A1192
-M993 A698
-M995 A1199
-M996 A562
-M997 A351
-M998 A383
-M999 A628
-M1000 A954
-M1001 A245
-M1003 A1290
-M1004 A977
-M1005 A1043
-M1006 A1345
-M1008 A1373
-M1009 A1048
-M1010 A550
-M1011 A1235
-M1012 A337
-M1013 A419
-M1014 A250
-M1015 A1024
-M1017 A1039
-M1018 A1043
-M1019 A1287
-M1020 A778
-M1021 A70
-M1022 A1043
-M1025 A66
-M1026 A908
-M1027 A1032
-M1028 A591
-M1029 A1141
-M1030 A9
-M1031 A1310
-M1032 A903
-M1033 A520
-M1034 A816
-M1036 A619
-M1037 A831
-M1038 A542
-M1039 A566
-M1040 A375
-M1041 A1054
-M1042 A1179
-M1043 A885
-M1045 A1310
-M1046 A713
-M1048 A1235
-M1049 A1054
-M1050 A691
-M1051 A903
-M1052 A826
-M1053 A419
-M1054 A961
-M1055 A1131
-M1058 A296
-M1059 A57
-M1060 A363
-M1063 A246
-M1065 A246
-M1066 A197
-M1067 A443
-M1068 A539
-M1069 A1179
-M1070 A1215
-M1072 A1139
-M1073 A697
-M1074 A419
-M1075 A586
-M1076 A1076
-M1077 A490
-M1078 A243
-M1079 A1179
-M1080 A670
-M1081 A9
-M1082 A566
-M1083 A482
-M1084 A1054
-M1085 A1179
-M1086 A996
-M1087 A987
-M1088 A14
-M1089 A963
-M1090 A349
-M1091 A320
-M1092 A916
-M1093 A578
-M1094 A51
-M1095 A1235
-M1096 A742
-M1097 A363
-M1098 A223
-M1100 A490
-M1102 A236
-M1103 A1141
-M1104 A473
-M1105 A1272
-M1106 A714
-M1107 A208
-M1108 A1384
-M1109 A306
-M1110 A1353
-M1111 A903
-M1112 A561
-M1113 A81
-M1114 A219
-M1115 A245
-M1116 A1373
-M1118 A864
-M1120 A86
-M1122 A95
-M1123 A1046
-M1124 A351
-M1125 A808
-M1126 A817
-M1127 A318
-M1128 A338
-M1129 A954
-M1130 A903
-M1131 A588
-M1134 A194
-M1137 A740
-M1138 A566
-M1139 A908
-M1140 A577
-M1141 A363
-M1142 A338
-M1144 A340
-M1145 A1038
-M1146 A1073
-M1147 A335
-M1148 A1345
-M1149 A9
-M1151 A1050
-M1153 A816
-M1154 A349
-M1155 A246
-M1156 A894
-M1157 A831
-M1158 A920
-M1159 A313
-M1161 A1382
-M1162 A9
-M1163 A340
-M1164 A1065
-M1166 A501
-M1167 A70
-M1168 A577
-M1170 A903
-M1171 A19
-M1172 A948
-M1173 A910
-M1176 A679
-M1177 A1146
-M1179 A389
-M1181 A245
-M1183 A1300
-M1184 A209
-M1185 A287
-M1186 A1117
-M1187 A993
-M1188 A1365
-M1189 A974
-M1190 A1376
-M1191 A363
-M1192 A1235
-M1195 A224
-M1196 A925
-M1201 A1365
-M1203 A1127
-M1204 A714
-M1205 A566
-M1206 A831
-M1207 A1048
-M1208 A1024
-M1209 A577
-M1210 A722
-M1211 A1151
-M1212 A1023
-M1213 A861
-M1214 A546
-M1215 A698
-M1216 A1097
-M1217 A1275
-M1218 A1018
-M1219 A698
-M1220 A336
-M1221 A111
-M1222 A223
-M1224 A224
-M1225 A559
-M1227 A768
-M1228 A762
-M1230 A580
-M1231 A550
-M1232 A419
-M1233 A826
-M1234 A1235
-M1237 A271
-M1238 A695
-M1239 A450
-M1240 A742
-M1242 A607
-M1243 A645
-M1246 A351
-M1247 A1040
-M1248 A808
-M1250 A303
-M1251 A876
-M1253 A1329
-M1255 A390
-M1256 A1106
-M1257 A903
-M1258 A471
-M1263 A1235
-M1264 A143
-M1266 A768
-M1267 A419
-M1268 A697
-M1269 A224
-M1271 A997
-M1272 A66
-M1273 A861
-M1274 A754
-M1275 A1274
-M1276 A119
-M1277 A257
-M1278 A1039
-M1280 A767
-M1281 A577
-M1282 A1381
-M1283 A1232
-M1284 A70
-M1285 A577
-M1286 A46
-M1287 A556
-M1288 A197
-M1290 A762
-M1291 A903
-M1292 A363
-M1293 A1043
-M1295 A1048
-M1296 A133
-M1298 A92
-M1299 A217
-M1301 A259
-M1303 A681
-M1304 A490
-M1305 A1073
-M1306 A652
-M1307 A1007
-M1308 A704
-M1310 A844
-M1311 A631
-M1312 A754
-M1313 A521
-M1314 A1035
-M1317 A947
-M1318 A196
-M1319 A104
-M1320 A133
-M1321 A392
-M1322 A117
-M1325 A1035
-M1326 A593
-M1328 A471
-M1332 A1024
-M1333 A968
-M1334 A709
-M1336 A1395
-M1337 A577
-M1338 A850
-M1339 A1
-M1340 A908
-M1341 A65
-M1342 A1345
-M1344 A78
-M1345 A39
-M1348 A1179
-M1349 A14
-M1353 A817
-M1354 A742
-M1355 A698
-M1356 A55
-M1357 A648
-M1358 A340
-M1359 A444
-M1360 A1373
-M1362 A107
-M1363 A1048
-M1365 A617
-M1366 A546
-M1368 A363
-M1369 A317
-M1370 A589
-M1371 A698
-M1373 A99
-M1374 A1025
-M1375 A245
-M1377 A9
-M1378 A885
-M1379 A1179
-M1380 A956
-M1383 A750
-M1386 A9
-M1387 A242
-M1388 A1373
-M1389 A562
-M1390 A741
-M1391 A275
-M1392 A808
-M1393 A312
-M1394 A327
-M1395 A678
-M1396 A1007
-M1397 A123
-M1399 A628
-M1400 A671
-M1401 A1232
-M1402 A252
-M1405 A741
-M1406 A1235
-M1407 A351
-M1408 A714
-M1409 A425
-M1410 A1393
-M1411 A577
-M1412 A1235
-M1414 A1007
-M1415 A51
-M1416 A1235
-M1417 A705
-M1418 A328
-M1419 A654
-M1420 A742
-M1421 A847
-M1422 A219
-M1423 A880
-M1424 A1035
-M1425 A400
-M1426 A1157
-M1427 A1018
-M1428 A224
-M1429 A1179
-M1431 A591
-M1433 A591
-M1434 A1367
-M1436 A742
-M1437 A1141
-M1438 A1207
-M1439 A698
-M1440 A1157
-M1441 A379
-M1443 A1124
-M1445 A1111
-M1449 A742
-M1450 A38
-M1451 A815
-M1452 A1039
-M1453 A219
-M1454 A448
-M1456 A709
-M1457 A308
-M1458 A691
-M1460 A1010
-M1461 A1312
-M1462 A1235
-M1463 A577
-M1464 A102
-M1467 A243
-M1471 A1040
-M1472 A630
-M1474 A1149
-M1475 A338
-M1476 A87
-M1477 A318
-M1478 A816
-M1479 A762
-M1480 A691
-M1483 A714
-M1484 A618
-M1485 A180
-M1486 A1235
-M1487 A201
-M1488 A109
-M1489 A107
-M1490 A178
-M1492 A211
-M1493 A389
-M1495 A760
-M1496 A338
-M1497 A1056
-M1498 A716
-M1500 A709
-M1501 A352
-M1502 A1213
-M1503 A139
-M1505 A1104
-M1508 A1341
-M1510 A14
-M1512 A1111
-M1513 A202
-M1514 A1054
-M1516 A903
-M1519 A742
-M1520 A281
-M1521 A908
-M1522 A858
-M1523 A1014
-M1524 A1050
-M1525 A636
-M1526 A1022
-M1530 A724
-M1533 A582
-M1534 A816
-M1535 A133
-M1538 A156
-M1539 A652
-M1540 A418
-M1541 A935
-M1542 A622
-M1543 A1048
-M1545 A995
-M1546 A349
-M1548 A577
-M1549 A1024
-M1551 A729
-M1552 A641
-M1553 A59
-M1554 A441
-M1555 A432
-M1556 A962
-M1558 A250
-M1559 A816
-M1560 A219
-M1561 A1008
-M1562 A903
-M1564 A326
-M1565 A1163
-M1567 A367
-M1568 A1048
-M1569 A1263
-M1571 A566
-M1573 A857
-M1574 A70
-M1575 A714
-M1576 A1179
-M1577 A1055
-M1578 A1048
-M1579 A808
-M1580 A64
-M1581 A546
-M1582 A1381
-M1584 A352
-M1585 A118
-M1586 A1345
-M1587 A1245
-M1588 A224
-M1589 A1127
-M1591 A566
-M1592 A383
-M1593 A880
-M1594 A196
-M1595 A571
-M1596 A661
-M1597 A6
-M1598 A243
-M1599 A327
-M1600 A9
-M1601 A1348
-M1603 A1308
-M1604 A977
-M1605 A577
-M1606 A383
-M1607 A1310
-M1610 A931
-M1611 A808
-M1612 A698
-M1613 A237
-M1615 A338
-M1616 A643
-M1617 A390
-M1618 A714
-M1619 A1157
-M1621 A863
-M1624 A243
-M1626 A645
-M1627 A249
-M1630 A964
-M1633 A367
-M1636 A908
-M1637 A993
-M1638 A856
-M1639 A929
-M1640 A1010
-M1642 A742
-M1645 A641
-M1646 A103
-M1647 A566
-M1648 A1163
-M1649 A277
-M1650 A18
-M1651 A29
-M1652 A1165
-M1654 A1319
-M1655 A577
-M1657 A1048
-M1660 A1390
-M1661 A224
-M1662 A1018
-M1665 A36
-M1666 A1007
-M1668 A349
-M1672 A1377
-M1673 A564
-M1674 A714
-M1675 A1179
-M1676 A560
-M1677 A722
-M1680 A1106
-M1681 A1007
-M1682 A1122
-M1683 A1141
-M1685 A46
-M1688 A641
-M1689 A197
-M1690 A317
-M1691 A245
-M1692 A1379
-M1694 A1376
-M1696 A112
-M1699 A826
-M1701 A1040
-M1703 A708
-M1705 A918
-M1707 A9
-M1708 A359
-M1711 A411
-M1712 A89
-M1713 A1179
-M1714 A577
-M1715 A28
-M1716 A197
-M1718 A349
-M1719 A376
-M1722 A178
-M1725 A816
-M1726 A908
-M1728 A858
-M1734 A589
-M1735 A843
-M1738 A1213
-M1739 A519
-M1741 A111
-M1742 A418
-M1743 A1179
-M1744 A588
-M1747 A591
-M1749 A566
-M1751 A9
-M1752 A714
-M1753 A742
-M1754 A996
-M1755 A1175
-M1756 A383
-M1757 A894
-M1759 A147
-M1760 A550
-M1762 A1048
-M1763 A740
-M1764 A577
-M1765 A530
-M1768 A1141
-M1770 A410
-M1771 A154
-M1772 A1171
-M1773 A985
-M1774 A908
-M1775 A473
-M1777 A566
-M1778 A419
-M1779 A1290
-M1780 A935
-M1783 A1051
-M1785 A399
-M1786 A452
-M1787 A816
-M1789 A321
-M1792 A1018
-M1794 A1024
-M1797 A1145
-M1798 A648
-M1801 A698
-M1803 A425
-M1804 A44
-M1805 A974
-M1806 A9
-M1807 A62
-M1808 A422
-M1811 A284
-M1812 A245
-M1813 A546
-M1814 A929
-M1815 A173
-M1816 A754
-M1820 A550
-M1821 A123
-M1822 A692
-M1824 A741
-M1825 A246
-M1826 A301
-M1827 A577
-M1828 A1272
-M1829 A577
-M1830 A1352
-M1832 A893
-M1833 A1089
-M1834 A590
-M1835 A1111
-M1837 A758
-M1838 A910
-M1839 A661
-M1840 A935
-M1841 A246
-M1844 A72
-M1845 A408
-M1846 A1110
-M1847 A546
-M1848 A36
-M1849 A974
-M1850 A66
-M1851 A880
-M1852 A1039
-M1853 A789
-M1854 A577
-M1857 A602
-M1859 A1317
-M1861 A881
-M1862 A641
-M1864 A1117
-M1865 A716
-M1867 A591
-M1868 A1050
-M1869 A977
-M1870 A519
-M1871 A419
-M1872 A311
-M1873 A14
-M1874 A419
-M1875 A562
-M1876 A856
-M1877 A414
-M1879 A266
-M1881 A1326
-M1882 A1066
-M1884 A577
-M1885 A375
-M1886 A934
-M1888 A756
-M1889 A584
-M1890 A748
-M1891 A1039
-M1892 A66
-M1893 A19
-M1895 A874
-M1896 A747
-M1898 A742
-M1899 A857
-M1900 A419
-M1901 A577
-M1903 A577
-M1904 A789
-M1905 A51
-M1906 A197
-M1907 A772
-M1908 A1046
-M1909 A698
-M1910 A240
-M1914 A149
-M1915 A1066
-M1917 A562
-M1918 A400
-M1919 A1350
-M1920 A340
-M1921 A51
-M1924 A19
-M1925 A714
-M1926 A349
-M1928 A267
-M1929 A1157
-M1933 A97
-M1935 A273
-M1937 A9
-M1939 A1048
-M1940 A367
-M1941 A856
-M1943 A634
-M1944 A1054
-M1945 A1179
-M1948 A349
-M1949 A1035
-M1952 A982
-M1958 A49
-M1959 A90
-M1961 A467
-M1962 A965
-M1965 A683
-M1966 A713
-M1967 A1155
-M1969 A376
-M1970 A645
-M1971 A56
-M1973 A964
-M1974 A389
-M1976 A688
-M1978 A713
-M1979 A996
-M1980 A777
-M1981 A338
-M1982 A687
-M1984 A93
-M1985 A691
-M1986 A329
-M1987 A237
-M1989 A250
-M1992 A841
-M1993 A808
-M1994 A724
-M1995 A1043
-M1998 A301
-M1999 A156
-M2000 A51
-M2003 A405
-M2005 A903
-M2007 A360
-M2008 A1186
-M2011 A539
-M2015 A209
-M2016 A1127
-M2017 A885
-M2018 A425
-M2019 A363
-M2020 A947
-M2021 A468
-M2022 A1039
-M2023 A382
-M2024 A245
-M2025 A513
-M2026 A29
-M2027 A340
-M2028 A628
-M2031 A9
-M2033 A789
-M2034 A133
-M2035 A695
-M2036 A863
-M2038 A284
-M2039 A1116
-M2040 A722
-M2041 A482
-M2042 A1235
-M2043 A1024
-M2046 A36
-M2047 A826
-M2048 A562
-M2049 A1043
-M2050 A982
-M2051 A566
-M2052 A1009
-M2053 A367
-M2054 A789
-M2055 A383
-M2056 A1373
-M2057 A1054
-M2058 A974
-M2059 A1054
-M2060 A1024
-M2062 A1039
-M2063 A918
-M2065 A349
-M2069 A767
-M2070 A464
-M2071 A910
-M2072 A924
-M2076 A973
-M2081 A1199
-M2082 A946
-M2083 A1052
-M2084 A546
-M2085 A36
-M2086 A977
-M2087 A222
-M2088 A1076
-M2090 A1112
-M2091 A513
-M2093 A1038
-M2094 A373
-M2095 A351
-M2096 A808
-M2099 A1073
-M2101 A350
-M2102 A743
-M2103 A46
-M2104 A29
-M2105 A974
-M2107 A974
-M2108 A861
-M2109 A816
-M2110 A202
-M2111 A1043
-M2113 A66
-M2115 A691
-M2118 A109
-M2121 A1010
-M2122 A51
-M2123 A870
-M2124 A577
-M2126 A826
-M2127 A1111
-M2128 A444
-M2129 A340
-M2132 A977
-M2133 A677
-M2134 A418
-M2136 A1127
-M2137 A287
-M2138 A43
-M2139 A1318
-M2140 A816
-M2141 A318
-M2144 A1339
-M2146 A641
-M2149 A287
-M2151 A117
-M2152 A566
-M2153 A876
-M2154 A651
-M2156 A655
-M2157 A1019
-M2158 A808
-M2160 A867
-M2161 A870
-M2162 A1089
-M2163 A428
-M2164 A1022
-M2165 A571
-M2166 A104
-M2167 A17
-M2168 A775
-M2170 A785
-M2171 A1172
-M2172 A1154
-M2173 A772
-M2174 A229
-M2175 A1109
-M2176 A636
-M2177 A742
-M2179 A1168
-M2180 A870
-M2181 A1157
-M2184 A1047
-M2189 A842
-M2190 A1068
-M2192 A151
-M2193 A300
-M2194 A191
-M2195 A1068
-M2196 A414
-M2198 A1127
-M2199 A591
-M2203 A324
-M2204 A178
-M2205 A531
-M2208 A1090
-M2209 A61
-M2212 A910
-M2214 A167
-M2217 A782
-M2218 A1111
-M2219 A630
-M2220 A1241
-M2221 A778
-M2222 A816
-M2223 A156
-M2224 A1089
-M2225 A1127
-M2226 A710
-M2230 A833
-M2231 A593
-M2232 A996
-M2234 A322
-M2236 A842
-M2237 A523
-M2239 A250
-M2242 A1336
-M2243 A556
-M2244 A196
-M2245 A80
-M2246 A690
-M2247 A1131
-M2248 A284
-M2249 A982
-M2250 A885
-M2251 A260
-M2253 A855
-M2255 A1040
-M2260 A1043
-M2262 A885
-M2263 A730
-M2267 A246
-M2268 A1319
-M2269 A453
-M2270 A1048
-M2273 A423
-M2274 A816
-M2275 A1213
-M2277 A691
-M2278 A903
-M2279 A948
-M2280 A802
-M2282 A1018
-M2284 A129
-M2285 A506
-M2286 A156
-M2287 A881
-M2289 A123
-M2290 A577
-M2291 A987
-M2292 A418
-M2293 A749
-M2294 A577
-M2296 A1178
-M2297 A432
-M2298 A818
-M2305 A741
-M2306 A1346
-M2308 A1213
-M2310 A539
-M2311 A1040
-M2312 A908
-M2314 A592
-M2315 A769
-M2316 A1047
-M2317 A1304
-M2319 A826
-M2321 A743
-M2323 A677
-M2324 A1237
-M2326 A947
-M2329 A826
-M2330 A94
-M2331 A242
-M2333 A178
-M2334 A1006
-M2337 A165
-M2343 A318
-M2344 A2
-M2346 A640
-M2352 A1141
-M2354 A519
-M2355 A699
-M2356 A31
-M2357 A411
-M2358 A197
-M2359 A12
-M2360 A35
-M2361 A1149
-M2363 A1365
-M2364 A153
-M2367 A827
-M2368 A366
-M2372 A94
-M2375 A419
-M2377 A426
-M2378 A482
-M2380 A695
-M2381 A1110
-M2383 A1345
-M2386 A903
-M2387 A310
-M2389 A742
-M2391 A529
-M2392 A956
-M2393 A224
-M2395 A718
-M2396 A425
-M2397 A1111
-M2398 A423
-M2399 A903
-M2400 A703
-M2401 A692
-M2402 A1007
-M2403 A29
-M2404 A13
-M2405 A743
-M2407 A1021
-M2408 A346
-M2409 A531
-M2411 A795
-M2413 A903
-M2414 A562
-M2415 A1259
-M2418 A1179
-M2420 A1089
-M2421 A546
-M2424 A690
-M2425 A577
-M2428 A548
-M2429 A910
-M2430 A881
-M2431 A295
-M2432 A1122
-M2433 A885
-M2434 A991
-M2435 A156
-M2436 A1254
-M2437 A1345
-M2439 A1073
-M2442 A1153
-M2443 A1354
-M2446 A727
-M2449 A742
-M2450 A698
-M2452 A713
-M2454 A685
-M2455 A993
-M2456 A762
-M2457 A881
-M2463 A593
-M2465 A1079
-M2469 A700
-M2470 A303
-M2473 A1059
-M2474 A577
-M2475 A562
-M2476 A527
-M2481 A1075
-M2482 A624
-M2485 A1176
-M2486 A66
-M2488 A628
-M2489 A227
-M2490 A561
-M2492 A1089
-M2493 A19
-M2495 A1317
-M2496 A1117
-M2497 A86
-M2498 A1012
-M2499 A340
-M2501 A196
-M2502 A12
-M2503 A158
-M2508 A603
-M2509 A571
-M2510 A1374
-M2513 A961
-M2516 A13
-M2517 A1310
-M2520 A1024
-M2521 A20
-M2523 A754
-M2527 A937
-M2531 A876
-M2532 A1040
-M2533 A754
-M2534 A1327
-M2535 A996
-M2536 A243
-M2538 A1008
-M2539 A591
-M2542 A1352
-M2543 A1092
-M2544 A1024
-M2546 A70
-M2547 A245
-M2548 A379
-M2553 A533
-M2554 A648
-M2556 A415
-M2560 A974
-M2562 A1157
-M2563 A1245
-M2565 A698
-M2567 A79
-M2568 A903
-M2570 A847
-M2571 A1367
-M2572 A1179
-M2577 A881
-M2578 A987
-M2579 A775
-M2582 A317
-M2583 A929
-M2584 A861
-M2589 A1206
-M2590 A1046
-M2592 A1091
-M2599 A1239
-M2600 A1052
-M2601 A876
-M2602 A328
-M2603 A1048
-M2604 A1050
-M2606 A1235
-M2607 A51
-M2608 A772
-M2611 A1340
-M2613 A9
-M2614 A856
-M2615 A1256
-M2616 A1155
-M2617 A1018
-M2619 A1089
-M2620 A1123
-M2622 A900
-M2623 A1339
-M2624 A636
-M2625 A826
-M2626 A1073
-M2627 A399
-M2628 A1227
-M2629 A577
-M2630 A577
-M2632 A36
-M2634 A1355
-M2636 A577
-M2638 A641
-M2639 A562
-M2640 A1024
-M2641 A214
-M2642 A58
-M2644 A1182
-M2645 A1186
-M2646 A109
-M2647 A807
-M2648 A800
-M2650 A419
-M2651 A964
-M2652 A1179
-M2653 A383
-M2655 A281
-M2658 A1312
-M2659 A400
-M2660 A677
-M2664 A974
-M2666 A808
-M2667 A961
-M2668 A477
-M2669 A549
-M2672 A1129
-M2675 A833
-M2677 A383
-M2678 A311
-M2679 A254
-M2680 A317
-M2681 A240
-M2682 A178
-M2683 A831
-M2687 A775
-M2689 A243
-M2691 A1373
-M2692 A222
-M2694 A1353
-M2695 A881
-M2698 A1339
-M2700 A577
-M2703 A831
-M2704 A411
-M2705 A927
-M2706 A133
-M2707 A910
-M2709 A51
-M2710 A117
-M2714 A562
-M2716 A351
-M2717 A186
-M2719 A1307
-M2721 A1040
-M2723 A1043
-M2725 A808
-M2726 A359
-M2728 A882
-M2731 A1043
-M2733 A506
-M2736 A1345
-M2738 A655
-M2741 A539
-M2742 A161
-M2746 A396
-M2755 A1048
-M2757 A219
-M2760 A977
-M2761 A851
-M2763 A1222
-M2764 A246
-M2768 A1050
-M2771 A93
-M2772 A363
-M2773 A368
-M2776 A857
-M2777 A1127
-M2778 A841
-M2779 A492
-M2780 A29
-M2781 A714
-M2784 A1303
-M2785 A920
-M2787 A1155
-M2788 A1319
-M2789 A99
-M2790 A422
-M2793 A778
-M2794 A577
-M2795 A400
-M2796 A1345
-M2798 A1227
-M2800 A412
-M2803 A156
-M2804 A844
-M2805 A250
-M2806 A502
-M2808 A499
-M2809 A51
-M2811 A1202
-M2813 A120
-M2815 A1081
-M2816 A351
-M2818 A399
-M2819 A641
-M2820 A773
-M2821 A1004
-M2822 A546
-M2823 A482
-M2825 A156
-M2826 A1281
-M2827 A118
-M2829 A882
-M2831 A974
-M2832 A591
-M2833 A1193
-M2835 A1039
-M2836 A70
-M2837 A577
-M2838 A386
-M2840 A645
-M2841 A645
-M2842 A1190
-M2844 A1175
-M2845 A338
-M2848 A653
-M2850 A949
-M2851 A1146
-M2852 A889
-M2856 A533
-M2857 A519
-M2859 A1116
-M2860 A648
-M2862 A762
-M2864 A698
-M2865 A1272
-M2867 A256
-M2868 A889
-M2869 A1117
-M2876 A1345
-M2877 A692
-M2878 A531
-M2880 A419
-M2882 A1039
-M2883 A418
-M2886 A519
-M2887 A1141
-M2888 A318
-M2891 A1265
-M2895 A270
-M2896 A1179
-M2898 A178
-M2900 A1235
-M2901 A281
-M2902 A1381
-M2905 A383
-M2907 A257
-M2908 A1111
-M2910 A803
-M2912 A1235
-M2915 A958
-M2916 A44
-M2917 A870
-M2918 A1245
-M2920 A688
-M2921 A368
-M2922 A386
-M2923 A425
-M2926 A566
-M2928 A349
-M2929 A245
-M2930 A577
-M2931 A1252
-M2932 A928
-M2933 A390
-M2934 A1184
-M2937 A1357
-M2938 A591
-M2939 A453
-M2942 A80
-M2944 A1125
-M2945 A1382
-M2948 A351
-M2949 A861
-M2951 A59
-M2953 A208
-M2954 A857
-M2955 A1139
-M2957 A162
-M2959 A1033
-M2961 A773
-M2962 A242
-M2963 A816
-M2965 A1110
-M2966 A1064
-M2967 A243
-M2968 A661
-M2969 A974
-M2970 A957
-M2971 A546
-M2974 A850
-M2979 A775
-M2980 A1371
-M2981 A717
-M2983 A1187
-M2984 A1265
-M2985 A117
-M2986 A5
-M2989 A935
-M2994 A305
-M2995 A954
-M2997 A398
-M2998 A1263
-M3000 A899
-M3001 A93
-M3002 A48
-M3003 A513
-M3004 A707
-M3007 A515
-M3008 A485
-M3010 A274
-M3013 A868
-M3016 A849
-M3017 A816
-M3019 A1321
-M3023 A1170
-M3024 A566
-M3026 A685
-M3027 A438
-M3028 A334
-M3029 A826
-M3030 A867
-M3031 A1222
-M3035 A394
-M3036 A94
-M3037 A712
-M3038 A765
-M3039 A414
-M3040 A86
-M3041 A265
-M3042 A625
-M3045 A363
-M3046 A1069
-M3048 A945
-M3050 A29
-M3053 A550
-M3054 A1117
-M3055 A419
-M3058 A742
-M3060 A794
-M3061 A971
-M3062 A661
-M3063 A411
-M3064 A1146
-M3066 A56
-M3067 A156
-M3069 A1207
-M3070 A726
-M3071 A820
-M3075 A525
-M3076 A677
-M3078 A756
-M3080 A37
-M3082 A1129
-M3086 A813
-M3089 A591
-M3091 A86
-M3092 A1048
-M3093 A762
-M3094 A965
-M3095 A1024
-M3096 A648
-M3097 A86
-M3098 A656
-M3099 A199
-M3100 A1050
-M3102 A755
-M3103 A414
-M3104 A982
-M3105 A1256
-M3106 A1127
-M3107 A581
-M3108 A1111
-M3111 A739
-M3112 A771
-M3113 A103
-M3114 A920
-M3116 A79
-M3117 A816
-M3119 A273
-M3120 A351
-M3122 A1315
-M3123 A543
-M3124 A445
-M3125 A421
-M3127 A1
-M3128 A1050
-M3130 A107
-M3132 A1318
-M3133 A1056
-M3135 A714
-M3138 A228
-M3139 A1048
-M3140 A980
-M3143 A1086
-M3144 A1179
-M3148 A546
-M3150 A550
-M3155 A214
-M3156 A482
-M3158 A785
-M3160 A1157
-M3161 A571
-M3163 A478
-M3164 A1393
-M3165 A590
-M3166 A717
-M3168 A1293
-M3170 A742
-M3173 A523
-M3175 A1092
-M3179 A1155
-M3182 A475
-M3184 A1376
-M3185 A988
-M3186 A826
-M3188 A1310
-M3192 A101
-M3193 A961
-M3196 A42
-M3200 A90
-M3201 A1043
-M3202 A1384
-M3203 A942
-M3205 A687
-M3206 A1035
-M3209 A857
-M3210 A550
-M3212 A831
-M3213 A803
-M3214 A1039
-M3215 A1123
-M3216 A993
-M3219 A723
-M3220 A1294
-M3227 A1330
-M3228 A778
-M3231 A1180
-M3233 A282
-M3240 A1303
-M3253 A1260
-M3255 A1057
-M3256 A199
-M3258 A885
-M3260 A920
-M3262 A1083
-M3264 A1294
-M3267 A697
-M3271 A1040
-M3272 A880
-M3274 A329
-M3276 A525
-M3277 A107
-M3278 A1090
-M3281 A1055
-M3282 A490
-M3284 A865
-M3285 A1317
-M3287 A649
-M3288 A342
-M3290 A1
-M3291 A340
-M3294 A1274
-M3296 A1364
-M3297 A831
-M3300 A477
-M3302 A14
-M3304 A661
-M3306 A191
-M3307 A208
-M3308 A826
-M3314 A1202
-M3317 A577
-M3320 A359
-M3331 A513
-M3332 A340
-M3337 A1131
-M3338 A524
-M3343 A1281
-M3344 A525
-M3345 A513
-M3347 A404
-M3351 A209
-M3353 A741
-M3355 A245
-M3358 A415
-M3360 A224
-M3362 A1238
-M3366 A199
-M3368 A338
-M3370 A259
-M3374 A1127
-M3377 A1321
-M3378 A1114
-M3380 A149
-M3382 A351
-M3384 A641
-M3386 A844
-M3387 A100
-M3389 A562
-M3390 A445
-M3391 A396
-M3393 A296
-M3394 A1091
-M3396 A808
-M3399 A1213
-M3400 A257
-M3401 A1046
-M3402 A937
-M3404 A506
-M3405 A51
-M3406 A523
-M3409 A555
-M3414 A1162
-M3417 A220
-M3421 A705
-M3424 A1024
-M3428 A137
-M3430 A1352
-M3434 A9
-M3435 A464
-M3438 A698
-M3439 A1068
-M3440 A28
-M3441 A40
-M3442 A1156
-M3443 A1145
-M3447 A698
-M3448 A340
-M3450 A614
-M3452 A961
-M3453 A305
-M3457 A689
-M3459 A1127
-M3461 A698
-M3463 A698
-M3465 A1055
-M3466 A197
-M3467 A369
-M3468 A338
-M3469 A1110
-M3470 A349
-M3472 A76
-M3473 A21
-M3475 A631
-M3476 A219
-M3477 A1040
-M3478 A831
-M3479 A855
-M3480 A1141
-M3483 A962
-M3484 A739
-M3485 A715
-M3488 A884
-M3489 A737
-M3491 A1351
-M3492 A1272
-M3493 A594
-M3494 A737
-M3496 A1036
-M3498 A64
-M3499 A677
-M3503 A355
-M3505 A1021
-M3508 A243
-M3510 A525
-M3511 A657
-M3512 A595
-M3514 A1381
-M3515 A617
-M3516 A1147
-M3519 A1050
-M3520 A1043
-M3527 A202
-M3530 A590
-M3531 A1118
-M3535 A844
-M3537 A1032
-M3540 A977
-M3542 A1141
-M3543 A334
-M3553 A438
-M3554 A400
-M3559 A417
-M3560 A914
-M3561 A1379
-M3562 A505
-M3566 A539
-M3567 A713
-M3569 A329
-M3570 A964
-M3574 A726
-M3575 A66
-M3577 A1115
-M3578 A612
-M3580 A1092
-M3582 A742
-M3586 A709
-M3587 A1213
-M3589 A1048
-M3591 A468
-M3592 A1235
-M3594 A1117
-M3595 A296
-M3596 A97
-M3597 A1371
-M3598 A1083
-M3599 A544
-M3602 A971
-M3603 A246
-M3605 A142
-M3607 A1047
-M3609 A550
-M3611 A1081
-M3618 A762
-M3619 A1213
-M3621 A274
-M3627 A617
-M3629 A327
-M3630 A952
-M3631 A217
-M3633 A614
-M3637 A1376
-M3638 A887
-M3640 A604
-M3641 A714
-M3647 A687
-M3648 A1290
-M3649 A814
-M3654 A383
-M3662 A716
-M3663 A1178
-M3664 A203
-M3668 A1256
-M3669 A406
-M3671 A621
-M3673 A977
-M3674 A974
-M3675 A156
-M3679 A1198
-M3680 A461
-M3681 A414
-M3688 A1350
-M3689 A506
-M3696 A724
-M3697 A1012
-M3699 A867
-M3704 A785
-M3709 A742
-M3711 A573
-M3713 A808
-M3714 A550
-M3716 A432
-M3718 A1092
-M3719 A1044
-M3726 A588
-M3727 A1129
-M3729 A1146
-M3730 A1086
-M3731 A199
-M3732 A228
-M3733 A885
-M3737 A13
-M3743 A550
-M3744 A1083
-M3746 A319
-M3753 A483
-M3760 A439
-M3762 A566
-M3765 A987
-M3767 A1393
-M3768 A1235
-M3769 A17
-M3771 A993
-M3773 A996
-M3775 A156
-M3776 A617
-M3778 A933
-M3779 A826
-M3781 A317
-M3782 A513
-M3783 A876
-M3784 A1078
-M3788 A250
-M3795 A1092
-M3803 A1053
-M3804 A1075
-M3805 A1245
-M3811 A1185
-M3812 A903
-M3817 A742
-M3821 A935
-M3823 A1068
-M3824 A208
-M3825 A659
-M3830 A622
-M3831 A383
-M3841 A319
-M3842 A598
-M3843 A795
-M3844 A117
-M3845 A1249
-M3846 A831
-M3848 A517
-M3849 A44
-M3852 A817
-M3853 A385
-M3854 A641
-M3857 A304
-M3859 A644
-M3869 A1088
-M3873 A783
-M3874 A947
-M3876 A38
-M3879 A1054
-M3880 A334
-M3882 A254
-M3883 A172
-M3884 A1213
-M3886 A627
-M3888 A974
-M3889 A881
-M3891 A826
-M3892 A389
-M3894 A903
-M3896 A612
-M3897 A65
-M3898 A127
-M3907 A974
-M3908 A1046
-M3910 A648
-M3911 A706
-M3912 A787
-M3913 A591
-M3915 A965
-M3916 A611
-M3918 A991
-M3919 A959
-M3920 A847
-M3921 A1382
-M3922 A1183
-M3924 A471
-M3925 A1339
-M3927 A150
-M3928 A83
-M3936 A375
-M3938 A903
-M3940 A250
-M3942 A895
-M3944 A620
-M3945 A1162
-M3947 A964
-M3952 A1127
-M3956 A485
-M3961 A1065
-M3967 A9
-M3968 A439
-M3969 A935
-M3972 A622
-M3980 A28
-M3981 A419
-M3984 A1111
-M3987 A1022
-M3988 A445
-M3991 A615
-M3994 A1336
-M3999 A691
-M4001 A700
-M4003 A365
-M4009 A886
-M4011 A1355
-M4019 A348
-M4023 A265
-M4025 A1235
-M4029 A657
-M4032 A910
-M4034 A707
-M4035 A553
-M4036 A1286
-M4041 A875
-M4044 A567
-M4046 A62
-M4050 A375
-M4051 A1007
-M4053 A604
-M4054 A649
-M4055 A954
-M4057 A64
-M4058 A356
-M4060 A951
-M4063 A760
-M4066 A952
-M4067 A1291
-M4068 A462
-M4072 A1064
-M4074 A312
-M4077 A261
-M4078 A1004
-M4085 A391
-M4086 A422
-M4091 A355
-M4092 A284
-M4096 A176
-M4097 A550
-M4098 A820
-M4099 A70
-M4100 A166
-M4101 A858
-M4102 A831
-M4103 A944
-M4104 A742
-M4105 A644
-M4110 A960
-M4112 A166
-M4118 A367
-M4120 A972
-M4124 A66
-M4128 A742
-M4132 A517
-M4135 A875
-M4137 A342
-M4141 A783
-M4145 A512
-M4151 A686
-M4158 A265
-M4159 A1158
-M4161 A101
-M4162 A830
-M4165 A875
-M4167 A24
-M4168 A265
-M4171 A922
-M4173 A742
-M4180 A601
-M4182 A1186
-M4183 A224
-M4188 A1127
-M4189 A553
-M4195 A90
-M4198 A1129
-M4202 A571
-M4208 A718
-M4212 A1007
-M4216 A1115
-M4220 A1
-M4222 A107
-M4223 A695
-M4226 A870
-M4230 A37
-M4231 A741
-M4239 A83
-M4247 A439
-M4250 A881
-M4252 A881
-M4254 A138
-M4256 A573
-M4259 A120
-M4265 A490
-M4266 A694
-M4268 A1110
-M4269 A655
-M4274 A351
-M4277 A698
-M4278 A1047
-M4279 A246
-M4282 A685
-M4286 A1194
-M4290 A993
-M4293 A1298
-M4295 A775
-M4302 A506
-M4303 A286
-M4308 A589
-M4313 A683
-M4315 A539
-M4316 A1008
-M4324 A881
-M4330 A641
-M4331 A342
-M4338 A649
-M4344 A139
-M4345 A801
-M4346 A1117
-M4350 A1018
-M4352 A742
-M4357 A795
-M4358 A119
-M4359 A1357
-M4362 A1286
-M4366 A855
-M4369 A977
-M4375 A699
-M4379 A422
-M4380 A1286
-M4381 A1111
-M4384 A919
-M4388 A787
-M4389 A200
-M4396 A566
-M4401 A147
-M4404 A1182
-M4405 A861
-M4408 A515
-M4411 A993
-M4412 A661
-M4413 A724
-M4419 A770
-M4421 A705
-M4423 A437
-M4432 A602
-M4437 A737
-M4446 A1050
-M4458 A816
-M4461 A332
-M4462 A1345
-M4466 A245
-M4472 A1143
-M4474 A1139
-M4481 A1286
-M4483 A255
-M4484 A1359
-M4485 A1214
-M4486 A1366
-M4489 A1127
-M4491 A1131
-M4493 A245
-M4494 A987
-M4496 A855
-M4498 A977
-M4511 A1298
-M4516 A919
-M4530 A363
-M4532 A1089
-M4533 A594
-M4535 A577
-M4541 A848
-M4543 A383
-M4551 A955
-M4556 A831
-M4557 A698
-M4563 A1040
-M4564 A672
-M4565 A1111
-M4568 A789
-M4569 A1072
-M4573 A265
-M4574 A746
-M4575 A588
-M4577 A1008
-M4584 A1265
-M4589 A46
-M4594 A514
-M4596 A871
-M4599 A1310
-M4606 A1333
-M4614 A581
-M4623 A61
-M4631 A861
-M4636 A464
-M4638 A119
-M4643 A659
-M4649 A36
-M4650 A338
-M4654 A273
-M4657 A944
-M4661 A1071
-M4667 A164
-M4674 A566
-M4677 A584
-M4679 A715
-M4680 A514
-M4681 A70
-M4685 A731
-M4690 A418
-M4691 A1087
-M4694 A1043
-M4695 A815
-M4710 A1218
-M4730 A418
-M4734 A793
-M4738 A793
-M4741 A463
-M4744 A672
-M4745 A1122
-M4752 A274
-M4761 A94
-M4762 A993
-M4765 A286
-M4773 A1014
-M4779 A999
-M4781 A604
-M4791 A260
-M4792 A1033
-M4795 A141
-M4804 A1093
-M4813 A365
-M4821 A1089
-M4822 A837
-M4830 A284
-M4839 A105
-M4842 A81
-M4852 A351
-M4853 A455
-M4857 A10
-M4860 A485
-M4866 A489
-M4868 A1191
-M4882 A117
-M4888 A713
-M4894 A573
-M4896 A1393
-M4897 A977
-M4905 A349
-M4909 A166
-M4923 A811
-M4927 A695
-M4936 A1090
-M4941 A649
-M4942 A1381
-M4949 A751
-M4950 A751
-M4955 A385
-M4960 A477
-M4971 A807
-M4973 A296
-M4979 A589
-M4997 A554
-M5007 A710
-M5008 A1193
-M5012 A211
-M5014 A735
-M5021 A840
-M5026 A345
-M0 A752
-M1 A263
-M2 A1105
-M3 A826
-M5 A614
-M6 A910
-M8 A1040
-M9 A1073
-M10 A771
-M11 A119
-M12 A60
-M13 A263
-M14 A639
-M15 A517
-M17 A1040
-M18 A1049
-M19 A1036
-M20 A140
-M21 A359
-M22 A1129
-M23 A140
-M24 A1125
-M25 A882
-M26 A156
-M27 A974
-M28 A329
-M29 A1310
-M30 A235
-M31 A910
-M32 A1393
-M33 A996
-M34 A548
-M36 A433
-M37 A1038
-M38 A1157
-M39 A359
-M40 A1122
-M41 A882
-M42 A15
-M44 A851
-M45 A1024
-M46 A178
-M47 A418
-M48 A993
-M49 A832
-M50 A1210
-M51 A868
-M52 A958
-M54 A947
-M58 A483
-M59 A545
-M60 A1216
-M61 A51
-M62 A1157
-M63 A329
-M65 A539
-M66 A389
-M67 A352
-M68 A560
-M69 A340
-M70 A962
-M71 A1382
-M72 A691
-M73 A241
-M74 A9
-M77 A546
-M78 A1269
-M79 A1048
-M80 A974
-M83 A1310
-M86 A1035
-M88 A780
-M92 A782
-M93 A422
-M94 A579
-M95 A1373
-M96 A243
-M97 A713
-M99 A140
-M100 A1373
-M101 A448
-M102 A539
-M103 A831
-M104 A788
-M105 A1066
-M106 A996
-M107 A141
-M108 A432
-M111 A234
-M112 A542
-M113 A133
-M114 A1073
-M115 A1073
-M116 A242
-M117 A657
-M118 A962
-M119 A53
-M120 A419
-M121 A445
-M122 A255
-M123 A1117
-M127 A1039
-M128 A242
-M129 A1188
-M130 A1039
-M134 A1018
-M135 A317
-M137 A329
-M141 A1050
-M142 A381
-M143 A736
-M144 A156
-M145 A633
-M147 A263
-M148 A1326
-M149 A272
-M151 A1345
-M152 A861
-M153 A808
-M155 A317
-M157 A306
-M158 A807
-M159 A1302
-M160 A425
-M161 A910
-M162 A318
-M164 A572
-M165 A191
-M166 A252
-M168 A36
-M169 A560
-M171 A28
-M174 A885
-M175 A178
-M176 A131
-M178 A860
-M179 A713
-M181 A947
-M183 A1392
-M184 A585
-M185 A783
-M186 A490
-M187 A885
-M188 A401
-M189 A768
-M190 A431
-M192 A190
-M193 A2
-M194 A377
-M195 A70
-M197 A154
-M200 A255
-M202 A205
-M203 A340
-M205 A263
-M207 A831
-M210 A1028
-M211 A1373
-M212 A794
-M213 A448
-M216 A880
-M217 A865
-M218 A1173
-M219 A546
-M220 A279
-M221 A404
-M222 A466
-M223 A757
-M226 A1007
-M227 A242
-M228 A1329
-M229 A1348
-M230 A82
-M231 A1281
-M234 A603
-M235 A1050
-M236 A962
-M237 A962
-M238 A53
-M240 A419
-M241 A562
-M245 A1029
-M248 A773
-M249 A1144
-M250 A255
-M251 A466
-M252 A272
-M253 A709
-M254 A246
-M255 A133
-M256 A498
-M257 A482
-M259 A1111
-M264 A498
-M266 A951
-M267 A15
-M268 A597
-M270 A263
-M271 A208
-M272 A445
-M274 A993
-M275 A263
-M276 A1234
-M277 A1196
-M278 A1331
-M280 A1345
-M281 A1377
-M282 A133
-M283 A1290
-M284 A965
-M285 A220
-M286 A144
-M287 A706
-M288 A568
-M289 A826
-M290 A461
-M291 A950
-M292 A1221
-M293 A747
-M294 A29
-M295 A622
-M296 A478
-M297 A1097
-M298 A509
-M299 A363
-M300 A553
-M303 A633
-M305 A133
-M306 A296
-M307 A318
-M308 A550
-M311 A81
-M312 A1150
-M314 A754
-M315 A1326
-M316 A1162
-M318 A1392
-M321 A902
-M324 A846
-M325 A691
-M326 A419
-M329 A1131
-M330 A176
-M332 A245
-M334 A57
-M335 A54
-M336 A1322
-M337 A721
-M338 A1010
-M339 A859
-M340 A263
-M341 A318
-M342 A621
-M344 A469
-M345 A1047
-M347 A981
-M348 A603
-M349 A703
-M350 A90
-M351 A861
-M352 A602
-M353 A53
-M354 A861
-M356 A1192
-M357 A532
-M359 A609
-M360 A1272
-M361 A351
-M363 A208
-M364 A567
-M366 A794
-M369 A1107
-M370 A1111
-M372 A982
-M373 A1129
-M374 A247
-M375 A107
-M376 A933
-M378 A232
-M379 A133
-M380 A9
-M382 A49
-M383 A1290
-M384 A315
-M385 A284
-M386 A314
-M387 A178
-M389 A1144
-M390 A690
-M391 A934
-M392 A1234
-M393 A705
-M394 A1254
-M395 A1373
-M397 A1054
-M398 A1153
-M399 A70
-M400 A1007
-M401 A441
-M402 A982
-M405 A1117
-M406 A141
-M407 A570
-M408 A754
-M409 A1027
-M410 A43
-M411 A351
-M412 A1065
-M413 A498
-M415 A861
-M416 A943
-M417 A690
-M420 A178
-M423 A1234
-M424 A477
-M425 A221
-M427 A718
-M428 A333
-M430 A1327
-M431 A734
-M432 A617
-M434 A641
-M435 A313
-M436 A590
-M437 A363
-M438 A795
-M439 A490
-M441 A984
-M442 A590
-M443 A761
-M444 A246
-M445 A1393
-M446 A808
-M448 A98
-M449 A982
-M450 A36
-M451 A1393
-M453 A935
-M455 A1177
-M456 A1319
-M457 A117
-M458 A748
-M460 A412
-M461 A1
-M462 A1039
-M463 A1157
-M465 A701
-M467 A1349
-M468 A1350
-M470 A441
-M471 A314
-M474 A533
-M475 A1129
-M476 A216
-M478 A982
-M482 A1278
-M483 A961
-M484 A347
-M486 A1367
-M487 A389
-M488 A1326
-M490 A1317
-M491 A1033
-M493 A250
-M498 A908
-M500 A288
-M501 A890
-M503 A1087
-M505 A684
-M508 A65
-M509 A1234
-M510 A36
-M512 A808
-M513 A250
-M514 A982
-M515 A981
-M516 A910
-M517 A825
-M520 A1230
-M521 A17
-M522 A709
-M523 A759
-M524 A1135
-M525 A1165
-M526 A576
-M527 A9
-M528 A66
-M534 A1064
-M535 A982
-M536 A538
-M538 A480
-M539 A880
-M540 A224
-M542 A624
-M544 A337
-M545 A699
-M546 A107
-M547 A768
-M548 A1210
-M549 A429
-M550 A260
-M551 A1219
-M556 A937
-M557 A993
-M558 A582
-M560 A1224
-M561 A1263
-M562 A245
-M564 A1281
-M565 A79
-M568 A947
-M569 A1280
-M570 A1008
-M571 A533
-M572 A1021
-M574 A872
-M576 A1327
-M577 A706
-M579 A1310
-M580 A936
-M581 A1002
-M583 A732
-M584 A517
-M587 A352
-M588 A178
-M589 A246
-M590 A754
-M591 A1168
-M592 A863
-M594 A253
-M595 A238
-M596 A272
-M598 A603
-M599 A9
-M600 A378
-M602 A549
-M603 A680
-M604 A856
-M605 A351
-M607 A65
-M609 A245
-M610 A1383
-M611 A690
-M612 A21
-M613 A606
-M618 A981
-M620 A520
-M624 A284
-M627 A355
-M628 A997
-M629 A979
-M631 A885
-M632 A119
-M633 A17
-M635 A9
-M639 A602
-M640 A278
-M641 A169
-M642 A168
-M643 A28
-M644 A1353
-M646 A1080
-M648 A1373
-M650 A384
-M651 A1010
-M652 A542
-M653 A1327
-M655 A1321
-M657 A749
-M658 A214
-M660 A482
-M662 A1280
-M663 A774
-M664 A932
-M665 A16
-M666 A836
-M667 A884
-M668 A1068
-M669 A762
-M670 A802
-M671 A410
-M672 A216
-M673 A609
-M674 A822
-M675 A987
-M678 A199
-M679 A1230
-M681 A528
-M682 A490
-M683 A301
-M686 A622
-M687 A649
-M690 A1046
-M691 A245
-M692 A795
-M693 A294
-M696 A242
-M697 A1291
-M698 A1339
-M701 A415
-M702 A882
-M703 A836
-M706 A117
-M707 A423
-M708 A1137
-M709 A539
-M710 A747
-M711 A889
-M712 A298
-M714 A483
-M715 A782
-M717 A271
-M718 A479
-M719 A620
-M721 A384
-M722 A1133
-M723 A1137
-M724 A242
-M725 A857
-M726 A881
-M728 A935
-M730 A701
-M731 A246
-M732 A961
-M733 A9
-M734 A546
-M735 A795
-M736 A445
-M737 A1305
-M738 A1068
-M739 A935
-M740 A313
-M743 A1361
-M745 A1254
-M746 A94
-M747 A752
-M748 A1319
-M749 A537
-M750 A967
-M751 A750
-M753 A245
-M754 A572
-M755 A327
-M760 A258
-M761 A577
-M763 A471
-M764 A1197
-M766 A513
-M767 A29
-M768 A324
-M769 A908
-M771 A284
-M772 A9
-M773 A692
-M774 A1111
-M775 A488
-M778 A443
-M781 A947
-M783 A1098
-M784 A197
-M785 A1321
-M786 A1010
-M787 A1009
-M788 A588
-M789 A1305
-M793 A1156
-M794 A1040
-M796 A1287
-M798 A352
-M801 A760
-M803 A1332
-M804 A593
-M805 A1115
-M806 A234
-M807 A722
-M808 A1016
-M809 A1185
-M814 A1109
-M815 A966
-M817 A253
-M818 A753
-M821 A982
-M822 A124
-M823 A516
-M824 A199
-M827 A197
-M828 A833
-M829 A1390
-M830 A961
-M831 A11
-M834 A1329
-M835 A892
-M836 A1096
-M838 A747
-M840 A1395
-M841 A622
-M842 A313
-M843 A278
-M844 A1204
-M845 A362
-M846 A910
-M848 A1018
-M849 A40
-M850 A250
-M851 A249
-M853 A847
-M854 A190
-M855 A865
-M859 A1190
-M860 A714
-M862 A572
-M864 A1369
-M865 A796
-M866 A569
-M868 A681
-M869 A484
-M871 A879
-M873 A1215
-M876 A552
-M877 A928
-M878 A9
-M879 A249
-M880 A677
-M881 A1281
-M882 A1293
-M886 A531
-M887 A73
-M890 A786
-M891 A777
-M892 A882
-M893 A861
-M895 A778
-M896 A762
-M898 A1197
-M900 A808
-M901 A464
-M902 A548
-M903 A1257
-M904 A1185
-M905 A856
-M906 A780
-M907 A533
-M908 A861
-M909 A1351
-M910 A935
-M911 A224
-M912 A1317
-M913 A1371
-M914 A714
-M915 A1120
-M917 A990
-M918 A1122
-M920 A256
-M923 A239
-M924 A315
-M926 A885
-M929 A9
-M930 A772
-M931 A547
-M933 A831
-M934 A29
-M935 A982
-M936 A721
-M937 A594
-M938 A593
-M939 A529
-M940 A1050
-M941 A588
-M942 A882
-M943 A281
-M944 A718
-M945 A714
-M946 A1343
-M947 A656
-M948 A271
-M949 A351
-M950 A402
-M951 A550
-M953 A562
-M955 A732
-M956 A964
-M957 A9
-M958 A625
-M960 A794
-M964 A730
-M965 A1370
-M966 A1132
-M969 A210
-M970 A1347
-M972 A566
-M973 A296
-M974 A1199
-M975 A1316
-M977 A1393
-M978 A236
-M979 A181
-M981 A1283
-M982 A316
-M983 A935
-M984 A569
-M987 A963
-M989 A923
-M991 A846
-M993 A772
-M996 A1310
-M997 A608
-M998 A511
-M999 A363
-M1001 A1070
-M1002 A1209
-M1003 A1229
-M1004 A661
-M1005 A1219
-M1006 A1041
-M1008 A1229
-M1009 A661
-M1011 A156
-M1012 A527
-M1013 A1064
-M1014 A1043
-M1015 A674
-M1017 A588
-M1018 A201
-M1019 A747
-M1020 A610
-M1021 A935
-M1022 A147
-M1024 A648
-M1025 A643
-M1026 A441
-M1028 A984
-M1029 A29
-M1031 A1335
-M1032 A308
-M1034 A19
-M1037 A474
-M1038 A849
-M1039 A530
-M1041 A1348
-M1042 A133
-M1043 A183
-M1045 A280
-M1048 A340
-M1049 A987
-M1050 A853
-M1051 A506
-M1052 A441
-M1053 A279
-M1054 A1315
-M1055 A937
-M1057 A1377
-M1058 A1084
-M1059 A578
-M1060 A450
-M1065 A560
-M1066 A1215
-M1067 A400
-M1069 A351
-M1070 A393
-M1072 A543
-M1073 A1137
-M1074 A731
-M1077 A691
-M1078 A936
-M1079 A246
-M1080 A1167
-M1081 A342
-M1083 A1226
-M1084 A641
-M1085 A821
-M1086 A852
-M1087 A825
-M1089 A799
-M1090 A898
-M1092 A770
-M1094 A419
-M1095 A265
-M1096 A448
-M1097 A174
-M1098 A732
-M1100 A29
-M1103 A1303
-M1104 A982
-M1105 A857
-M1106 A148
-M1108 A341
-M1111 A714
-M1112 A957
-M1114 A156
-M1115 A980
-M1116 A1352
-M1118 A1152
-M1122 A333
-M1123 A1093
-M1124 A37
-M1125 A506
-M1126 A1000
-M1127 A448
-M1128 A1318
-M1131 A937
-M1134 A543
-M1135 A1323
-M1137 A193
-M1138 A789
-M1139 A216
-M1140 A178
-M1142 A1199
-M1144 A194
-M1145 A1263
-M1146 A11
-M1148 A250
-M1149 A19
-M1151 A1110
-M1153 A807
-M1156 A942
-M1157 A108
-M1158 A507
-M1161 A1276
-M1162 A482
-M1163 A65
-M1164 A907
-M1165 A1246
-M1166 A68
-M1167 A308
-M1168 A242
-M1171 A78
-M1172 A818
-M1173 A583
-M1176 A1160
-M1177 A788
-M1179 A754
-M1181 A1163
-M1183 A164
-M1185 A826
-M1186 A373
-M1189 A9
-M1190 A731
-M1191 A1382
-M1192 A550
-M1193 A277
-M1194 A1375
-M1196 A853
-M1199 A1319
-M1203 A808
-M1204 A584
-M1205 A517
-M1206 A351
-M1207 A889
-M1208 A993
-M1209 A243
-M1210 A1270
-M1211 A599
-M1212 A914
-M1213 A218
-M1214 A1392
-M1215 A1393
-M1216 A934
-M1218 A1022
-M1219 A178
-M1220 A774
-M1221 A43
-M1223 A453
-M1224 A1003
-M1225 A1333
-M1227 A587
-M1228 A1314
-M1230 A674
-M1232 A90
-M1233 A1054
-M1234 A478
-M1237 A1055
-M1238 A657
-M1240 A177
-M1243 A1072
-M1246 A655
-M1247 A177
-M1248 A9
-M1250 A1241
-M1253 A592
-M1256 A354
-M1257 A780
-M1258 A480
-M1259 A583
-M1263 A1163
-M1267 A242
-M1268 A1181
-M1271 A739
-M1272 A1046
-M1276 A15
-M1278 A974
-M1281 A151
-M1282 A420
-M1285 A910
-M1286 A1234
-M1287 A1068
-M1288 A968
-M1289 A1067
-M1290 A291
-M1291 A313
-M1292 A942
-M1293 A611
-M1295 A714
-M1296 A327
-M1297 A971
-M1301 A1154
-M1303 A153
-M1304 A1382
-M1305 A11
-M1306 A1192
-M1307 A734
-M1308 A897
-M1310 A546
-M1311 A1146
-M1312 A521
-M1314 A913
-M1317 A1003
-M1318 A1293
-M1319 A115
-M1320 A948
-M1322 A14
-M1325 A832
-M1326 A651
-M1328 A233
-M1331 A1114
-M1332 A1373
-M1333 A1339
-M1337 A363
-M1338 A78
-M1339 A584
-M1340 A1371
-M1342 A741
-M1345 A478
-M1348 A1174
-M1349 A1161
-M1351 A1236
-M1353 A1042
-M1354 A90
-M1355 A437
-M1356 A47
-M1357 A570
-M1360 A1191
-M1363 A519
-M1366 A476
-M1369 A1350
-M1370 A546
-M1371 A178
-M1373 A1122
-M1374 A126
-M1375 A828
-M1378 A1120
-M1379 A1346
-M1380 A149
-M1383 A579
-M1385 A1267
-M1387 A674
-M1388 A1161
-M1389 A123
-M1390 A561
-M1393 A75
-M1394 A1234
-M1395 A481
-M1396 A847
-M1397 A227
-M1399 A1380
-M1400 A1190
-M1401 A279
-M1402 A347
-M1403 A129
-M1405 A732
-M1406 A1243
-M1407 A340
-M1408 A243
-M1409 A1244
-M1412 A81
-M1413 A583
-M1415 A795
-M1417 A772
-M1418 A444
-M1420 A724
-M1421 A381
-M1423 A728
-M1424 A414
-M1425 A460
-M1426 A411
-M1427 A1301
-M1428 A639
-M1429 A352
-M1431 A1007
-M1434 A260
-M1436 A197
-M1438 A812
-M1439 A224
-M1440 A838
-M1441 A1122
-M1443 A456
-M1445 A953
-M1448 A56
-M1449 A448
-M1451 A370
-M1452 A246
-M1453 A1043
-M1455 A756
-M1456 A501
-M1458 A1155
-M1460 A79
-M1461 A664
-M1462 A276
-M1463 A1107
-M1465 A1311
-M1467 A208
-M1471 A602
-M1472 A828
-M1475 A236
-M1476 A41
-M1478 A958
-M1479 A1393
-M1480 A598
-M1483 A1050
-M1486 A880
-M1487 A1282
-M1488 A1213
-M1489 A348
-M1490 A226
-M1492 A339
-M1493 A753
-M1495 A602
-M1498 A594
-M1500 A1204
-M1502 A1111
-M1503 A845
-M1505 A360
-M1510 A749
-M1513 A611
-M1514 A44
-M1516 A467
-M1518 A663
-M1519 A242
-M1521 A546
-M1523 A1377
-M1524 A1199
-M1525 A739
-M1526 A1135
-M1530 A481
-M1534 A1276
-M1535 A249
-M1537 A333
-M1539 A883
-M1540 A588
-M1541 A93
-M1542 A1244
-M1543 A400
-M1544 A597
-M1545 A809
-M1546 A910
-M1548 A242
-M1549 A617
-M1551 A1263
-M1552 A1100
-M1554 A561
-M1555 A627
-M1556 A62
-M1558 A1337
-M1559 A448
-M1564 A634
-M1567 A916
-M1568 A929
-M1569 A838
-M1571 A119
-M1574 A1232
-M1575 A671
-M1576 A993
-M1578 A1111
-M1579 A1278
-M1581 A190
-M1584 A252
-M1586 A982
-M1587 A1149
-M1588 A53
-M1589 A1254
-M1591 A691
-M1592 A1179
-M1593 A1021
-M1594 A1047
-M1595 A1010
-M1597 A276
-M1598 A80
-M1599 A1067
-M1600 A178
-M1601 A289
-M1603 A878
-M1604 A9
-M1605 A643
-M1606 A728
-M1608 A464
-M1610 A747
-M1611 A178
-M1612 A589
-M1615 A245
-M1616 A753
-M1618 A590
-M1619 A565
-M1621 A310
-M1624 A456
-M1626 A1310
-M1630 A1011
-M1632 A326
-M1633 A1309
-M1636 A1229
-M1637 A1227
-M1641 A186
-M1642 A962
-M1645 A997
-M1646 A549
-M1647 A109
-M1648 A935
-M1650 A1278
-M1654 A324
-M1655 A76
-M1657 A817
-M1660 A68
-M1661 A227
-M1662 A1310
-M1665 A228
-M1666 A32
-M1668 A388
-M1674 A1129
-M1675 A86
-M1676 A396
-M1677 A216
-M1681 A282
-M1682 A170
-M1683 A769
-M1685 A958
-M1688 A243
-M1689 A857
-M1690 A242
-M1691 A111
-M1692 A485
-M1694 A898
-M1699 A873
-M1701 A20
-M1707 A1332
-M1708 A394
-M1711 A1081
-M1713 A9
-M1714 A915
-M1715 A1294
-M1716 A377
-M1718 A1066
-M1722 A317
-M1725 A705
-M1726 A399
-M1735 A541
-M1739 A21
-M1741 A988
-M1742 A89
-M1743 A1353
-M1744 A400
-M1747 A916
-M1751 A308
-M1753 A1382
-M1754 A245
-M1755 A1042
-M1756 A826
-M1757 A357
-M1758 A620
-M1759 A760
-M1760 A1310
-M1762 A661
-M1763 A380
-M1764 A9
-M1767 A934
-M1768 A329
-M1775 A590
-M1776 A810
-M1778 A448
-M1783 A796
-M1784 A776
-M1785 A1192
-M1787 A740
-M1788 A405
-M1789 A574
-M1792 A1004
-M1796 A683
-M1798 A1303
-M1799 A497
-M1801 A641
-M1803 A523
-M1804 A1176
-M1805 A1048
-M1806 A807
-M1808 A747
-M1811 A1224
-M1812 A9
-M1813 A880
-M1814 A1156
-M1816 A1129
-M1820 A926
-M1821 A432
-M1822 A357
-M1824 A1300
-M1825 A774
-M1827 A831
-M1828 A958
-M1829 A964
-M1830 A272
-M1834 A382
-M1837 A1098
-M1838 A571
-M1839 A279
-M1840 A899
-M1841 A284
-M1846 A1154
-M1849 A147
-M1850 A1344
-M1852 A641
-M1853 A1227
-M1854 A1064
-M1856 A1053
-M1859 A79
-M1861 A1067
-M1862 A34
-M1864 A502
-M1865 A1165
-M1867 A452
-M1868 A1041
-M1870 A880
-M1872 A173
-M1873 A79
-M1874 A1299
-M1875 A851
-M1876 A1212
-M1877 A718
-M1879 A819
-M1881 A1166
-M1882 A1086
-M1885 A1010
-M1886 A564
-M1889 A934
-M1890 A1138
-M1891 A166
-M1893 A352
-M1895 A269
-M1898 A156
-M1899 A564
-M1900 A965
-M1901 A847
-M1902 A29
-M1903 A79
-M1904 A1227
-M1905 A375
-M1906 A335
-M1907 A971
-M1908 A230
-M1910 A1014
-M1914 A944
-M1915 A422
-M1918 A285
-M1920 A714
-M1924 A1128
-M1925 A49
-M1928 A1017
-M1929 A701
-M1931 A1220
-M1935 A1203
-M1937 A298
-M1938 A626
-M1939 A724
-M1941 A714
-M1945 A1141
-M1948 A37
-M1949 A1145
-M1952 A376
-M1955 A227
-M1959 A976
-M1961 A970
-M1962 A666
-M1965 A1297
-M1968 A752
-M1970 A538
-M1973 A1111
-M1974 A705
-M1975 A392
-M1976 A1289
-M1978 A965
-M1979 A419
-M1980 A960
-M1981 A349
-M1982 A588
-M1985 A1268
-M1986 A302
-M1987 A204
-M1989 A242
-M1992 A520
-M1994 A481
-M1998 A1396
-M1999 A478
-M2000 A549
-M2005 A341
-M2008 A233
-M2011 A178
-M2015 A325
-M2016 A1212
-M2017 A250
-M2018 A1244
-M2019 A915
-M2020 A1392
-M2021 A445
-M2022 A620
-M2023 A586
-M2024 A111
-M2025 A511
-M2027 A1103
-M2028 A1146
-M2031 A1148
-M2033 A944
-M2034 A740
-M2035 A571
-M2036 A335
-M2039 A344
-M2040 A785
-M2042 A506
-M2043 A579
-M2046 A777
-M2047 A747
-M2048 A1048
-M2049 A882
-M2050 A21
-M2051 A1262
-M2052 A221
-M2054 A1360
-M2055 A831
-M2056 A94
-M2057 A361
-M2062 A48
-M2065 A539
-M2070 A950
-M2071 A242
-M2074 A658
-M2079 A1134
-M2081 A807
-M2082 A724
-M2083 A1242
-M2084 A1068
-M2085 A93
-M2086 A714
-M2090 A941
-M2091 A391
-M2093 A158
-M2094 A1280
-M2095 A1300
-M2096 A674
-M2099 A11
-M2102 A247
-M2103 A1370
-M2104 A1149
-M2105 A1345
-M2107 A349
-M2108 A835
-M2109 A910
-M2110 A1248
-M2111 A884
-M2113 A1396
-M2114 A694
-M2115 A1130
-M2118 A120
-M2121 A259
-M2124 A1141
-M2125 A413
-M2126 A176
-M2127 A373
-M2128 A670
-M2129 A878
-M2131 A497
-M2133 A141
-M2134 A832
-M2136 A1063
-M2137 A577
-M2138 A880
-M2139 A391
-M2144 A87
-M2146 A1110
-M2151 A1379
-M2152 A829
-M2153 A477
-M2154 A830
-M2156 A1052
-M2158 A880
-M2160 A457
-M2161 A940
-M2163 A701
-M2164 A828
-M2165 A840
-M2166 A115
-M2167 A946
-M2168 A1046
-M2170 A234
-M2174 A193
-M2175 A587
-M2176 A1312
-M2177 A513
-M2179 A1128
-M2181 A1068
-M2184 A1111
-M2189 A1177
-M2192 A1319
-M2193 A690
-M2196 A104
-M2198 A931
-M2199 A531
-M2205 A822
-M2207 A909
-M2209 A1110
-M2210 A821
-M2212 A373
-M2214 A806
-M2217 A1251
-M2219 A701
-M2221 A1123
-M2222 A299
-M2223 A762
-M2224 A1101
-M2225 A1035
-M2227 A1346
-M2230 A311
-M2231 A1227
-M2232 A328
-M2234 A1337
-M2237 A585
-M2238 A675
-M2239 A116
-M2240 A1046
-M2243 A366
-M2244 A993
-M2245 A743
-M2246 A637
-M2249 A961
-M2250 A1129
-M2251 A946
-M2253 A1382
-M2255 A1110
-M2260 A794
-M2262 A1120
-M2263 A212
-M2265 A649
-M2267 A928
-M2268 A239
-M2269 A1309
-M2270 A1359
-M2274 A958
-M2275 A643
-M2277 A187
-M2278 A237
-M2279 A1151
-M2281 A892
-M2282 A966
-M2284 A667
-M2285 A465
-M2286 A844
-M2287 A1118
-M2289 A679
-M2290 A1199
-M2291 A50
-M2292 A832
-M2293 A772
-M2294 A674
-M2297 A907
-M2298 A13
-M2304 A1366
-M2305 A1384
-M2307 A892
-M2308 A500
-M2310 A250
-M2311 A1127
-M2312 A822
-M2315 A314
-M2316 A1124
-M2317 A873
-M2319 A389
-M2321 A261
-M2323 A445
-M2324 A139
-M2326 A621
-M2328 A1268
-M2329 A376
-M2331 A44
-M2333 A133
-M2334 A241
-M2337 A761
-M2352 A1305
-M2354 A1259
-M2355 A100
-M2357 A1371
-M2358 A542
-M2359 A724
-M2360 A1062
-M2361 A1299
-M2362 A1184
-M2363 A518
-M2364 A1359
-M2367 A1375
-M2369 A28
-M2372 A148
-M2375 A808
-M2377 A181
-M2380 A66
-M2381 A57
-M2383 A117
-M2386 A714
-M2387 A869
-M2389 A899
-M2393 A1367
-M2396 A1244
-M2397 A118
-M2398 A944
-M2399 A1365
-M2400 A554
-M2401 A386
-M2402 A239
-M2403 A63
-M2405 A247
-M2406 A739
-M2407 A335
-M2408 A592
-M2409 A423
-M2411 A651
-M2413 A889
-M2414 A1043
-M2417 A1146
-M2418 A9
-M2420 A1327
-M2422 A651
-M2425 A1098
-M2429 A1093
-M2430 A1367
-M2432 A978
-M2433 A1211
-M2434 A617
-M2435 A539
-M2437 A273
-M2439 A1089
-M2441 A88
-M2442 A1289
-M2449 A1048
-M2450 A156
-M2452 A908
-M2454 A1364
-M2456 A176
-M2457 A327
-M2463 A301
-M2465 A262
-M2470 A380
-M2474 A519
-M2475 A1175
-M2476 A285
-M2481 A279
-M2488 A414
-M2489 A442
-M2492 A1083
-M2493 A78
-M2495 A963
-M2496 A1064
-M2497 A269
-M2498 A410
-M2499 A1122
-M2501 A217
-M2502 A700
-M2508 A450
-M2509 A1224
-M2510 A1
-M2513 A701
-M2516 A906
-M2517 A546
-M2520 A643
-M2521 A941
-M2523 A794
-M2527 A914
-M2531 A1089
-M2532 A147
-M2533 A794
-M2535 A156
-M2536 A1044
-M2538 A933
-M2539 A863
-M2542 A573
-M2543 A508
-M2546 A395
-M2547 A1048
-M2548 A1355
-M2553 A1019
-M2554 A821
-M2560 A433
-M2562 A1068
-M2563 A350
-M2565 A1363
-M2568 A1365
-M2571 A1245
-M2572 A224
-M2577 A1065
-M2578 A724
-M2579 A1296
-M2582 A47
-M2584 A11
-M2589 A451
-M2590 A104
-M2592 A1154
-M2595 A798
-M2599 A845
-M2600 A1282
-M2601 A1033
-M2602 A259
-M2603 A320
-M2604 A107
-M2606 A1359
-M2607 A1376
-M2608 A1031
-M2611 A270
-M2613 A1188
-M2614 A1105
-M2615 A1315
-M2616 A1123
-M2617 A743
-M2619 A1083
-M2620 A902
-M2623 A1021
-M2624 A171
-M2625 A762
-M2626 A900
-M2628 A1004
-M2629 A714
-M2630 A762
-M2632 A1317
-M2634 A1382
-M2636 A363
-M2638 A589
-M2639 A123
-M2640 A533
-M2641 A376
-M2642 A586
-M2644 A493
-M2645 A576
-M2647 A412
-M2648 A1065
-M2650 A448
-M2651 A854
-M2652 A228
-M2653 A1315
-M2655 A779
-M2659 A963
-M2660 A181
-M2664 A1013
-M2668 A618
-M2669 A646
-M2672 A1155
-M2677 A402
-M2681 A1357
-M2682 A176
-M2689 A81
-M2691 A943
-M2692 A323
-M2695 A186
-M2698 A1027
-M2700 A506
-M2701 A380
-M2703 A340
-M2705 A1119
-M2706 A716
-M2707 A130
-M2709 A1111
-M2710 A1194
-M2714 A123
-M2716 A1327
-M2717 A971
-M2719 A183
-M2721 A558
-M2723 A326
-M2725 A1043
-M2726 A474
-M2731 A1096
-M2736 A14
-M2738 A880
-M2746 A1396
-M2755 A949
-M2757 A1317
-M2760 A9
-M2761 A1183
-M2763 A765
-M2764 A80
-M2768 A777
-M2769 A16
-M2770 A234
-M2772 A450
-M2773 A302
-M2774 A1055
-M2776 A391
-M2777 A1212
-M2780 A430
-M2781 A1213
-M2783 A927
-M2785 A1062
-M2786 A454
-M2787 A248
-M2788 A89
-M2789 A1141
-M2794 A23
-M2795 A352
-M2798 A1004
-M2800 A253
-M2802 A60
-M2803 A620
-M2805 A96
-M2806 A858
-M2809 A795
-M2813 A92
-M2816 A477
-M2818 A380
-M2819 A389
-M2820 A1029
-M2821 A136
-M2822 A660
-M2823 A976
-M2826 A570
-M2832 A1311
-M2835 A1157
-M2836 A1281
-M2837 A197
-M2838 A599
-M2840 A307
-M2841 A584
-M2844 A586
-M2845 A389
-M2848 A616
-M2850 A1285
-M2851 A368
-M2852 A1231
-M2856 A792
-M2857 A82
-M2860 A865
-M2862 A1068
-M2864 A561
-M2865 A445
-M2867 A837
-M2868 A194
-M2869 A29
-M2876 A942
-M2882 A641
-M2883 A1373
-M2886 A5
-M2887 A1379
-M2888 A549
-M2895 A1085
-M2896 A728
-M2897 A1299
-M2898 A70
-M2900 A197
-M2901 A1197
-M2902 A461
-M2905 A83
-M2907 A1396
-M2908 A1150
-M2910 A388
-M2915 A732
-M2917 A439
-M2918 A1149
-M2921 A382
-M2923 A34
-M2924 A1396
-M2926 A250
-M2928 A277
-M2929 A1199
-M2930 A737
-M2932 A766
-M2933 A921
-M2934 A1201
-M2937 A1277
-M2938 A571
-M2939 A1068
-M2942 A483
-M2944 A144
-M2945 A1305
-M2948 A968
-M2949 A382
-M2951 A1164
-M2954 A421
-M2957 A633
-M2959 A689
-M2961 A1192
-M2962 A1025
-M2963 A424
-M2965 A841
-M2966 A302
-M2967 A784
-M2968 A1200
-M2969 A51
-M2971 A603
-M2973 A682
-M2974 A640
-M2979 A523
-M2980 A1339
-M2984 A930
-M2986 A410
-M2988 A1057
-M2989 A1093
-M2994 A1032
-M2995 A78
-M2997 A237
-M2998 A767
-M3000 A167
-M3001 A52
-M3002 A1118
-M3003 A840
-M3004 A792
-M3007 A998
-M3008 A941
-M3010 A401
-M3017 A178
-M3019 A1222
-M3023 A200
-M3030 A457
-M3031 A934
-M3035 A970
-M3037 A139
-M3038 A660
-M3039 A386
-M3040 A720
-M3041 A1088
-M3045 A1181
-M3050 A920
-M3053 A218
-M3054 A1008
-M3055 A284
-M3056 A1224
-M3058 A219
-M3060 A1259
-M3061 A1119
-M3062 A643
-M3064 A1388
-M3066 A244
-M3067 A1345
-M3069 A812
-M3070 A1295
-M3071 A86
-M3076 A970
-M3078 A292
-M3080 A1255
-M3089 A351
-M3091 A720
-M3093 A1243
-M3095 A740
-M3096 A570
-M3097 A720
-M3099 A1312
-M3100 A14
-M3103 A740
-M3104 A647
-M3106 A931
-M3108 A860
-M3110 A6
-M3111 A138
-M3112 A967
-M3113 A1023
-M3114 A1303
-M3117 A178
-M3119 A1198
-M3120 A937
-M3123 A744
-M3124 A1017
-M3127 A883
-M3128 A300
-M3132 A914
-M3135 A561
-M3138 A729
-M3139 A1290
-M3140 A1004
-M3141 A137
-M3143 A282
-M3144 A146
-M3146 A1291
-M3148 A151
-M3150 A226
-M3155 A519
-M3156 A150
-M3158 A1358
-M3160 A762
-M3161 A1010
-M3163 A1165
-M3164 A170
-M3165 A741
-M3168 A913
-M3170 A133
-M3173 A1010
-M3175 A1123
-M3184 A976
-M3185 A1253
-M3193 A391
-M3198 A313
-M3200 A1006
-M3205 A243
-M3206 A910
-M3209 A717
-M3212 A677
-M3213 A790
-M3215 A543
-M3216 A142
-M3220 A360
-M3227 A702
-M3231 A625
-M3240 A1253
-M3241 A814
-M3249 A1188
-M3251 A753
-M3256 A426
-M3258 A683
-M3260 A1303
-M3267 A1088
-M3270 A705
-M3271 A1213
-M3274 A762
-M3277 A76
-M3284 A445
-M3285 A963
-M3288 A708
-M3290 A483
-M3291 A70
-M3293 A763
-M3294 A1061
-M3296 A1322
-M3297 A980
-M3300 A904
-M3304 A1189
-M3305 A1308
-M3306 A1208
-M3307 A423
-M3308 A396
-M3317 A1107
-M3331 A372
-M3332 A55
-M3337 A520
-M3338 A606
-M3343 A554
-M3344 A504
-M3345 A511
-M3353 A370
-M3355 A770
-M3366 A963
-M3368 A589
-M3370 A1327
-M3374 A1179
-M3378 A1285
-M3380 A906
-M3382 A210
-M3384 A588
-M3386 A820
-M3388 A446
-M3389 A12
-M3390 A572
-M3391 A1396
-M3392 A1369
-M3393 A313
-M3394 A1022
-M3396 A929
-M3399 A894
-M3400 A1294
-M3401 A950
-M3405 A1213
-M3414 A684
-M3417 A970
-M3424 A1122
-M3430 A413
-M3435 A1146
-M3438 A769
-M3439 A712
-M3440 A1000
-M3441 A521
-M3442 A838
-M3443 A538
-M3447 A1157
-M3448 A732
-M3450 A33
-M3452 A391
-M3453 A77
-M3457 A1185
-M3459 A574
-M3461 A910
-M3463 A645
-M3465 A668
-M3466 A119
-M3468 A1018
-M3470 A196
-M3471 A0
-M3473 A1227
-M3475 A391
-M3476 A1210
-M3477 A1108
-M3479 A1345
-M3480 A1379
-M3492 A468
-M3493 A235
-M3498 A202
-M3499 A625
-M3501 A613
-M3503 A294
-M3505 A18
-M3512 A1034
-M3514 A971
-M3515 A143
-M3516 A1096
-M3523 A128
-M3530 A863
-M3531 A486
-M3537 A689
-M3540 A511
-M3542 A593
-M3543 A1211
-M3559 A727
-M3560 A40
-M3561 A608
-M3566 A476
-M3567 A1259
-M3569 A458
-M3570 A735
-M3574 A614
-M3575 A1390
-M3578 A1385
-M3580 A1193
-M3581 A327
-M3582 A777
-M3586 A1171
-M3587 A208
-M3591 A1253
-M3592 A643
-M3594 A914
-M3597 A1339
-M3598 A1362
-M3599 A805
-M3602 A465
-M3603 A701
-M3605 A1353
-M3607 A169
-M3618 A889
-M3621 A54
-M3627 A861
-M3631 A115
-M3637 A154
-M3638 A797
-M3640 A1000
-M3641 A1226
-M3647 A1122
-M3648 A20
-M3652 A686
-M3654 A511
-M3663 A463
-M3668 A439
-M3671 A1316
-M3674 A1345
-M3675 A1213
-M3679 A222
-M3680 A1011
-M3681 A770
-M3688 A1042
-M3689 A919
-M3696 A613
-M3697 A310
-M3699 A457
-M3704 A1358
-M3709 A1128
-M3713 A52
-M3714 A645
-M3716 A730
-M3717 A1338
-M3718 A1193
-M3719 A1306
-M3722 A719
-M3726 A482
-M3727 A1263
-M3729 A1388
-M3732 A510
-M3733 A43
-M3734 A1279
-M3741 A475
-M3746 A170
-M3751 A1168
-M3753 A1303
-M3760 A763
-M3762 A1386
-M3765 A672
-M3766 A1313
-M3767 A18
-M3769 A372
-M3771 A1379
-M3773 A238
-M3775 A289
-M3778 A945
-M3779 A117
-M3780 A725
-M3781 A429
-M3782 A1283
-M3783 A1287
-M3784 A1285
-M3788 A778
-M3795 A1081
-M3811 A165
-M3814 A194
-M3817 A1163
-M3821 A183
-M3824 A291
-M3825 A341
-M3831 A795
-M3844 A1052
-M3846 A861
-M3849 A296
-M3852 A1011
-M3854 A31
-M3857 A267
-M3858 A291
-M3867 A69
-M3869 A1231
-M3873 A947
-M3879 A633
-M3880 A1211
-M3882 A472
-M3884 A942
-M3886 A869
-M3888 A1048
-M3889 A854
-M3891 A1302
-M3892 A1181
-M3894 A1041
-M3896 A1385
-M3898 A1284
-M3907 A1353
-M3908 A706
-M3910 A1303
-M3911 A1014
-M3912 A358
-M3913 A572
-M3915 A666
-M3917 A601
-M3918 A1084
-M3920 A538
-M3921 A857
-M3922 A983
-M3924 A839
-M3925 A483
-M3928 A1391
-M3929 A705
-M3936 A299
-M3938 A226
-M3940 A755
-M3942 A1008
-M3945 A1219
-M3947 A1043
-M3955 A398
-M3961 A784
-M3967 A1188
-M3969 A1084
-M3972 A615
-M3980 A552
-M3981 A90
-M3984 A753
-M3987 A920
-M3988 A522
-M3991 A897
-M3994 A344
-M3998 A629
-M3999 A1077
-M4009 A879
-M4011 A662
-M4023 A537
-M4025 A1199
-M4032 A1028
-M4035 A1221
-M4040 A667
-M4041 A1313
-M4044 A1391
-M4046 A829
-M4049 A738
-M4050 A291
-M4051 A1135
-M4053 A1383
-M4057 A922
-M4063 A14
-M4067 A1249
-M4068 A689
-M4069 A488
-M4072 A149
-M4076 A459
-M4077 A787
-M4085 A1351
-M4086 A10
-M4092 A1204
-M4097 A293
-M4098 A1310
-M4099 A471
-M4102 A1367
-M4104 A1163
-M4110 A635
-M4112 A917
-M4124 A712
-M4128 A996
-M4132 A423
-M4135 A417
-M4141 A1322
-M4158 A1181
-M4159 A869
-M4171 A763
-M4173 A242
-M4180 A1298
-M4182 A576
-M4183 A242
-M4188 A1396
-M4189 A1389
-M4198 A195
-M4202 A1010
-M4208 A560
-M4212 A692
-M4214 A1133
-M4216 A890
-M4220 A542
-M4223 A44
-M4230 A1069
-M4231 A352
-M4239 A650
-M4252 A1189
-M4254 A6
-M4256 A786
-M4257 A179
-M4260 A385
-M4265 A494
-M4274 A132
-M4277 A826
-M4278 A1037
-M4279 A852
-M4282 A1364
-M4290 A405
-M4293 A651
-M4295 A1237
-M4300 A596
-M4303 A121
-M4313 A1297
-M4315 A774
-M4330 A579
-M4331 A299
-M4336 A218
-M4344 A159
-M4346 A836
-M4350 A1310
-M4352 A724
-M4358 A1277
-M4359 A615
-M4369 A689
-M4381 A1155
-M4396 A830
-M4401 A760
-M4408 A998
-M4411 A1287
-M4412 A723
-M4413 A613
-M4423 A179
-M4437 A1255
-M4446 A935
-M4458 A448
-M4462 A930
-M4466 A1321
-M4472 A1292
-M4485 A412
-M4486 A122
-M4489 A1350
-M4491 A904
-M4493 A9
-M4494 A740
-M4496 A1345
-M4508 A676
-M4511 A849
-M4526 A1182
-M4532 A1327
-M4535 A1315
-M4536 A316
-M4541 A571
-M4543 A1066
-M4551 A306
-M4557 A645
-M4564 A314
-M4565 A753
-M4568 A805
-M4569 A966
-M4573 A1088
-M4575 A554
-M4577 A992
-M4584 A721
-M4589 A948
-M4594 A166
-M4596 A1200
-M4604 A280
-M4623 A777
-M4631 A11
-M4636 A600
-M4638 A1277
-M4645 A1342
-M4649 A1317
-M4650 A293
-M4663 A1342
-M4677 A840
-M4678 A623
-M4681 A395
-M4690 A771
-M4694 A882
-M4695 A370
-M4730 A1175
-M4741 A1071
-M4744 A492
-M4745 A589
-M4752 A401
-M4761 A1254
-M4762 A154
-M4765 A1000
-M4773 A1325
-M4778 A1157
-M4791 A324
-M4795 A1201
-M4821 A1083
-M4830 A609
-M4852 A1092
-M4853 A810
-M4860 A627
-M4863 A1356
-M4866 A618
-M4882 A1194
-M4895 A491
-M4905 A388
-M4914 A1169
-M4927 A44
-M4930 A781
-M4942 A461
-M4944 A494
-M4949 A485
-M4950 A485
-M4960 A195
-M4971 A412
-M4979 A514
-M5001 A1307
-M5012 A339
-M5014 A266
-M5021 A268
-M0 A1151
-M1 A800
-M3 A338
-M6 A762
-M8 A974
-M9 A220
-M11 A90
-M12 A1105
-M13 A800
-M14 A1111
-M15 A692
-M17 A974
-M18 A1115
-M20 A604
-M23 A604
-M24 A795
-M25 A901
-M27 A1035
-M29 A679
-M30 A1105
-M31 A762
-M32 A961
-M33 A243
-M36 A1221
-M37 A548
-M38 A910
-M39 A551
-M40 A674
-M41 A396
-M44 A1064
-M45 A1373
-M46 A211
-M47 A1054
-M50 A491
-M53 A433
-M54 A1213
-M58 A84
-M60 A891
-M61 A70
-M62 A214
-M65 A226
-M66 A9
-M69 A1393
-M70 A947
-M71 A233
-M72 A797
-M73 A1301
-M74 A590
-M75 A621
-M79 A1208
-M80 A1393
-M83 A632
-M84 A804
-M86 A783
-M87 A1273
-M92 A1324
-M93 A895
-M95 A318
-M96 A428
-M97 A338
-M99 A604
-M100 A1355
-M101 A458
-M102 A284
-M103 A490
-M105 A190
-M106 A243
-M107 A970
-M109 A321
-M111 A690
-M112 A433
-M113 A1364
-M114 A205
-M115 A220
-M116 A794
-M118 A1383
-M119 A450
-M120 A9
-M121 A1135
-M122 A208
-M127 A808
-M128 A154
-M129 A1316
-M130 A808
-M134 A962
-M135 A1295
-M136 A638
-M140 A802
-M141 A794
-M142 A387
-M143 A912
-M145 A258
-M147 A530
-M148 A1135
-M149 A275
-M150 A733
-M151 A907
-M152 A602
-M155 A427
-M158 A664
-M159 A402
-M160 A993
-M161 A762
-M162 A422
-M164 A1150
-M165 A983
-M167 A660
-M169 A352
-M171 A783
-M172 A1055
-M174 A1120
-M175 A1064
-M176 A1129
-M178 A757
-M179 A862
-M181 A1010
-M182 A503
-M183 A594
-M184 A736
-M185 A85
-M186 A843
-M187 A1120
-M188 A824
-M189 A898
-M190 A1232
-M191 A1363
-M193 A441
-M194 A1263
-M195 A220
-M197 A148
-M200 A208
-M202 A1380
-M203 A168
-M205 A800
-M207 A490
-M208 A957
-M210 A781
-M211 A36
-M213 A82
-M214 A1238
-M215 A1258
-M216 A1218
-M218 A563
-M219 A782
-M220 A1150
-M221 A276
-M222 A190
-M224 A101
-M225 A251
-M226 A652
-M227 A1295
-M228 A717
-M229 A657
-M231 A1376
-M234 A1232
-M235 A106
-M236 A617
-M237 A617
-M238 A622
-M240 A866
-M241 A903
-M243 A620
-M247 A187
-M248 A1329
-M250 A208
-M251 A716
-M252 A271
-M253 A1111
-M256 A1146
-M257 A40
-M264 A154
-M270 A859
-M271 A686
-M272 A572
-M274 A538
-M276 A1021
-M277 A1187
-M278 A913
-M280 A318
-M281 A1208
-M282 A935
-M283 A581
-M285 A2
-M286 A253
-M287 A893
-M288 A811
-M289 A1115
-M290 A1280
-M291 A1128
-M292 A575
-M293 A1185
-M295 A30
-M296 A251
-M298 A206
-M299 A245
-M303 A258
-M304 A110
-M305 A754
-M306 A1396
-M307 A1114
-M308 A1393
-M309 A930
-M310 A499
-M311 A1213
-M312 A1167
-M314 A888
-M315 A471
-M316 A532
-M318 A352
-M321 A915
-M324 A1194
-M326 A1213
-M330 A832
-M332 A70
-M334 A1246
-M336 A182
-M337 A891
-M338 A1329
-M339 A484
-M340 A859
-M341 A1000
-M344 A924
-M345 A89
-M346 A665
-M347 A917
-M348 A1232
-M351 A208
-M352 A1335
-M353 A585
-M354 A1092
-M356 A990
-M359 A540
-M360 A8
-M361 A947
-M363 A729
-M364 A1156
-M365 A357
-M368 A436
-M369 A98
-M370 A882
-M371 A647
-M372 A54
-M373 A1058
-M374 A675
-M375 A623
-M376 A540
-M378 A729
-M379 A318
-M380 A963
-M382 A684
-M385 A1371
-M386 A50
-M387 A681
-M390 A353
-M391 A784
-M392 A202
-M393 A542
-M394 A372
-M395 A1355
-M397 A288
-M399 A655
-M401 A263
-M402 A1367
-M405 A176
-M406 A1248
-M407 A1324
-M410 A405
-M411 A178
-M412 A1268
-M413 A1146
-M415 A832
-M416 A665
-M420 A381
-M423 A1029
-M426 A949
-M429 A96
-M430 A1362
-M434 A159
-M435 A807
-M436 A768
-M437 A19
-M438 A929
-M440 A994
-M442 A1339
-M443 A1372
-M444 A1340
-M445 A935
-M449 A1367
-M450 A243
-M453 A659
-M455 A1153
-M457 A281
-M458 A647
-M460 A149
-M462 A687
-M463 A70
-M465 A561
-M471 A1363
-M474 A146
-M476 A449
-M482 A112
-M483 A176
-M486 A971
-M487 A377
-M488 A862
-M491 A199
-M496 A240
-M498 A1
-M499 A522
-M508 A1324
-M510 A923
-M512 A376
-M513 A131
-M514 A180
-M516 A428
-M517 A941
-M520 A1263
-M521 A208
-M522 A31
-M524 A503
-M526 A1028
-M527 A1064
-M528 A1023
-M534 A928
-M535 A1382
-M539 A402
-M541 A133
-M542 A1187
-M544 A930
-M545 A60
-M546 A981
-M547 A1339
-M548 A290
-M550 A1339
-M552 A1220
-M556 A1234
-M557 A756
-M558 A761
-M560 A1297
-M561 A990
-M562 A928
-M564 A880
-M565 A1081
-M568 A1198
-M570 A711
-M571 A571
-M572 A164
-M576 A298
-M579 A1129
-M580 A1181
-M583 A931
-M584 A80
-M587 A252
-M588 A478
-M589 A506
-M590 A888
-M591 A1094
-M592 A1272
-M594 A621
-M595 A651
-M597 A98
-M598 A520
-M599 A94
-M601 A1221
-M602 A594
-M603 A231
-M604 A794
-M605 A651
-M606 A404
-M609 A467
-M611 A688
-M612 A225
-M614 A500
-M618 A1338
-M623 A1103
-M624 A817
-M627 A596
-M631 A589
-M632 A1368
-M633 A565
-M635 A547
-M636 A967
-M639 A930
-M640 A185
-M642 A993
-M644 A472
-M646 A298
-M648 A351
-M650 A391
-M651 A1314
-M653 A934
-M655 A750
-M656 A1238
-M657 A1242
-M658 A243
-M659 A149
-M661 A557
-M662 A595
-M663 A766
-M664 A192
-M666 A276
-M667 A122
-M668 A553
-M669 A486
-M671 A320
-M672 A980
-M673 A173
-M675 A1371
-M676 A300
-M677 A437
-M678 A887
-M679 A1153
-M684 A872
-M686 A202
-M687 A759
-M690 A567
-M691 A1125
-M692 A14
-M695 A344
-M696 A535
-M698 A5
-M699 A155
-M701 A920
-M702 A1333
-M703 A509
-M706 A106
-M707 A260
-M709 A1001
-M710 A1185
-M711 A526
-M712 A899
-M714 A1363
-M717 A221
-M718 A1102
-M721 A84
-M722 A1309
-M723 A390
-M725 A545
-M726 A52
-M728 A902
-M729 A277
-M731 A191
-M732 A1192
-M734 A1272
-M736 A930
-M737 A484
-M738 A524
-M739 A1047
-M740 A299
-M746 A292
-M747 A923
-M751 A320
-M753 A1293
-M754 A432
-M755 A155
-M761 A123
-M763 A155
-M766 A327
-M767 A1106
-M768 A1093
-M769 A550
-M771 A75
-M773 A1389
-M774 A373
-M775 A1200
-M776 A339
-M778 A378
-M781 A373
-M784 A997
-M785 A1378
-M786 A247
-M787 A242
-M788 A705
-M789 A446
-M793 A750
-M794 A974
-M795 A387
-M796 A79
-M801 A1177
-M804 A21
-M806 A222
-M807 A332
-M809 A620
-M815 A1042
-M817 A1116
-M818 A1029
-M821 A1367
-M823 A227
-M824 A869
-M827 A242
-M828 A886
-M830 A511
-M831 A775
-M834 A1108
-M838 A1185
-M840 A1197
-M841 A149
-M842 A1058
-M843 A45
-M844 A483
-M846 A590
-M847 A1098
-M850 A1180
-M853 A30
-M854 A118
-M856 A280
-M860 A5
-M862 A1311
-M863 A150
-M864 A705
-M865 A736
-M869 A33
-M871 A784
-M874 A492
-M876 A1364
-M877 A118
-M878 A433
-M879 A361
-M880 A942
-M881 A1232
-M887 A91
-M890 A668
-M891 A177
-M892 A1263
-M893 A877
-M894 A745
-M895 A1224
-M896 A1310
-M897 A350
-M900 A433
-M901 A122
-M902 A213
-M903 A1150
-M905 A482
-M906 A891
-M907 A928
-M908 A106
-M909 A1284
-M911 A645
-M913 A512
-M914 A1044
-M915 A243
-M916 A1389
-M917 A1119
-M918 A1228
-M920 A164
-M924 A909
-M925 A1078
-M926 A1120
-M929 A471
-M930 A1223
-M931 A107
-M933 A412
-M934 A590
-M935 A1359
-M937 A27
-M938 A730
-M940 A762
-M942 A1158
-M945 A589
-M946 A680
-M947 A192
-M949 A168
-M950 A1209
-M951 A588
-M953 A590
-M955 A1027
-M956 A178
-M957 A252
-M958 A193
-M959 A912
-M960 A922
-M964 A678
-M966 A888
-M969 A1159
-M970 A388
-M972 A584
-M973 A880
-M974 A1371
-M977 A18
-M978 A987
-M979 A1096
-M982 A1170
-M983 A833
-M987 A1169
-M991 A791
-M995 A437
-M996 A208
-M997 A686
-M998 A992
-M1001 A135
-M1003 A779
-M1005 A1316
-M1008 A704
-M1011 A387
-M1012 A83
-M1013 A934
-M1015 A892
-M1017 A910
-M1018 A113
-M1021 A1199
-M1022 A606
-M1024 A1392
-M1025 A1335
-M1026 A549
-M1027 A1148
-M1028 A1029
-M1029 A652
-M1031 A1208
-M1032 A659
-M1034 A352
-M1036 A184
-M1041 A449
-M1042 A975
-M1048 A400
-M1049 A670
-M1050 A1152
-M1051 A1222
-M1052 A1082
-M1053 A813
-M1057 A492
-M1058 A1314
-M1060 A362
-M1065 A623
-M1066 A1061
-M1067 A378
-M1069 A880
-M1072 A1099
-M1073 A537
-M1074 A1063
-M1075 A764
-M1078 A80
-M1079 A807
-M1080 A262
-M1081 A734
-M1083 A1015
-M1084 A966
-M1085 A777
-M1087 A503
-M1089 A1396
-M1090 A815
-M1092 A1328
-M1094 A22
-M1095 A53
-M1096 A1045
-M1098 A252
-M1100 A935
-M1102 A1142
-M1104 A1300
-M1105 A479
-M1107 A1333
-M1111 A246
-M1112 A493
-M1113 A1195
-M1119 A180
-M1124 A176
-M1125 A1153
-M1126 A479
-M1127 A1151
-M1128 A302
-M1131 A468
-M1134 A146
-M1137 A1108
-M1138 A782
-M1139 A956
-M1140 A677
-M1141 A209
-M1142 A314
-M1144 A1294
-M1145 A1322
-M1146 A102
-M1147 A369
-M1148 A1137
-M1149 A506
-M1151 A444
-M1154 A227
-M1155 A475
-M1157 A1065
-M1159 A1384
-M1162 A1028
-M1166 A244
-M1167 A561
-M1168 A548
-M1171 A345
-M1172 A1133
-M1181 A1000
-M1184 A198
-M1185 A123
-M1186 A703
-M1188 A518
-M1189 A210
-M1190 A432
-M1191 A741
-M1192 A284
-M1193 A1223
-M1196 A241
-M1199 A586
-M1200 A951
-M1201 A518
-M1203 A70
-M1204 A527
-M1205 A476
-M1206 A800
-M1207 A227
-M1209 A252
-M1210 A934
-M1212 A692
-M1214 A1129
-M1219 A246
-M1220 A766
-M1221 A730
-M1224 A823
-M1227 A788
-M1228 A316
-M1230 A471
-M1231 A1008
-M1232 A20
-M1233 A974
-M1234 A1243
-M1237 A638
-M1238 A1132
-M1239 A959
-M1240 A479
-M1243 A1064
-M1246 A561
-M1247 A484
-M1248 A550
-M1253 A526
-M1258 A750
-M1261 A932
-M1263 A417
-M1266 A595
-M1267 A547
-M1271 A675
-M1272 A1356
-M1277 A767
-M1278 A856
-M1284 A141
-M1285 A1308
-M1287 A337
-M1288 A1224
-M1289 A501
-M1290 A602
-M1291 A150
-M1295 A235
-M1296 A116
-M1299 A460
-M1303 A1396
-M1304 A304
-M1305 A102
-M1306 A871
-M1307 A1303
-M1308 A1356
-M1311 A1034
-M1312 A922
-M1313 A440
-M1319 A449
-M1325 A1057
-M1326 A360
-M1328 A184
-M1331 A912
-M1332 A1355
-M1333 A395
-M1345 A674
-M1348 A656
-M1351 A189
-M1352 A134
-M1353 A8
-M1354 A756
-M1355 A145
-M1357 A4
-M1358 A1
-M1360 A1023
-M1362 A1037
-M1366 A177
-M1368 A185
-M1370 A730
-M1371 A1058
-M1373 A100
-M1375 A164
-M1377 A1229
-M1378 A589
-M1379 A314
-M1380 A1253
-M1381 A730
-M1383 A1311
-M1385 A1061
-M1386 A1229
-M1387 A1102
-M1388 A522
-M1389 A1048
-M1390 A703
-M1391 A122
-M1392 A519
-M1396 A828
-M1397 A1303
-M1399 A344
-M1400 A1006
-M1401 A394
-M1402 A557
-M1403 A933
-M1405 A955
-M1409 A911
-M1410 A85
-M1411 A271
-M1412 A54
-M1415 A957
-M1416 A437
-M1417 A879
-M1418 A883
-M1421 A116
-M1423 A371
-M1424 A690
-M1426 A108
-M1428 A551
-M1431 A512
-M1433 A1269
-M1434 A180
-M1436 A314
-M1437 A272
-M1438 A703
-M1439 A1010
-M1441 A411
-M1445 A1001
-M1449 A1045
-M1451 A323
-M1454 A530
-M1455 A6
-M1458 A1026
-M1460 A931
-M1463 A1063
-M1465 A437
-M1474 A1132
-M1475 A802
-M1478 A1217
-M1479 A1079
-M1483 A359
-M1488 A882
-M1490 A205
-M1495 A518
-M1498 A868
-M1500 A1081
-M1501 A155
-M1502 A458
-M1508 A804
-M1510 A1123
-M1513 A557
-M1514 A296
-M1519 A740
-M1521 A1310
-M1523 A704
-M1524 A164
-M1525 A934
-M1530 A508
-M1533 A49
-M1534 A105
-M1535 A381
-M1539 A1175
-M1540 A29
-M1542 A704
-M1544 A374
-M1546 A2
-M1548 A313
-M1549 A1068
-M1551 A343
-M1554 A462
-M1555 A497
-M1556 A1093
-M1558 A1129
-M1561 A1288
-M1564 A362
-M1565 A1261
-M1567 A1327
-M1568 A376
-M1569 A975
-M1571 A935
-M1576 A1270
-M1577 A668
-M1578 A335
-M1579 A237
-M1581 A620
-M1582 A733
-M1584 A813
-M1586 A397
-M1587 A987
-M1589 A572
-M1591 A1254
-M1592 A1092
-M1593 A932
-M1594 A929
-M1595 A752
-M1597 A1293
-M1598 A1071
-M1599 A1114
-M1600 A1144
-M1601 A333
-M1604 A600
-M1605 A563
-M1606 A607
-M1608 A597
-M1611 A458
-M1616 A315
-M1617 A959
-M1621 A1176
-M1624 A1076
-M1637 A704
-M1638 A810
-M1641 A734
-M1646 A1276
-M1647 A44
-M1648 A1012
-M1655 A150
-M1657 A544
-M1661 A583
-M1662 A343
-M1665 A433
-M1666 A759
-M1673 A455
-M1674 A912
-M1675 A799
-M1676 A315
-M1677 A1392
-M1680 A112
-M1681 A1014
-M1683 A347
-M1685 A999
-M1688 A1310
-M1689 A656
-M1690 A1155
-M1691 A654
-M1694 A965
-M1699 A1151
-M1701 A437
-M1708 A1380
-M1711 A388
-M1712 A330
-M1713 A1131
-M1714 A1093
-M1716 A22
-M1717 A221
-M1718 A456
-M1722 A176
-M1725 A1097
-M1739 A284
-M1741 A1174
-M1742 A446
-M1743 A860
-M1744 A31
-M1751 A412
-M1755 A532
-M1759 A1146
-M1760 A1130
-M1762 A492
-M1763 A1027
-M1764 A382
-M1765 A1055
-M1768 A29
-M1771 A651
-M1772 A575
-M1773 A1309
-M1775 A645
-M1777 A704
-M1778 A770
-M1783 A18
-M1784 A806
-M1786 A889
-M1787 A654
-M1788 A193
-M1789 A726
-M1794 A1280
-M1798 A1135
-M1799 A638
-M1801 A714
-M1803 A1244
-M1805 A1208
-M1806 A309
-M1812 A216
-M1813 A176
-M1814 A757
-M1815 A215
-M1816 A6
-M1820 A183
-M1824 A955
-M1825 A780
-M1827 A780
-M1829 A885
-M1833 A1116
-M1835 A74
-M1838 A57
-M1840 A627
-M1841 A237
-M1846 A122
-M1847 A177
-M1848 A575
-M1849 A546
-M1852 A748
-M1853 A1383
-M1856 A1353
-M1857 A1215
-M1859 A497
-M1861 A680
-M1862 A1123
-M1864 A843
-M1868 A55
-M1870 A934
-M1871 A75
-M1875 A1066
-M1877 A1154
-M1879 A707
-M1881 A1261
-M1885 A1091
-M1886 A1292
-M1890 A912
-M1891 A1269
-M1892 A218
-M1894 A125
-M1895 A1061
-M1898 A375
-M1900 A1347
-M1901 A105
-M1902 A590
-M1903 A1014
-M1905 A894
-M1906 A404
-M1908 A688
-M1909 A862
-M1910 A18
-M1914 A538
-M1915 A1337
-M1917 A1123
-M1920 A329
-M1921 A772
-M1925 A942
-M1926 A322
-M1929 A191
-M1937 A276
-M1938 A896
-M1939 A758
-M1941 A246
-M1944 A1323
-M1945 A861
-M1948 A592
-M1949 A653
-M1955 A551
-M1959 A660
-M1961 A59
-M1965 A511
-M1973 A1010
-M1974 A380
-M1975 A849
-M1978 A433
-M1979 A246
-M1981 A1281
-M1982 A807
-M1989 A935
-M1994 A508
-M1995 A1161
-M1998 A1250
-M2000 A508
-M2005 A673
-M2011 A567
-M2015 A849
-M2016 A393
-M2017 A762
-M2018 A911
-M2019 A994
-M2020 A352
-M2022 A175
-M2024 A654
-M2025 A487
-M2026 A1133
-M2028 A14
-M2034 A929
-M2035 A1234
-M2036 A35
-M2038 A968
-M2040 A516
-M2041 A516
-M2042 A810
-M2046 A1303
-M2047 A459
-M2049 A901
-M2055 A1376
-M2056 A981
-M2059 A1323
-M2062 A673
-M2065 A705
-M2070 A223
-M2071 A1091
-M2076 A1320
-M2079 A669
-M2082 A758
-M2083 A1181
-M2084 A1019
-M2085 A877
-M2088 A1220
-M2091 A759
-M2095 A518
-M2099 A102
-M2103 A79
-M2104 A1193
-M2105 A318
-M2107 A1032
-M2109 A965
-M2110 A406
-M2111 A1294
-M2113 A1094
-M2115 A141
-M2118 A290
-M2122 A163
-M2124 A40
-M2127 A320
-M2128 A782
-M2129 A237
-M2133 A746
-M2134 A22
-M2137 A123
-M2138 A894
-M2139 A1351
-M2151 A432
-M2152 A1006
-M2153 A1033
-M2156 A1282
-M2160 A128
-M2164 A608
-M2165 A446
-M2166 A449
-M2168 A521
-M2170 A415
-M2171 A838
-M2174 A366
-M2176 A171
-M2177 A27
-M2181 A445
-M2184 A1340
-M2192 A483
-M2193 A353
-M2195 A72
-M2199 A423
-M2205 A1226
-M2207 A746
-M2209 A1229
-M2210 A563
-M2212 A334
-M2221 A150
-M2222 A102
-M2223 A1111
-M2225 A342
-M2227 A1108
-M2230 A616
-M2231 A1051
-M2232 A402
-M2237 A45
-M2238 A666
-M2240 A261
-M2244 A356
-M2245 A1226
-M2248 A545
-M2249 A864
-M2250 A1327
-M2251 A524
-M2253 A732
-M2255 A319
-M2260 A1165
-M2262 A589
-M2263 A392
-M2268 A1301
-M2270 A283
-M2274 A151
-M2275 A513
-M2276 A7
-M2278 A575
-M2279 A745
-M2282 A632
-M2284 A1342
-M2286 A958
-M2287 A497
-M2289 A1140
-M2292 A22
-M2293 A424
-M2294 A179
-M2297 A1135
-M2298 A1119
-M2307 A84
-M2308 A1202
-M2310 A113
-M2311 A1338
-M2312 A232
-M2314 A297
-M2315 A1165
-M2316 A1315
-M2317 A827
-M2321 A989
-M2323 A191
-M2326 A1165
-M2328 A1344
-M2329 A582
-M2331 A1084
-M2337 A1394
-M2344 A627
-M2352 A326
-M2356 A472
-M2358 A361
-M2361 A335
-M2364 A521
-M2369 A413
-M2372 A440
-M2375 A1139
-M2377 A1142
-M2380 A736
-M2383 A942
-M2389 A361
-M2391 A1301
-M2392 A1144
-M2393 A1081
-M2395 A1298
-M2396 A911
-M2399 A129
-M2401 A481
-M2402 A978
-M2403 A344
-M2406 A1386
-M2409 A126
-M2411 A682
-M2413 A1153
-M2414 A1002
-M2418 A1131
-M2420 A1336
-M2421 A703
-M2425 A629
-M2428 A1297
-M2429 A1289
-M2430 A313
-M2434 A795
-M2435 A1036
-M2436 A869
-M2437 A483
-M2439 A296
-M2442 A1349
-M2448 A163
-M2449 A968
-M2450 A606
-M2451 A515
-M2452 A1111
-M2454 A416
-M2457 A948
-M2463 A1167
-M2465 A1391
-M2470 A295
-M2473 A353
-M2474 A529
-M2475 A672
-M2476 A502
-M2481 A332
-M2486 A568
-M2488 A314
-M2489 A248
-M2490 A579
-M2491 A1030
-M2492 A1020
-M2493 A345
-M2495 A841
-M2496 A1130
-M2498 A835
-M2499 A674
-M2502 A410
-M2503 A776
-M2508 A1028
-M2509 A225
-M2510 A603
-M2513 A553
-M2517 A520
-M2520 A253
-M2521 A88
-M2535 A1111
-M2538 A806
-M2539 A523
-M2542 A837
-M2543 A835
-M2547 A651
-M2548 A926
-M2553 A1014
-M2556 A1309
-M2558 A125
-M2560 A204
-M2562 A445
-M2567 A96
-M2568 A129
-M2571 A1031
-M2572 A897
-M2577 A1093
-M2579 A137
-M2583 A1077
-M2584 A534
-M2590 A1153
-M2592 A990
-M2601 A1339
-M2602 A319
-M2603 A830
-M2607 A1322
-M2611 A545
-M2613 A597
-M2615 A215
-M2616 A1307
-M2617 A247
-M2619 A1020
-M2620 A1268
-M2624 A435
-M2625 A770
-M2627 A872
-M2628 A790
-M2629 A1011
-M2630 A1371
-M2632 A1270
-M2634 A740
-M2636 A937
-M2638 A422
-M2639 A1048
-M2641 A169
-M2642 A312
-M2644 A25
-M2645 A1395
-M2646 A50
-M2648 A45
-M2650 A82
-M2652 A1081
-M2659 A468
-M2663 A693
-M2664 A860
-M2666 A696
-M2669 A711
-M2675 A711
-M2678 A907
-M2680 A872
-M2681 A1333
-M2682 A402
-M2689 A757
-M2691 A635
-M2692 A975
-M2695 A667
-M2698 A1184
-M2700 A1213
-M2701 A819
-M2703 A1345
-M2704 A811
-M2706 A1074
-M2709 A970
-M2714 A54
-M2716 A599
-M2718 A1136
-M2721 A32
-M2723 A1346
-M2725 A832
-M2726 A1352
-M2731 A91
-M2736 A1093
-M2738 A16
-M2741 A1053
-M2755 A1116
-M2757 A450
-M2760 A1008
-M2763 A1356
-M2764 A1097
-M2770 A424
-M2772 A362
-M2773 A692
-M2776 A627
-M2777 A393
-M2778 A280
-M2780 A563
-M2786 A1197
-M2787 A694
-M2788 A563
-M2789 A1056
-M2790 A10
-M2793 A1328
-M2794 A1273
-M2795 A167
-M2796 A1217
-M2798 A790
-M2803 A725
-M2809 A22
-M2818 A20
-M2819 A243
-M2823 A1267
-M2826 A1036
-M2827 A455
-M2829 A100
-M2833 A575
-M2835 A296
-M2836 A1376
-M2837 A935
-M2840 A294
-M2844 A1292
-M2845 A1388
-M2851 A391
-M2852 A461
-M2857 A298
-M2859 A1273
-M2860 A1253
-M2864 A689
-M2865 A149
-M2868 A799
-M2869 A762
-M2876 A1240
-M2877 A139
-M2878 A204
-M2882 A1092
-M2883 A1130
-M2886 A237
-M2887 A291
-M2888 A279
-M2891 A30
-M2896 A502
-M2898 A313
-M2902 A1081
-M2905 A554
-M2908 A1258
-M2910 A759
-M2915 A934
-M2917 A940
-M2918 A987
-M2921 A586
-M2922 A1135
-M2923 A1244
-M2924 A727
-M2926 A862
-M2929 A1308
-M2930 A326
-M2931 A1123
-M2934 A1262
-M2938 A1290
-M2939 A236
-M2942 A1173
-M2944 A253
-M2945 A1041
-M2948 A1082
-M2949 A1203
-M2954 A1153
-M2955 A319
-M2959 A886
-M2961 A32
-M2962 A1267
-M2963 A449
-M2965 A568
-M2966 A114
-M2967 A160
-M2969 A1111
-M2971 A1232
-M2979 A958
-M2980 A1381
-M2982 A47
-M2984 A1082
-M2995 A345
-M2997 A1194
-M2998 A972
-M3001 A1315
-M3002 A616
-M3003 A176
-M3004 A972
-M3008 A134
-M3010 A824
-M3017 A448
-M3024 A1262
-M3026 A611
-M3028 A1211
-M3030 A128
-M3031 A755
-M3035 A1341
-M3036 A41
-M3039 A67
-M3041 A1231
-M3042 A624
-M3044 A845
-M3045 A174
-M3046 A198
-M3050 A180
-M3054 A1171
-M3055 A847
-M3056 A1299
-M3058 A744
-M3060 A894
-M3061 A1260
-M3062 A723
-M3064 A1081
-M3067 A396
-M3069 A703
-M3070 A407
-M3076 A941
-M3080 A139
-M3089 A677
-M3091 A71
-M3092 A1052
-M3095 A477
-M3096 A4
-M3098 A1016
-M3099 A312
-M3100 A404
-M3102 A195
-M3103 A365
-M3108 A137
-M3111 A270
-M3113 A1366
-M3116 A285
-M3117 A448
-M3118 A110
-M3119 A115
-M3120 A931
-M3123 A8
-M3124 A177
-M3125 A819
-M3130 A244
-M3132 A1303
-M3135 A884
-M3138 A1387
-M3139 A1000
-M3140 A415
-M3143 A1271
-M3148 A902
-M3150 A176
-M3155 A458
-M3156 A237
-M3158 A1217
-M3160 A1319
-M3161 A324
-M3163 A1296
-M3164 A142
-M3165 A43
-M3166 A845
-M3170 A754
-M3173 A1078
-M3175 A1218
-M3179 A1085
-M3188 A270
-M3193 A1370
-M3196 A992
-M3198 A445
-M3200 A1220
-M3203 A1077
-M3205 A1213
-M3206 A501
-M3209 A647
-M3212 A472
-M3213 A886
-M3228 A153
-M3233 A392
-M3241 A32
-M3249 A1322
-M3256 A889
-M3258 A509
-M3262 A1055
-M3267 A24
-M3270 A374
-M3271 A795
-M3274 A1352
-M3281 A668
-M3284 A312
-M3287 A152
-M3290 A1344
-M3294 A1261
-M3300 A160
-M3302 A1228
-M3306 A47
-M3307 A35
-M3317 A1063
-M3331 A1145
-M3345 A487
-M3351 A198
-M3353 A1014
-M3355 A447
-M3358 A1030
-M3366 A1217
-M3368 A851
-M3370 A1362
-M3374 A1364
-M3380 A1102
-M3382 A1159
-M3384 A1154
-M3386 A958
-M3390 A522
-M3391 A684
-M3392 A354
-M3393 A854
-M3394 A1027
-M3404 A1259
-M3414 A938
-M3417 A407
-M3421 A1006
-M3424 A448
-M3435 A865
-M3444 A134
-M3447 A958
-M3448 A689
-M3452 A1370
-M3457 A526
-M3461 A762
-M3463 A560
-M3466 A935
-M3468 A588
-M3470 A550
-M3475 A302
-M3476 A491
-M3477 A978
-M3479 A1382
-M3480 A291
-M3483 A1055
-M3484 A1386
-M3488 A184
-M3489 A926
-M3492 A1022
-M3493 A1105
-M3498 A487
-M3502 A673
-M3503 A496
-M3510 A496
-M3511 A449
-M3512 A268
-M3514 A249
-M3515 A55
-M3530 A731
-M3531 A136
-M3537 A283
-M3542 A981
-M3560 A314
-M3566 A1347
-M3569 A819
-M3570 A1155
-M3574 A100
-M3580 A397
-M3581 A1088
-M3582 A480
-M3586 A1324
-M3587 A686
-M3589 A1152
-M3591 A41
-M3592 A164
-M3597 A971
-M3599 A145
-M3603 A27
-M3618 A880
-M3619 A810
-M3621 A1389
-M3624 A1261
-M3627 A1100
-M3641 A929
-M3647 A1155
-M3649 A981
-M3654 A992
-M3662 A496
-M3663 A479
-M3668 A488
-M3669 A1220
-M3674 A1173
-M3675 A22
-M3680 A969
-M3681 A397
-M3696 A71
-M3697 A720
-M3699 A128
-M3704 A1217
-M3709 A76
-M3713 A500
-M3714 A373
-M3726 A208
-M3729 A1081
-M3732 A26
-M3733 A499
-M3734 A459
-M3741 A757
-M3746 A361
-M3751 A602
-M3762 A1094
-M3768 A913
-M3779 A1144
-M3788 A1084
-M3805 A817
-M3811 A267
-M3814 A1327
-M3817 A987
-M3821 A1311
-M3824 A475
-M3830 A897
-M3841 A669
-M3843 A74
-M3844 A984
-M3846 A781
-M3858 A1352
-M3867 A1069
-M3869 A213
-M3879 A258
-M3884 A767
-M3889 A1184
-M3891 A402
-M3894 A64
-M3907 A407
-M3908 A893
-M3910 A18
-M3911 A673
-M3913 A904
-M3918 A958
-M3921 A701
-M3925 A32
-M3928 A951
-M3929 A1010
-M3936 A64
-M3942 A776
-M3947 A402
-M3952 A1082
-M3955 A1295
-M3956 A798
-M3967 A597
-M3969 A952
-M3970 A1005
-M3980 A555
-M3981 A20
-M3987 A239
-M3988 A927
-M3994 A786
-M4009 A784
-M4011 A193
-M4023 A1134
-M4025 A1124
-M4035 A189
-M4046 A1070
-M4047 A668
-M4049 A1005
-M4050 A832
-M4053 A1176
-M4054 A152
-M4055 A998
-M4057 A763
-M4058 A362
-M4059 A1343
-M4063 A899
-M4066 A25
-M4068 A192
-M4077 A24
-M4085 A157
-M4097 A921
-M4102 A847
-M4112 A733
-M4124 A618
-M4128 A243
-M4132 A565
-M4137 A254
-M4141 A182
-M4145 A1005
-M4150 A125
-M4158 A830
-M4165 A837
-M4168 A537
-M4171 A74
-M4173 A740
-M4180 A1203
-M4182 A1395
-M4183 A1
-M4202 A752
-M4212 A190
-M4216 A774
-M4222 A905
-M4223 A161
-M4230 A1136
-M4231 A877
-M4239 A262
-M4247 A668
-M4252 A1077
-M4256 A668
-M4268 A536
-M4277 A342
-M4279 A312
-M4282 A416
-M4290 A291
-M4293 A280
-M4303 A387
-M4313 A511
-M4331 A995
-M4344 A813
-M4350 A343
-M4357 A163
-M4379 A792
-M4381 A829
-M4387 A738
-M4401 A1146
-M4404 A1005
-M4411 A1379
-M4421 A943
-M4437 A69
-M4446 A1047
-M4458 A500
-M4462 A854
-M4466 A1378
-M4485 A720
-M4486 A1030
-M4490 A1320
-M4491 A50
-M4492 A1249
-M4493 A216
-M4494 A841
-M4496 A511
-M4530 A1386
-M4532 A1336
-M4535 A1181
-M4536 A1255
-M4537 A668
-M4543 A1308
-M4556 A45
-M4557 A560
-M4564 A1174
-M4569 A1042
-M4573 A1231
-M4575 A341
-M4577 A1136
-M4584 A891
-M4594 A1255
-M4599 A280
-M4604 A825
-M4631 A534
-M4636 A507
-M4638 A174
-M4643 A463
-M4649 A1270
-M4650 A1099
-M4667 A479
-M4677 A84
-M4679 A596
-M4690 A893
-M4694 A901
-M4695 A323
-M4730 A1344
-M4739 A813
-M4741 A1118
-M4744 A1273
-M4752 A824
-M4761 A1297
-M4765 A919
-M4778 A910
-M4791 A813
-M4821 A1020
-M4822 A206
-M4827 A765
-M4830 A502
-M4842 A1195
-M4852 A397
-M4863 A1288
-M4927 A161
-M4930 A231
-M4942 A1081
-M4949 A801
-M4950 A801
-M4960 A607
-M4979 A1285
-M5014 A1250
diff --git a/gammagl/datasets/acm4dhn.py b/gammagl/datasets/acm4dhn.py
new file mode 100644
index 00000000..e69de29b
diff --git a/gammagl/models/dhn.py b/gammagl/models/dhn.py
new file mode 100644
index 00000000..e69de29b
diff --git a/gammagl/utils/simple_path.py b/gammagl/utils/simple_path.py
new file mode 100644
index 00000000..e69de29b
diff --git a/tests/utils/test_find_all_simple_paths.py b/tests/utils/test_find_all_simple_paths.py
new file mode 100644
index 00000000..e69de29b

From ec1e5f1771df92d366745827ea2bdc0cbaf6f00b Mon Sep 17 00:00:00 2001
From: liberty <2700472871@qq.com>
Date: Mon, 1 Jul 2024 21:11:10 +0800
Subject: [PATCH 4/8] Add dataset and test code

---
 docs/source/api/gammagl.utils.rst         |   3 +-
 examples/dhn/dhn_trainer.py               | 119 +++++-----------------
 gammagl/datasets/__init__.py              |   4 +-
 gammagl/datasets/acm4dhn.py               |  83 +++++++++++++++
 gammagl/layers/conv/__init__.py           |   5 +-
 gammagl/layers/conv/dhn_conv.py           |  69 +++++--------
 gammagl/models/__init__.py                |   8 +-
 gammagl/models/dhn.py                     |  27 +++++
 gammagl/utils/__init__.py                 |   5 +-
 gammagl/utils/simple_path.py              |  50 +++++++++
 tests/utils/test_find_all_simple_paths.py |  23 +++++
 11 files changed, 251 insertions(+), 145 deletions(-)

diff --git a/docs/source/api/gammagl.utils.rst b/docs/source/api/gammagl.utils.rst
index 9f988f14..185e93d7 100644
--- a/docs/source/api/gammagl.utils.rst
+++ b/docs/source/api/gammagl.utils.rst
@@ -27,4 +27,5 @@ gammagl.utils
     gammagl.utils.to_scipy_sparse_matrix
     gammagl.utils.read_embeddings
     gammagl.utils.homophily
-    gammagl.utils.get_train_val_test_split
\ No newline at end of file
+    gammagl.utils.get_train_val_test_split
+    gammagl.utils.find_all_simple_paths
\ No newline at end of file
diff --git a/examples/dhn/dhn_trainer.py b/examples/dhn/dhn_trainer.py
index d0cb1ee2..c62417b1 100644
--- a/examples/dhn/dhn_trainer.py
+++ b/examples/dhn/dhn_trainer.py
@@ -5,98 +5,30 @@
 from tensorlayerx.model import TrainOneStep
 from sklearn.metrics import roc_auc_score
 
-from gammagl.data import HeteroGraph
-from gammagl.layers.conv.dhn_conv import DHNModel, NODE_TYPE, K_HOP, NUM_FEA, NUM_NEIGHBOR, BATCH_SIZE, type2idx
-from gammagl.utils import k_hop_subgraph
+from gammagl.models import DHNModel
+from gammagl.datasets import ACM4DHN
+from gammagl.utils import k_hop_subgraph,find_all_simple_paths
 
 # The UserWarning is ignored globally
 import warnings
 
-warnings.filterwarnings("ignore", category=UserWarning)
-
-random.seed(0)
-
-G = HeteroGraph()
-
-
-def load_ACM(test_ratio=0.2):
-    edge_index_M = []
-    edge_index_A = []
-
-    with open(args.dataset_path, 'r') as f:
-        for line in f.readlines():
-            src, dst = line.strip().split()
-            src_type, src_id = src[0], src[1:]  # Resolves the source node type and ID
-            dst_type, dst_id = dst[0], dst[1:]  # Resolve the target node type and ID
-
-            # Convert the node ID to an integer index and place it in a list
-            if src[0] == 'M':
-                edge_index_M.append(int(src_id))
-            elif src[0] == 'A':
-                edge_index_A.append(-int(src_id) - 1)
-
-            if dst[0] == 'M':
-                edge_index_M.append(int(dst_id))
-            elif dst[0] == 'A':
-                edge_index_A.append(-int(dst_id) - 1)
-
-    edge_index = tlx.convert_to_tensor([edge_index_M, edge_index_A])
-    G['M', 'MA', 'A'].edge_index = edge_index
-
-    # Computed split point
-    sp = 1 - test_ratio * 2
-    num_edge = len(edge_index_M)
-    sp1 = int(num_edge * sp)
-    sp2 = int(num_edge * test_ratio)
-
-    G_train = HeteroGraph()
-    G_val = HeteroGraph()
-    G_test = HeteroGraph()
+type2idx = {
+    'M': 0,
+    'A': 1,
+    # 'C': 2,
+    # 'T': 3
+}
 
-    # Divide the training set, the verification set, and the test set
-    G_train['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor([edge_index_M[:sp1], edge_index_A[:sp1]])
-    G_val['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor([edge_index_M[sp1:sp1 + sp2], edge_index_A[sp1:sp1 + sp2]])
-    G_test['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor([edge_index_M[sp1 + sp2:], edge_index_A[sp1 + sp2:]])
+NODE_TYPE = len(type2idx)
+K_HOP = 2
 
-    print(
-        f"all edge: {len(G['M', 'MA', 'A'].edge_index[0])}, train edge: {len(G_train['M', 'MA', 'A'].edge_index[0])}, val edge: {len(G_val['M', 'MA', 'A'].edge_index[0])}, test edge: {len(G_test['M', 'MA', 'A'].edge_index[0])}")
+NUM_FEA = (K_HOP + 2) * 4 + NODE_TYPE
+NUM_NEIGHBOR = 5
+BATCH_SIZE=32
 
-    return G_train, G_val, G_test
-
-
-def find_all_simple_paths(edge_index, src, dest, max_length):
-    # Converts edge_index to an adjacency list representation
-    num_nodes = max(edge_index[0].max().item(),
-                    edge_index[1].max().item(),
-                    -edge_index[0].min().item(),
-                    -edge_index[1].min().item(),
-                    abs(src.item())) + 1
-    adj_list = [[] for _ in range(num_nodes)]
-    for u, v in zip(edge_index[0].tolist(), edge_index[1].tolist()):
-        adj_list[u].append(v)
-
-    src = src.item()
-
-    paths = []
-    visited = set()
-    stack = [(src, [src])]
-
-    while stack:
-        (node, path) = stack.pop()
-
-        if node == dest:
-            paths.append(path)
-        elif len(path) < max_length:
-            for neighbor in adj_list[node]:
-                if neighbor not in path:
-                    visited.add((node, neighbor))
-                    stack.append((neighbor, path + [neighbor]))
-            for neighbor in adj_list[node]:
-                if (node, neighbor) in visited:
-                    visited.remove((node, neighbor))
-
-    return paths
+warnings.filterwarnings("ignore", category=UserWarning)
 
+tlx.set_seed(10)
 
 def dist_encoder(src, dest, G, K_HOP):
     if (G.size(1) == 0):
@@ -318,10 +250,16 @@ def result(self):
 def main(args):
     if str.lower(args.dataset) not in ['acm']:
         raise ValueError('Unknown dataset: {}'.format(args.dataset))
-    if str.lower(args.dataset) == 'acm':
-        G_train, G_val, G_test = load_ACM(test_ratio=args.test_ratio)
+    if str.lower(args.dataset)=='acm':
+        data = ACM4DHN(args.dataset_path)
+
+    graph=data[0]
+
+    G_train=graph['train']
+    G_val=graph['val']
+    G_test=graph['test']
 
-    m = DHNModel()
+    m = DHNModel(NUM_FEA,BATCH_SIZE,NUM_NEIGHBOR)
 
     optim = tlx.optimizers.Adam(lr=args.lr, weight_decay=args.drop_rate)
     train_weights = m.trainable_weights
@@ -375,13 +313,12 @@ def main(args):
 if __name__ == '__main__':
     # parameters setting
     parser = argparse.ArgumentParser()
-    parser.add_argument("--test_ratio", type=float, default=0.3, help="ratio of dividing the data set")
-    parser.add_argument("--one_hot", type=bool, default=True, help="use one-hot encoding")
+    parser.add_argument("--one_hot",type=bool,default=True,help="use one-hot encoding")
     parser.add_argument("--lr", type=float, default=0.001, help="learnin rate")
     parser.add_argument("--n_epoch", type=int, default=100, help="number of epoch")
     parser.add_argument("--drop_rate", type=float, default=0.01, help="drop_rate")
     parser.add_argument('--dataset', type=str, default='acm', help='dataset')
-    parser.add_argument("--dataset_path", type=str, default=r"MA.txt")
+    parser.add_argument("--dataset_path", type = str, default = r"",help='dataset_path')
 
     args = parser.parse_args()
-    main(args)
+    main(args)
\ No newline at end of file
diff --git a/gammagl/datasets/__init__.py b/gammagl/datasets/__init__.py
index 2dc598bb..4b0744b3 100644
--- a/gammagl/datasets/__init__.py
+++ b/gammagl/datasets/__init__.py
@@ -22,6 +22,7 @@
 from .molecule_net import MoleculeNet
 from .acm4heco import ACM4HeCo
 from .yelp import Yelp
+from .acm4dhn import ACM4DHN
 
 __all__ = [
     'ACM4HeCo',
@@ -46,7 +47,8 @@
     'WikiCS',
     'MoleculeNet',
     'NGSIM_US_101',
-    'Yelp'
+    'Yelp',
+    'ACM4DHN'
 ]
 
 classes = __all__
diff --git a/gammagl/datasets/acm4dhn.py b/gammagl/datasets/acm4dhn.py
index e69de29b..b65d2010 100644
--- a/gammagl/datasets/acm4dhn.py
+++ b/gammagl/datasets/acm4dhn.py
@@ -0,0 +1,83 @@
+import os.path as osp
+import tensorlayerx as tlx
+from gammagl.data import download_url, InMemoryDataset, HeteroGraph
+from typing import Callable, List, Optional
+
+
+class ACM4DHN(InMemoryDataset):
+    url = 'https://raw.githubusercontent.com/BUPT-GAMMA/HDE/main/ds/imdb'
+    test_ratio = 0.3
+
+    def __init__(self, root: Optional[str] = None, transform: Optional[Callable] = None,
+                 pre_transform: Optional[Callable] = None, pre_filter: Optional[Callable] = None,
+                 force_reload: bool = False):
+        super().__init__(root, transform, pre_transform, force_reload=force_reload)
+        self.data, self.slices = self.load_data(self.processed_paths[0])
+
+    @property
+    def raw_file_names(self) -> List[str]:
+        return [
+            'MA.txt'
+        ]
+
+    @property
+    def processed_file_names(self) -> str:
+        return tlx.BACKEND + '_data.pt'
+
+    def download(self):
+        for i in range(0, len(self.raw_file_names)):
+            download_url(f'{self.url}/{self.raw_file_names[i]}', self.raw_dir)
+
+    def process(self):
+        G = HeteroGraph()
+        edge_index_M = []
+        edge_index_A = []
+
+        path = osp.join(self.raw_dir, 'MA.txt')
+
+        with open(path, 'r') as f:
+            for line in f.readlines():
+                src, dst = line.strip().split()
+                src_type, src_id = src[0], src[1:]  # Resolves the source node type and ID
+                dst_type, dst_id = dst[0], dst[1:]  # Resolve the target node type and ID
+
+                # Convert the node ID to an integer index and place it in a list
+                if src[0] == 'M':
+                    edge_index_M.append(int(src_id))
+                elif src[0] == 'A':
+                    edge_index_A.append(-int(src_id) - 1)
+
+                if dst[0] == 'M':
+                    edge_index_M.append(int(dst_id))
+                elif dst[0] == 'A':
+                    edge_index_A.append(-int(dst_id) - 1)
+
+        edge_index = tlx.convert_to_tensor([edge_index_M, edge_index_A])
+        G['M', 'MA', 'A'].edge_index = edge_index
+
+        # Computed split point
+        sp = 1 - self.test_ratio * 2
+        num_edge = len(edge_index_M)
+        sp1 = int(num_edge * sp)
+        sp2 = int(num_edge * self.test_ratio)
+
+        G_train = HeteroGraph()
+        G_val = HeteroGraph()
+        G_test = HeteroGraph()
+
+        # Divide the training set, the verification set, and the test set
+        G_train['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor([edge_index_M[:sp1], edge_index_A[:sp1]])
+        G_val['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor(
+            [edge_index_M[sp1:sp1 + sp2], edge_index_A[sp1:sp1 + sp2]])
+        G_test['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor([edge_index_M[sp1 + sp2:], edge_index_A[sp1 + sp2:]])
+
+        print(
+            f"all edge: {len(G['M', 'MA', 'A'].edge_index[0])}, train edge: {len(G_train['M', 'MA', 'A'].edge_index[0])}, val edge: {len(G_val['M', 'MA', 'A'].edge_index[0])}, test edge: {len(G_test['M', 'MA', 'A'].edge_index[0])}")
+
+        G['train'] = G_train  # training set
+        G['val'] = G_val  # valuation set
+        G['test'] = G_test  # testing set
+
+        if self.pre_transform is not None:
+            G = self.pre_transform(G)
+        self.save_data(self.collate([G]), self.processed_paths[0])
diff --git a/gammagl/layers/conv/__init__.py b/gammagl/layers/conv/__init__.py
index 0162a6cb..3ba78108 100644
--- a/gammagl/layers/conv/__init__.py
+++ b/gammagl/layers/conv/__init__.py
@@ -33,6 +33,8 @@
 from .magcl_conv import MAGCLConv
 from .fusedgat_conv import FusedGATConv
 from .hid_conv import Hid_conv
+from .dhn_conv import DHNConv
+
 __all__ = [
     'MessagePassing',
     'GCNConv',
@@ -68,7 +70,8 @@
     'MAGCLConv',
     'FusedGATConv',
     'Hid_conv',
-    'HEATlayer'
+    'HEATlayer',
+    'DHNConv'
 ]
 
 classes = __all__
diff --git a/gammagl/layers/conv/dhn_conv.py b/gammagl/layers/conv/dhn_conv.py
index dc651a6a..be994c0c 100644
--- a/gammagl/layers/conv/dhn_conv.py
+++ b/gammagl/layers/conv/dhn_conv.py
@@ -1,43 +1,38 @@
 import tensorlayerx as tlx
 from gammagl.layers.conv import MessagePassing
 
-type2idx = {
-    'M': 0,
-    'A': 1,
-    # 'C': 2,
-    # 'T': 3
-}
-
-NODE_TYPE = len(type2idx)
-K_HOP = 2
-
-NUM_FEA = (K_HOP + 2) * 4 + NODE_TYPE
-NUM_NEIGHBOR = 5
-BATCH_SIZE=32
-
 
 class DHNConv(MessagePassing):
-    def __init__(self):
+    def __init__(self,
+                 num_fea,
+                 batch_size,
+                 num_neighbor):
         super().__init__()
-        self.lin1 = tlx.nn.Linear(in_features=2*NUM_FEA, out_features=2*BATCH_SIZE, act=tlx.nn.ELU(), W_init="xavier_uniform")
-        self.lin2 = tlx.nn.Linear(in_features=2*BATCH_SIZE+NUM_FEA, out_features=2*BATCH_SIZE, act=tlx.nn.ELU(), W_init="xavier_uniform")
-        self.lin3 = tlx.nn.Linear(in_features=2*BATCH_SIZE, out_features=2*BATCH_SIZE, act=tlx.nn.ELU(), W_init="xavier_uniform")
+        self.num_fea = num_fea
+        self.batch_size = batch_size
+        self.num_neighbor = num_neighbor
+        self.lin1 = tlx.nn.Linear(in_features=2 * num_fea, out_features=2 * batch_size, act=tlx.nn.ELU(),
+                                  W_init="xavier_uniform")
+        self.lin2 = tlx.nn.Linear(in_features=2 * batch_size + num_fea, out_features=2 * batch_size, act=tlx.nn.ELU(),
+                                  W_init="xavier_uniform")
+        self.lin3 = tlx.nn.Linear(in_features=2 * batch_size, out_features=2 * batch_size, act=tlx.nn.ELU(),
+                                  W_init="xavier_uniform")
 
     def aggregate(self, msg, edge_index, num_nodes=None, aggr=None):
-        if len(msg.shape)==4:
+        if len(msg.shape) == 4:
             return tlx.reduce_mean(msg, axis=2)
-        elif len(msg.shape)==3:
+        elif len(msg.shape) == 3:
             return tlx.reduce_mean(msg, axis=1)
 
     def forward(self, fea):
-        node = tlx.convert_to_tensor(fea[:, :NUM_FEA])
+        node = tlx.convert_to_tensor(fea[:, :self.num_fea])
 
         # Extract neigh1 and neigh2
-        neigh1 = tlx.convert_to_tensor(fea[:, NUM_FEA:NUM_FEA * (NUM_NEIGHBOR + 1)])
-        neigh1 = tlx.reshape(neigh1, [-1, NUM_NEIGHBOR, NUM_FEA])
+        neigh1 = tlx.convert_to_tensor(fea[:, self.num_fea:self.num_fea * (self.num_neighbor + 1)])
+        neigh1 = tlx.reshape(neigh1, [-1, self.num_neighbor, self.num_fea])
 
-        neigh2 = tlx.convert_to_tensor(fea[:, NUM_FEA * (NUM_NEIGHBOR + 1):])
-        neigh2 = tlx.reshape(neigh2, [-1, NUM_NEIGHBOR, NUM_NEIGHBOR, NUM_FEA])
+        neigh2 = tlx.convert_to_tensor(fea[:, self.num_fea * (self.num_neighbor + 1):])
+        neigh2 = tlx.reshape(neigh2, [-1, self.num_neighbor, self.num_neighbor, self.num_fea])
 
         # aggregate on neigh2
         neigh2_agg = self.propagate(edge_index=None, x=neigh2)
@@ -49,15 +44,15 @@ def forward(self, fea):
         )
 
         # transform into a two-dimensional tensor
-        flattened_tmp = tmp.view(2*BATCH_SIZE * NUM_NEIGHBOR, 2*NUM_FEA)
+        flattened_tmp = tmp.view(2 * self.batch_size * self.num_neighbor, 2 * self.num_fea)
         flattened_output = self.lin1(flattened_tmp)
         # Reshape the output tensor to a three-dimensional shape
-        tmp = flattened_output.view(2*BATCH_SIZE, NUM_NEIGHBOR, 2*BATCH_SIZE)
+        tmp = flattened_output.view(2 * self.batch_size, self.num_neighbor, 2 * self.batch_size)
 
         # Aggregate node and tmp
         emb = tlx.concat(
             [
-                node, self.propagate(edge_index=None,x=tmp)
+                node, self.propagate(edge_index=None, x=tmp)
             ],
             axis=1
         )
@@ -70,21 +65,3 @@ def forward(self, fea):
 
     def message(self, x, edge_index, edge_weight=None):
         return x
-
-
-class DHNModel(tlx.nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.dhn1 = DHNConv()
-        self.dhn2 = DHNConv()
-        self.lin1 = tlx.nn.Linear(in_features=4*BATCH_SIZE, out_features=BATCH_SIZE, act=tlx.nn.ELU(), W_init="xavier_uniform")
-        self.lin2 = tlx.nn.Linear(in_features=BATCH_SIZE, out_features=1, act=tlx.nn.ELU(), W_init="xavier_uniform")
-
-    def forward(self, n1, n2, label):
-        n1_emb = self.dhn1(n1)
-        n2_emb = self.dhn2(n2)
-
-        pred = self.lin1(tlx.concat([n1_emb, n2_emb], axis=1))
-        pred = self.lin2(pred)
-
-        return pred
diff --git a/gammagl/models/__init__.py b/gammagl/models/__init__.py
index c08b85c0..e6c925be 100644
--- a/gammagl/models/__init__.py
+++ b/gammagl/models/__init__.py
@@ -57,6 +57,7 @@
 from .fusedgat import FusedGATModel
 from .hid_net import Hid_net
 from .gnnlfhf import GNNLFHFModel
+from .dhn import DHNModel
 
 __all__ = [
     'HeCo',
@@ -105,8 +106,8 @@
     'MetaPath2Vec'
     'ieHGCNModel',
     'TADWModel',
-    'MGNNI_m_MLP', 
-    'MGNNI_m_att', 
+    'MGNNI_m_MLP',
+    'MGNNI_m_att',
     'NewGrace',
     'CAGCNModel',
     'CoGSLModel',
@@ -117,7 +118,8 @@
     'FusedGATModel',
     'hid_net',
     'HEAT',
-    'GNNLFHFModel'
+    'GNNLFHFModel',
+    'DHNModel'
 ]
 
 classes = __all__
diff --git a/gammagl/models/dhn.py b/gammagl/models/dhn.py
index e69de29b..2b717a17 100644
--- a/gammagl/models/dhn.py
+++ b/gammagl/models/dhn.py
@@ -0,0 +1,27 @@
+import tensorlayerx as tlx
+from gammagl.layers.conv import DHNConv
+
+
+class DHNModel(tlx.nn.Module):
+    def __init__(self,
+                 num_fea,
+                 batch_size,
+                 num_neighbor):
+        super().__init__()
+        self.num_fea = num_fea
+        self.batch_size = batch_size
+        self.num_neighbor = num_neighbor
+        self.dhn1 = DHNConv(num_fea, batch_size, num_neighbor)
+        self.dhn2 = DHNConv(num_fea, batch_size, num_neighbor)
+        self.lin1 = tlx.nn.Linear(in_features=4 * batch_size, out_features=batch_size, act=tlx.nn.ELU(),
+                                  W_init="xavier_uniform")
+        self.lin2 = tlx.nn.Linear(in_features=batch_size, out_features=1, act=tlx.nn.ELU(), W_init="xavier_uniform")
+
+    def forward(self, n1, n2, label):
+        n1_emb = self.dhn1(n1)
+        n2_emb = self.dhn2(n2)
+
+        pred = self.lin1(tlx.concat([n1_emb, n2_emb], axis=1))
+        pred = self.lin2(pred)
+
+        return pred
diff --git a/gammagl/utils/__init__.py b/gammagl/utils/__init__.py
index f57d8d09..ea3f5571 100644
--- a/gammagl/utils/__init__.py
+++ b/gammagl/utils/__init__.py
@@ -20,6 +20,7 @@
 from .shortest_path import shortest_path_distance, batched_shortest_path_distance
 from .get_split import get_train_val_test_split
 from .get_laplacian import get_laplacian
+from .simple_path import find_all_simple_paths
 
 __all__ = [
     'calc_A_norm_hat',
@@ -46,8 +47,8 @@
     'shortest_path_distance',
     'batched_shortest_path_distance',
     'get_train_val_test_split',
-    'get_laplacian'
-
+    'get_laplacian',
+    'find_all_simple_paths'
 ]
 
 classes = __all__
diff --git a/gammagl/utils/simple_path.py b/gammagl/utils/simple_path.py
index e69de29b..ee78af47 100644
--- a/gammagl/utils/simple_path.py
+++ b/gammagl/utils/simple_path.py
@@ -0,0 +1,50 @@
+def find_all_simple_paths(edge_index, src, dest, max_length):
+    r"""
+        The :obj:`find_all_simple_paths` function is used to find all simple paths (that is, paths that do not duplicate nodes) from the source node to the destination node in a given graph. The function accepts as parameters the edge index of the graph, the maximum length of the source node, the target node, and the path, and returns a list of all simple paths from the source node to the target node that do not exceed the maximum length.
+
+        Parameters
+        ----------
+        edge_index: tensor
+            A 2-D Tensor of the edge index of a graph with the shape [2, num_edges]. Each column contains two end-point indexes for one side.
+        src: tensor
+            The index of the source node.
+        dest: tensor
+            The index of the target node.
+        max_length: int
+            The maximum length of a path.
+
+        Return
+        -------
+        list[list[int]]
+            A list of all the simple paths from the source node to the destination node.
+    """
+    num_nodes = max(edge_index[0].max().item(),
+                    edge_index[1].max().item(),
+                    -edge_index[0].min().item(),
+                    -edge_index[1].min().item(),
+                    abs(src.item())) + 1
+    adj_list = [[] for _ in range(num_nodes)]
+    for u, v in zip(edge_index[0].tolist(), edge_index[1].tolist()):
+        adj_list[u].append(v)
+
+    src = src.item()
+
+    paths = []
+    visited = set()
+    stack = [(src, [src])]
+
+    while stack:
+        (node, path) = stack.pop()
+
+        if node == dest:
+            paths.append(path)
+        elif len(path) < max_length:
+            for neighbor in adj_list[node]:
+                if neighbor not in path:
+                    visited.add((node, neighbor))
+                    stack.append((neighbor, path + [neighbor]))
+            for neighbor in adj_list[node]:
+                if (node, neighbor) in visited:
+                    visited.remove((node, neighbor))
+
+    return paths
diff --git a/tests/utils/test_find_all_simple_paths.py b/tests/utils/test_find_all_simple_paths.py
index e69de29b..d39b872a 100644
--- a/tests/utils/test_find_all_simple_paths.py
+++ b/tests/utils/test_find_all_simple_paths.py
@@ -0,0 +1,23 @@
+import tensorlayerx as tlx
+from gammagl.utils.simple_path import find_all_simple_paths
+
+def test_find_all_simple_paths():
+    edge_index = tlx.convert_to_tensor([[0, 0, 1, 1, 2], [1, 2, 0, 3, 1]], dtype=tlx.int64)
+    src = tlx.convert_to_tensor([0], dtype=tlx.int64)
+    dest = tlx.convert_to_tensor([3], dtype=tlx.int64)
+    expected_paths = [[0, 1, 3], [0, 2, 1, 3]]
+    paths = find_all_simple_paths(edge_index, src, dest, max_length=4)
+    assert sorted(map(tuple, paths))==sorted(map(tuple, expected_paths))
+
+    # no path
+    src = tlx.convert_to_tensor([0], dtype=tlx.int64)
+    dest = tlx.convert_to_tensor([4], dtype=tlx.int64)
+    expected_paths = []
+    paths = find_all_simple_paths(edge_index, src, dest, max_length=4)
+    assert sorted(map(tuple, paths)) == sorted(map(tuple, expected_paths))
+
+    # same start and end point
+    dest = tlx.convert_to_tensor([0], dtype=tlx.int64)
+    expected_paths = [[0]]
+    paths = find_all_simple_paths(edge_index, src, dest, max_length=1)
+    assert sorted(map(tuple, paths))==sorted(map(tuple, expected_paths))
\ No newline at end of file

From 15e426534a951b5185a7b52a2f0ff34a6cc73360 Mon Sep 17 00:00:00 2001
From: liberty <2700472871@qq.com>
Date: Tue, 2 Jul 2024 15:50:14 +0800
Subject: [PATCH 5/8] modify the code and add test code

---
 examples/dhn/README.md         | 17 +++++++
 examples/dhn/dhn_trainer.py    | 92 ++++++++++++++++++----------------
 gammagl/datasets/acm4dhn.py    |  4 +-
 tests/datasets/test_acm4dhn.py | 15 ++++++
 4 files changed, 82 insertions(+), 46 deletions(-)
 create mode 100644 examples/dhn/README.md
 create mode 100644 tests/datasets/test_acm4dhn.py

diff --git a/examples/dhn/README.md b/examples/dhn/README.md
new file mode 100644
index 00000000..a365ebeb
--- /dev/null
+++ b/examples/dhn/README.md
@@ -0,0 +1,17 @@
+# Distance encoding based Heterogeneous graph neural Network (DHN)
+- Paper link: [https://ieeexplore.ieee.org/document/10209229](https://ieeexplore.ieee.org/document/10209229)
+- Author's code repo: [https://github.com/BUPT-GAMMA/HDE](https://github.com/BUPT-GAMMA/HDE)
+
+## Dataset Statics
+| Dataset  | # Nodes | # Edges | # Classes |
+|----------|---------|---------|-----------|
+| Cora     | 4435    | 7071    | 2         |
+
+## Results
+```bash
+TL_BACKEND="torch" python dhn_trainer.py --test_ratio 0.3 --one_hot True --k_hop 2 --num_neighbor 5 --batch_size 32 --lr 0.001 --n_epoch 100 --drop_rate 0.01 --dataset 'acm'
+```
+
+| Dataset  | Paper(AUC) | Our(th)(AUC)     |
+| -------- | ----- | ----------- | 
+| acm      | 95.07 |  95.54±0.18 | 
\ No newline at end of file
diff --git a/examples/dhn/dhn_trainer.py b/examples/dhn/dhn_trainer.py
index c62417b1..13b39689 100644
--- a/examples/dhn/dhn_trainer.py
+++ b/examples/dhn/dhn_trainer.py
@@ -7,7 +7,7 @@
 
 from gammagl.models import DHNModel
 from gammagl.datasets import ACM4DHN
-from gammagl.utils import k_hop_subgraph,find_all_simple_paths
+from gammagl.utils import k_hop_subgraph, find_all_simple_paths
 
 # The UserWarning is ignored globally
 import warnings
@@ -19,45 +19,41 @@
     # 'T': 3
 }
 
-NODE_TYPE = len(type2idx)
-K_HOP = 2
-
-NUM_FEA = (K_HOP + 2) * 4 + NODE_TYPE
-NUM_NEIGHBOR = 5
-BATCH_SIZE=32
-
 warnings.filterwarnings("ignore", category=UserWarning)
 
 tlx.set_seed(10)
 
-def dist_encoder(src, dest, G, K_HOP):
+
+def dist_encoder(src, dest, G, k_hop):
     if (G.size(1) == 0):
         paths = []
     else:
-        paths = find_all_simple_paths(G, src, dest, K_HOP + 2)
+        paths = find_all_simple_paths(G, src, dest, k_hop + 2)
 
-    cnt = [K_HOP + 1] * NODE_TYPE  # Default truncation for max_spd exceeded
+    node_type = len(type2idx)
+    cnt = [k_hop + 1] * node_type  # Default truncation for max_spd exceeded
     for path in paths:
-        res = [0] * NODE_TYPE
+        res = [0] * node_type
         for i in path:
             if i >= 0:
                 res[type2idx['M']] += 1
             else:
                 res[type2idx['A']] += 1
 
-        for k in range(NODE_TYPE):
+        for k in range(node_type):
             cnt[k] = min(cnt[k], res[k])
 
     # Generate one-hot encoding
     if args.one_hot:
-        one_hot_list = [np.eye(K_HOP + 2, dtype=np.float64)[cnt[i]]
-                        for i in range(NODE_TYPE)]
+        one_hot_list = [np.eye(k_hop + 2, dtype=np.float64)[cnt[i]]
+                        for i in range(node_type)]
         return np.concatenate(one_hot_list)
     return cnt
 
 
 def type_encoder(node):
-    res = [0] * NODE_TYPE
+    node_type = len(type2idx)
+    res = [0] * node_type
     if node.item() >= 0:
         res[type2idx['M']] = 1.0
     else:
@@ -69,15 +65,17 @@ def type_encoder(node):
 fea_batch = []
 
 
-def gen_fea_batch(G, root, fea_dict, hop):
+def gen_fea_batch(G, root, fea_dict, k_hop):
     fea_batch = []
     mini_batch.append(root)
 
-    a = [0] * (K_HOP + 2) * 4 + type_encoder(root)
+    a = [0] * (k_hop + 2) * 4 + type_encoder(root)
 
+    node_type = len(type2idx)
+    num_fea = (k_hop + 2) * 4 + node_type
     fea_batch.append(np.asarray(a,
                                 dtype=np.float32
-                                ).reshape(-1, NUM_FEA)
+                                ).reshape(-1, num_fea)
                      )
 
     # 1-order neighbor sampling
@@ -90,7 +88,7 @@ def gen_fea_batch(G, root, fea_dict, hop):
             neighbors_mask = dst == node
         neighbors = list(dst[neighbors_mask].numpy())
         neighbors.append(node.item())
-        random_choice_list = np.random.choice(neighbors, NUM_NEIGHBOR, replace=True)
+        random_choice_list = np.random.choice(neighbors, args.num_neighbor, replace=True)
         ns_1.append(random_choice_list.tolist())
     ns_1 = tlx.convert_to_tensor(ns_1)
     mini_batch.append(ns_1[0])
@@ -113,7 +111,7 @@ def gen_fea_batch(G, root, fea_dict, hop):
             neighbors_mask = dst == node
         neighbors = list(dst[neighbors_mask].numpy())
         neighbors.append(node.item())
-        random_choice_list = np.random.choice(neighbors, NUM_NEIGHBOR, replace=True)
+        random_choice_list = np.random.choice(neighbors, args.num_neighbor, replace=True)
         ns_2.append(random_choice_list.tolist())
     ns_2 = tlx.convert_to_tensor(ns_2)
 
@@ -133,13 +131,13 @@ def gen_fea_batch(G, root, fea_dict, hop):
     return np.concatenate(fea_batch, axis=1)
 
 
-def subgraph_sampling_with_DE_node_pair(G, node_pair, K_HOP=2):
+def subgraph_sampling_with_DE_node_pair(G, node_pair, k_hop=2):
     [A, B] = node_pair
 
     edge_index = tlx.concat([G['M', 'MA', 'A'].edge_index, reversed(G['M', 'MA', 'A'].edge_index)], axis=1)
 
     # Find k-hop subgraphs of A and B
-    sub_G_for_AB = k_hop_subgraph([A, B], K_HOP, edge_index)
+    sub_G_for_AB = k_hop_subgraph([A, B], k_hop, edge_index)
 
     # Remove edges using Boolean indexes
     # Note: Just remove the edges, the points remain
@@ -159,21 +157,20 @@ def subgraph_sampling_with_DE_node_pair(G, node_pair, K_HOP=2):
     # Distance from all points in the subgraph to the node pair
     SPD_based_on_node_pair = {}
     for node in sub_G_nodes:
-        tmpA = dist_encoder(A, node, sub_G_index, K_HOP)
-        tmpB = dist_encoder(B, node, sub_G_index, K_HOP)
+        tmpA = dist_encoder(A, node, sub_G_index, k_hop)
+        tmpB = dist_encoder(B, node, sub_G_index, k_hop)
 
         SPD_based_on_node_pair[node.item()] = np.concatenate([tmpA, tmpB], axis=0)
 
     A_fea_batch = gen_fea_batch(sub_G_index, A,
-                                SPD_based_on_node_pair, K_HOP)
+                                SPD_based_on_node_pair, k_hop)
     B_fea_batch = gen_fea_batch(sub_G_index, B,
-                                SPD_based_on_node_pair, K_HOP)
+                                SPD_based_on_node_pair, k_hop)
 
     return A_fea_batch, B_fea_batch
 
 
-def batch_data(G,
-               batch_size=3):
+def batch_data(G, batch_size):
     edge_index = G['M', 'MA', 'A'].edge_index
     nodes = set(tlx.convert_to_tensor(np.unique(edge_index[0].numpy()))) | set(
         tlx.convert_to_tensor(np.unique(edge_index[1].numpy())))
@@ -204,7 +201,7 @@ def batch_data(G,
             bx = batch_edge[:, i:i + 1]
 
             # Positive sample
-            posA, posB = subgraph_sampling_with_DE_node_pair(G, bx, K_HOP=K_HOP)
+            posA, posB = subgraph_sampling_with_DE_node_pair(G, bx, k_hop=args.k_hop)
             batch_A_fea.append(posA)
             batch_B_fea.append(posB)
             batch_y.append(np.asarray(by, dtype=np.float32))
@@ -213,7 +210,7 @@ def batch_data(G,
             neg_tmpB_id = random.choice(nodes_list)
             node_pair = tlx.convert_to_tensor([[bx[0].item()], [neg_tmpB_id]])
 
-            negA, negB = subgraph_sampling_with_DE_node_pair(G, node_pair, K_HOP=K_HOP)
+            negA, negB = subgraph_sampling_with_DE_node_pair(G, node_pair, k_hop=args.k_hop)
             batch_A_fea.append(negA)
             batch_B_fea.append(negB)
             batch_y.append(np.asarray(0.0, dtype=np.float32))
@@ -250,16 +247,19 @@ def result(self):
 def main(args):
     if str.lower(args.dataset) not in ['acm']:
         raise ValueError('Unknown dataset: {}'.format(args.dataset))
-    if str.lower(args.dataset)=='acm':
-        data = ACM4DHN(args.dataset_path)
+    if str.lower(args.dataset) == 'acm':
+        data = ACM4DHN(root=args.dataset_path, test_ratio=args.test_ratio)
+
+    graph = data[0]
 
-    graph=data[0]
+    G_train = graph['train']
+    G_val = graph['val']
+    G_test = graph['test']
 
-    G_train=graph['train']
-    G_val=graph['val']
-    G_test=graph['test']
+    node_type = len(type2idx)
+    num_fea = (args.k_hop + 2) * 4 + node_type
 
-    m = DHNModel(NUM_FEA,BATCH_SIZE,NUM_NEIGHBOR)
+    m = DHNModel(num_fea, args.batch_size, args.num_neighbor)
 
     optim = tlx.optimizers.Adam(lr=args.lr, weight_decay=args.drop_rate)
     train_weights = m.trainable_weights
@@ -276,7 +276,7 @@ def main(args):
 
         # train
         m.set_train()
-        tra_batch_A_fea, tra_batch_B_fea, tra_batch_y = batch_data(G_train, BATCH_SIZE).__next__()
+        tra_batch_A_fea, tra_batch_B_fea, tra_batch_y = batch_data(G_train, args.batch_size).__next__()
         tra_out = m(tra_batch_A_fea, tra_batch_B_fea, tra_batch_y)
 
         data = {
@@ -292,7 +292,7 @@ def main(args):
 
         # val
         m.set_eval()
-        val_batch_A_fea, val_batch_B_fea, val_batch_y = batch_data(G_val, BATCH_SIZE).__next__()
+        val_batch_A_fea, val_batch_B_fea, val_batch_y = batch_data(G_val, args.batch_size).__next__()
         val_out = m(val_batch_A_fea, val_batch_B_fea, val_batch_y)
 
         val_loss = tlx.losses.sigmoid_cross_entropy(output=val_out, target=tlx.convert_to_tensor(val_batch_y))
@@ -301,7 +301,7 @@ def main(args):
         print("val: ", val_loss.item(), val_auc)
 
         # test
-        test_batch_A_fea, test_batch_B_fea, test_batch_y = batch_data(G_test, BATCH_SIZE).__next__()
+        test_batch_A_fea, test_batch_B_fea, test_batch_y = batch_data(G_test, args.batch_size).__next__()
         test_out = m(test_batch_A_fea, test_batch_B_fea, test_batch_y)
 
         test_loss = tlx.losses.sigmoid_cross_entropy(output=test_out, target=tlx.convert_to_tensor(test_batch_y))
@@ -313,12 +313,16 @@ def main(args):
 if __name__ == '__main__':
     # parameters setting
     parser = argparse.ArgumentParser()
-    parser.add_argument("--one_hot",type=bool,default=True,help="use one-hot encoding")
-    parser.add_argument("--lr", type=float, default=0.001, help="learnin rate")
+    parser.add_argument("--test_ratio", type=float, default=0.3, help="ratio of dividing the data set")
+    parser.add_argument("--one_hot", type=bool, default=True, help="use one-hot encoding")
+    parser.add_argument("--k_hop", type=int, default=2, help="hops of the generated subgraph")
+    parser.add_argument("--num_neighbor", type=int, default=5, help="neighbor sample number")
+    parser.add_argument("--batch_size", type=int, default=32, help="batch size")
+    parser.add_argument("--lr", type=float, default=0.001, help="learning rate")
     parser.add_argument("--n_epoch", type=int, default=100, help="number of epoch")
     parser.add_argument("--drop_rate", type=float, default=0.01, help="drop_rate")
     parser.add_argument('--dataset', type=str, default='acm', help='dataset')
-    parser.add_argument("--dataset_path", type = str, default = r"",help='dataset_path')
+    parser.add_argument("--dataset_path", type=str, default=r"", help='dataset_path')
 
     args = parser.parse_args()
     main(args)
\ No newline at end of file
diff --git a/gammagl/datasets/acm4dhn.py b/gammagl/datasets/acm4dhn.py
index b65d2010..fbfb849e 100644
--- a/gammagl/datasets/acm4dhn.py
+++ b/gammagl/datasets/acm4dhn.py
@@ -6,13 +6,13 @@
 
 class ACM4DHN(InMemoryDataset):
     url = 'https://raw.githubusercontent.com/BUPT-GAMMA/HDE/main/ds/imdb'
-    test_ratio = 0.3
 
     def __init__(self, root: Optional[str] = None, transform: Optional[Callable] = None,
                  pre_transform: Optional[Callable] = None, pre_filter: Optional[Callable] = None,
-                 force_reload: bool = False):
+                 force_reload: bool = False, test_ratio: float = 0.3):
         super().__init__(root, transform, pre_transform, force_reload=force_reload)
         self.data, self.slices = self.load_data(self.processed_paths[0])
+        self.test_ratio = test_ratio
 
     @property
     def raw_file_names(self) -> List[str]:
diff --git a/tests/datasets/test_acm4dhn.py b/tests/datasets/test_acm4dhn.py
new file mode 100644
index 00000000..13102d3f
--- /dev/null
+++ b/tests/datasets/test_acm4dhn.py
@@ -0,0 +1,15 @@
+import tensorlayerx as tlx
+from gammagl.datasets import ACM4DHN
+
+root = "./data"
+
+
+def test_acm4dhn():
+    test_ratio = 0.3
+    dataset = ACM4DHN(root=root, test_ratio=test_ratio)
+    graph = dataset[0]
+    assert len(dataset) == 1
+    assert tlx.get_tensor_shape(graph['M', 'MA', 'A'].edge_index) == [2, 7071]
+    assert tlx.get_tensor_shape(graph['train']['M', 'MA', 'A'].edge_index) == [2, 2828]
+    assert tlx.get_tensor_shape(graph['val']['M', 'MA', 'A'].edge_index) == [2, 2121]
+    assert tlx.get_tensor_shape(graph['test']['M', 'MA', 'A'].edge_index) == [2, 2122]

From 986f94562e125cce9a06c61465caf66661113ef5 Mon Sep 17 00:00:00 2001
From: gyzhou2000 <gyzhou2000@gmail.com>
Date: Tue, 9 Jul 2024 22:15:37 +0800
Subject: [PATCH 6/8] update

---
 examples/dhn/README.md      |  6 +--
 examples/dhn/dhn_trainer.py | 85 +++++++++++++++++++++----------------
 gammagl/datasets/acm4dhn.py |  6 +--
 gammagl/models/dhn.py       |  5 ++-
 gammagl/utils/subgraph.py   |  4 +-
 5 files changed, 60 insertions(+), 46 deletions(-)

diff --git a/examples/dhn/README.md b/examples/dhn/README.md
index a365ebeb..a10d95ba 100644
--- a/examples/dhn/README.md
+++ b/examples/dhn/README.md
@@ -3,9 +3,9 @@
 - Author's code repo: [https://github.com/BUPT-GAMMA/HDE](https://github.com/BUPT-GAMMA/HDE)
 
 ## Dataset Statics
-| Dataset  | # Nodes | # Edges | # Classes |
-|----------|---------|---------|-----------|
-| Cora     | 4435    | 7071    | 2         |
+| Dataset  | # Nodes | # Edges |
+|----------|---------|---------|
+| acm      | 3908    | 4500    |
 
 ## Results
 ```bash
diff --git a/examples/dhn/dhn_trainer.py b/examples/dhn/dhn_trainer.py
index 13b39689..2c529143 100644
--- a/examples/dhn/dhn_trainer.py
+++ b/examples/dhn/dhn_trainer.py
@@ -1,16 +1,19 @@
+import os
+# os.environ['CUDA_VISIBLE_DEVICES'] = '0'
+# os.environ['TL_BACKEND'] = 'torch'
+os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2' 
+# 0:Output all; 1:Filter out INFO; 2:Filter out INFO and WARNING; 3:Filter out INFO, WARNING, and ERROR
+
 import argparse
 import random
 import numpy as np
 import tensorlayerx as tlx
 from tensorlayerx.model import TrainOneStep
 from sklearn.metrics import roc_auc_score
-
 from gammagl.models import DHNModel
 from gammagl.datasets import ACM4DHN
 from gammagl.utils import k_hop_subgraph, find_all_simple_paths
 
-# The UserWarning is ignored globally
-import warnings
 
 type2idx = {
     'M': 0,
@@ -19,10 +22,6 @@
     # 'T': 3
 }
 
-warnings.filterwarnings("ignore", category=UserWarning)
-
-tlx.set_seed(10)
-
 
 def dist_encoder(src, dest, G, k_hop):
     if (G.size(1) == 0):
@@ -86,7 +85,7 @@ def gen_fea_batch(G, root, fea_dict, k_hop):
             neighbors_mask = src == node
         else:
             neighbors_mask = dst == node
-        neighbors = list(dst[neighbors_mask].numpy())
+        neighbors = list(tlx.convert_to_numpy(dst[neighbors_mask]))
         neighbors.append(node.item())
         random_choice_list = np.random.choice(neighbors, args.num_neighbor, replace=True)
         ns_1.append(random_choice_list.tolist())
@@ -109,7 +108,7 @@ def gen_fea_batch(G, root, fea_dict, k_hop):
             neighbors_mask = src == node
         else:
             neighbors_mask = dst == node
-        neighbors = list(dst[neighbors_mask].numpy())
+        neighbors = list(tlx.convert_to_numpy(dst[neighbors_mask]))
         neighbors.append(node.item())
         random_choice_list = np.random.choice(neighbors, args.num_neighbor, replace=True)
         ns_2.append(random_choice_list.tolist())
@@ -141,17 +140,17 @@ def subgraph_sampling_with_DE_node_pair(G, node_pair, k_hop=2):
 
     # Remove edges using Boolean indexes
     # Note: Just remove the edges, the points remain
-    edge_index_np = sub_G_for_AB[1].numpy()
+    edge_index_np = tlx.convert_to_numpy(sub_G_for_AB[1])
     remove_indices = tlx.convert_to_tensor([
         ((edge_index_np[0, i] == A) & (edge_index_np[1, i] == B)) | (
                 (edge_index_np[0, i] == B) & (edge_index_np[1, i] == A))
         for i in range(sub_G_for_AB[1].shape[1])
     ])
-    remove_indices = remove_indices.numpy()
+    remove_indices = tlx.convert_to_numpy(remove_indices)
     sub_G_index = sub_G_for_AB[1][:, ~remove_indices]
 
-    sub_G_nodes = set(np.unique(sub_G_for_AB[0].numpy())) | set(
-        np.unique(sub_G_for_AB[1].numpy()))  # Gets the points in the graph
+    sub_G_nodes = set(np.unique(tlx.convert_to_numpy(sub_G_for_AB[0]))) | set(
+        np.unique(tlx.convert_to_numpy(sub_G_for_AB[1])))  # Gets the points in the graph
     sub_G_nodes = tlx.convert_to_tensor(list(sub_G_nodes))
 
     # Distance from all points in the subgraph to the node pair
@@ -172,8 +171,8 @@ def subgraph_sampling_with_DE_node_pair(G, node_pair, k_hop=2):
 
 def batch_data(G, batch_size):
     edge_index = G['M', 'MA', 'A'].edge_index
-    nodes = set(tlx.convert_to_tensor(np.unique(edge_index[0].numpy()))) | set(
-        tlx.convert_to_tensor(np.unique(edge_index[1].numpy())))
+    nodes = set(tlx.convert_to_tensor(np.unique(tlx.convert_to_numpy(edge_index[0])))) | set(
+        tlx.convert_to_tensor(np.unique(tlx.convert_to_numpy(edge_index[1]))))
 
     nodes_list = []
     for node in nodes:
@@ -182,7 +181,7 @@ def batch_data(G, batch_size):
     num_batch = int(len(edge_index[0]) / batch_size)
 
     # Shuffle the order of the edges
-    edge_index_np = np.array(edge_index)
+    edge_index_np = tlx.convert_to_numpy(edge_index)
     permutation = np.random.permutation(edge_index_np.shape[1])  # Generate a randomly arranged index
     edge_index_np = edge_index_np[:, permutation]  # Use this permutation index to scramble edge_index
     edge_index = tlx.convert_to_tensor(edge_index_np)
@@ -259,25 +258,25 @@ def main(args):
     node_type = len(type2idx)
     num_fea = (args.k_hop + 2) * 4 + node_type
 
-    m = DHNModel(num_fea, args.batch_size, args.num_neighbor)
+    model = DHNModel(num_fea, args.batch_size, args.num_neighbor, name="DHN")
 
     optim = tlx.optimizers.Adam(lr=args.lr, weight_decay=args.drop_rate)
-    train_weights = m.trainable_weights
+    train_weights = model.trainable_weights
 
-    net_with_loss = Loss(m, loss_fn=tlx.losses.sigmoid_cross_entropy)
+    net_with_loss = Loss(model, loss_fn=tlx.losses.sigmoid_cross_entropy)
     net_with_train = TrainOneStep(net_with_loss, optim, train_weights)
 
     tra_auc_metric = AUCMetric()
     val_auc_metric = AUCMetric()
     test_auc_metric = AUCMetric()
 
+    best_val_auc = 0
     for epoch in range(args.n_epoch):
-        print("-----Epoch {}/{}-----".format(epoch + 1, args.n_epoch))
 
         # train
-        m.set_train()
+        model.set_train()
         tra_batch_A_fea, tra_batch_B_fea, tra_batch_y = batch_data(G_train, args.batch_size).__next__()
-        tra_out = m(tra_batch_A_fea, tra_batch_B_fea, tra_batch_y)
+        tra_out = model(tra_batch_A_fea, tra_batch_B_fea, tra_batch_y)
 
         data = {
             "n1": tra_batch_A_fea,
@@ -286,28 +285,33 @@ def main(args):
         }
 
         tra_loss = net_with_train(data, tra_batch_y)
-        tra_auc_metric.update_state(y_true=tra_batch_y, y_pred=tlx.sigmoid(tra_out).detach().numpy())
+        tra_auc_metric.update_state(y_true=tra_batch_y, y_pred=tlx.convert_to_numpy(tlx.sigmoid(tra_out)))
         tra_auc = tra_auc_metric.result()
-        print('train: ', tra_loss, tra_auc)
 
         # val
-        m.set_eval()
+        model.set_eval()
         val_batch_A_fea, val_batch_B_fea, val_batch_y = batch_data(G_val, args.batch_size).__next__()
-        val_out = m(val_batch_A_fea, val_batch_B_fea, val_batch_y)
+        val_out = model(val_batch_A_fea, val_batch_B_fea, val_batch_y)
 
-        val_loss = tlx.losses.sigmoid_cross_entropy(output=val_out, target=tlx.convert_to_tensor(val_batch_y))
-        val_auc_metric.update_state(y_true=val_batch_y, y_pred=tlx.sigmoid(val_out).detach().numpy())
+        val_auc_metric.update_state(y_true=val_batch_y, y_pred=tlx.convert_to_numpy(tlx.sigmoid(val_out)))
         val_auc = val_auc_metric.result()
-        print("val: ", val_loss.item(), val_auc)
 
-        # test
-        test_batch_A_fea, test_batch_B_fea, test_batch_y = batch_data(G_test, args.batch_size).__next__()
-        test_out = m(test_batch_A_fea, test_batch_B_fea, test_batch_y)
+        print("Epoch [{:0>3d}] ".format(epoch+1)\
+              + "  train loss: {:.4f}".format(tra_loss.item())\
+              + "  val auc: {:.4f}".format(val_auc))
 
-        test_loss = tlx.losses.sigmoid_cross_entropy(output=test_out, target=tlx.convert_to_tensor(test_batch_y))
-        test_auc_metric.update_state(y_true=test_batch_y, y_pred=tlx.sigmoid(test_out).detach().numpy())
-        test_auc = test_auc_metric.result()
-        print("test: ", test_loss.item(), test_auc)
+        if val_auc > best_val_auc:
+            best_val_auc = val_auc
+            model.save_weights(args.best_model_path+model.name+".npz", format='npz_dict')
+
+    model.load_weights(args.best_model_path+model.name+".npz", format='npz_dict')
+    # test
+    test_batch_A_fea, test_batch_B_fea, test_batch_y = batch_data(G_test, args.batch_size).__next__()
+    test_out = model(test_batch_A_fea, test_batch_B_fea, test_batch_y)
+
+    test_auc_metric.update_state(y_true=test_batch_y, y_pred=tlx.convert_to_numpy(tlx.sigmoid(test_out)))
+    test_auc = test_auc_metric.result()
+    print("Test auc:  {:.4f}".format(test_auc))
 
 
 if __name__ == '__main__':
@@ -323,6 +327,13 @@ def main(args):
     parser.add_argument("--drop_rate", type=float, default=0.01, help="drop_rate")
     parser.add_argument('--dataset', type=str, default='acm', help='dataset')
     parser.add_argument("--dataset_path", type=str, default=r"", help='dataset_path')
+    parser.add_argument("--best_model_path", type=str, default=r'./', help="path to save best model")
+    parser.add_argument("--gpu", type=int, default=-1)
 
     args = parser.parse_args()
-    main(args)
\ No newline at end of file
+    if args.gpu >= 0:
+        tlx.set_device("GPU", args.gpu)
+    else:
+        tlx.set_device("CPU")
+
+    main(args)
diff --git a/gammagl/datasets/acm4dhn.py b/gammagl/datasets/acm4dhn.py
index fbfb849e..2305e798 100644
--- a/gammagl/datasets/acm4dhn.py
+++ b/gammagl/datasets/acm4dhn.py
@@ -10,9 +10,9 @@ class ACM4DHN(InMemoryDataset):
     def __init__(self, root: Optional[str] = None, transform: Optional[Callable] = None,
                  pre_transform: Optional[Callable] = None, pre_filter: Optional[Callable] = None,
                  force_reload: bool = False, test_ratio: float = 0.3):
+        self.test_ratio = test_ratio
         super().__init__(root, transform, pre_transform, force_reload=force_reload)
         self.data, self.slices = self.load_data(self.processed_paths[0])
-        self.test_ratio = test_ratio
 
     @property
     def raw_file_names(self) -> List[str]:
@@ -71,8 +71,8 @@ def process(self):
             [edge_index_M[sp1:sp1 + sp2], edge_index_A[sp1:sp1 + sp2]])
         G_test['M', 'MA', 'A'].edge_index = tlx.convert_to_tensor([edge_index_M[sp1 + sp2:], edge_index_A[sp1 + sp2:]])
 
-        print(
-            f"all edge: {len(G['M', 'MA', 'A'].edge_index[0])}, train edge: {len(G_train['M', 'MA', 'A'].edge_index[0])}, val edge: {len(G_val['M', 'MA', 'A'].edge_index[0])}, test edge: {len(G_test['M', 'MA', 'A'].edge_index[0])}")
+        # print(
+        #     f"all edge: {len(G['M', 'MA', 'A'].edge_index[0])}, train edge: {len(G_train['M', 'MA', 'A'].edge_index[0])}, val edge: {len(G_val['M', 'MA', 'A'].edge_index[0])}, test edge: {len(G_test['M', 'MA', 'A'].edge_index[0])}")
 
         G['train'] = G_train  # training set
         G['val'] = G_val  # valuation set
diff --git a/gammagl/models/dhn.py b/gammagl/models/dhn.py
index 2b717a17..615960b3 100644
--- a/gammagl/models/dhn.py
+++ b/gammagl/models/dhn.py
@@ -6,8 +6,9 @@ class DHNModel(tlx.nn.Module):
     def __init__(self,
                  num_fea,
                  batch_size,
-                 num_neighbor):
-        super().__init__()
+                 num_neighbor,
+                 name=None):
+        super().__init__(name=name)
         self.num_fea = num_fea
         self.batch_size = batch_size
         self.num_neighbor = num_neighbor
diff --git a/gammagl/utils/subgraph.py b/gammagl/utils/subgraph.py
index d2217ea7..1f9f3a40 100644
--- a/gammagl/utils/subgraph.py
+++ b/gammagl/utils/subgraph.py
@@ -42,7 +42,9 @@ def k_hop_subgraph(node_idx, num_hops, edge_index, relabel_nodes=False, num_node
         col, row = edge_index
 
     if isinstance(node_idx, (int, list, tuple)):
-        node_idx = tlx.convert_to_tensor(np.array([node_idx]).flatten(), dtype=edge_index.dtype)
+        # node_idx = tlx.convert_to_tensor(np.array([node_idx]).flatten(), dtype=edge_index.dtype)
+        node_idx = np.array([i.item() for i in node_idx]).flatten()
+        node_idx = tlx.convert_to_tensor(node_idx, dtype=tlx.int64)
     
     subsets = [node_idx]
     node_mask = tlx.zeros((num_nodes,), dtype=tlx.bool)

From 75194150426aa4ba9cced0f38c7b3dfa01c11384 Mon Sep 17 00:00:00 2001
From: gyzhou2000 <gyzhou2000@gmail.com>
Date: Sun, 14 Jul 2024 22:16:34 +0800
Subject: [PATCH 7/8] update

---
 tests/utils/test_find_all_simple_paths.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/tests/utils/test_find_all_simple_paths.py b/tests/utils/test_find_all_simple_paths.py
index d39b872a..7d2fff71 100644
--- a/tests/utils/test_find_all_simple_paths.py
+++ b/tests/utils/test_find_all_simple_paths.py
@@ -1,7 +1,10 @@
 import tensorlayerx as tlx
 from gammagl.utils.simple_path import find_all_simple_paths
 
+
 def test_find_all_simple_paths():
+    if tlx.BACKEND == 'tensorflow':
+        return
     edge_index = tlx.convert_to_tensor([[0, 0, 1, 1, 2], [1, 2, 0, 3, 1]], dtype=tlx.int64)
     src = tlx.convert_to_tensor([0], dtype=tlx.int64)
     dest = tlx.convert_to_tensor([3], dtype=tlx.int64)
@@ -20,4 +23,4 @@ def test_find_all_simple_paths():
     dest = tlx.convert_to_tensor([0], dtype=tlx.int64)
     expected_paths = [[0]]
     paths = find_all_simple_paths(edge_index, src, dest, max_length=1)
-    assert sorted(map(tuple, paths))==sorted(map(tuple, expected_paths))
\ No newline at end of file
+    assert sorted(map(tuple, paths))==sorted(map(tuple, expected_paths))

From 4dbba4960de0be86b967b6f1c06340bfcef5b14f Mon Sep 17 00:00:00 2001
From: gyzhou2000 <gyzhou@bupt.edu.cn>
Date: Sun, 14 Jul 2024 15:07:20 +0000
Subject: [PATCH 8/8] update

---
 gammagl/utils/subgraph.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/gammagl/utils/subgraph.py b/gammagl/utils/subgraph.py
index 1f9f3a40..29f10532 100644
--- a/gammagl/utils/subgraph.py
+++ b/gammagl/utils/subgraph.py
@@ -42,9 +42,9 @@ def k_hop_subgraph(node_idx, num_hops, edge_index, relabel_nodes=False, num_node
         col, row = edge_index
 
     if isinstance(node_idx, (int, list, tuple)):
-        # node_idx = tlx.convert_to_tensor(np.array([node_idx]).flatten(), dtype=edge_index.dtype)
-        node_idx = np.array([i.item() for i in node_idx]).flatten()
-        node_idx = tlx.convert_to_tensor(node_idx, dtype=tlx.int64)
+        node_idx = tlx.convert_to_tensor(np.array([node_idx]).flatten(), dtype=edge_index.dtype)
+        # node_idx = np.array([i.item() for i in node_idx]).flatten()
+        # node_idx = tlx.convert_to_tensor(node_idx, dtype=tlx.int64)
     
     subsets = [node_idx]
     node_mask = tlx.zeros((num_nodes,), dtype=tlx.bool)