DotaNP
Daniel Craig
2023-05-14
library(ggiraphExtra)
library(knitr)
library(data.table)
library(readr)
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:data.table':
##
## between, first, last## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(randomForest)## randomForest 4.7-1.1## Type rfNews() to see new features/changes/bug fixes.##
## Attaching package: 'randomForest'## The following object is masked from 'package:dplyr':
##
## combinelibrary(caret)## Loading required package: ggplot2##
## Attaching package: 'ggplot2'## The following object is masked from 'package:randomForest':
##
## margin## Loading required package: latticelibrary(ggplot2)
library(gbm)## Loaded gbm 2.1.8.1library(ggsci)
library(tree)
library(mlr3) #https://mlr3book.mlr-org.com/basics.html#learners
library(mlr3viz)
library(mlr3learners)
library(kknn)##
## Attaching package: 'kknn'## The following object is masked from 'package:caret':
##
## contr.dummyLoad In Data
# data<-read_csv("_Projects\\Project3\\datasets\\NPBase.csv")
#
# npData<-read_csv("NPBase.csv")
# npData<- as_tibble(npData)
# npData <- as.data.frame(npData)
# npData$win <- as.character(npData$win)
# #recode(npData$win, Yes = "1", No = "0") did not use
# #recode(npData$win, "Yes"="TRUE", "No" = "FALSE") did not use
#
# npData
#
# npData$win <- npData$win %>% mutate(win = ifelse(win == "No",0,1))
#
#
# npData <- data.frame(match_id = c(1234,5678,1245),
# start_time = c(1657339909,1657357190,1657366555),
# win = c('false','false','true'),
# hero_id = '53',
# account_id = c(2345438,259803,438689),
# leaguename = c('Titus','Destiny','Ultras'),
# gold_per_min = c(569,549,654),
# net_worth = c(15770,15317,16985),
# gold = c(1140,1559,3210),
# kills = c(4,12,7),
# tower_damage = c(1245,19823,4599),
# duration = c(1754,1759,1829),
# lane = c(1,3,3),
# lane_role = c(1, 3,3)
# )
npData<-read_csv("NPBase.csv")
npData$gold <- as.numeric(npData$gold)
dataModel <- data.frame(npData)
dataModel$win <- as.factor(dataModel$win)
dataModel$lane <- as.factor(dataModel$lane)
dataModel$lane_role <- as.factor(dataModel$lane_role)
dataModel## match_id start_time win hero_id account_id
## 1 6653637943 1657339909 FALSE 53 250544263
## 2 6653911377 1657357190 FALSE 53 76082922
## 3 6654090150 1657366555 TRUE 53 1312740399
## 4 6656721191 1657512120 FALSE 53 209165399
## 5 6657130147 1657540852 TRUE 53 132851371
## 6 6657972166 1657586272 TRUE 53 59752811
## 7 6658023411 1657591653 FALSE 53 85367644
## 8 6658141348 1657602904 FALSE 53 1312740399
## 9 6658904045 1657643886 TRUE 53 155370747
## 10 6659582275 1657695291 TRUE 53 412910858
## 11 6660871188 1657775009 TRUE 53 162539779
## 12 6660953769 1657782385 TRUE 53 176158708
## 13 6662754663 1657890925 TRUE 53 216284105
## 14 6663760635 1657948184 FALSE 53 254710104
## 15 6663831927 1657952923 TRUE 53 254710104
## 16 6663890419 1657956692 FALSE 53 128367019
## 17 6664002830 1657962564 TRUE 53 176158708
## 18 6664762074 1657994640 FALSE 53 191362875
## 19 6664870055 1657999696 TRUE 53 1261414529
## 20 6665353816 1658035345 FALSE 53 374150060
## 21 6665371502 1658036503 TRUE 53 247018516
## 22 6665457039 1658041546 TRUE 53 153727297
## 23 6665530799 1658045885 FALSE 53 923473386
## 24 6665530990 1658045359 TRUE 53 899495504
## 25 6665634983 1658050792 FALSE 53 330534326
## 26 6666225827 1658073951 TRUE 53 216284105
## 27 6666901923 1658120586 TRUE 53 254710104
## 28 6667145190 1658138752 TRUE 53 145993649
## 29 6667146750 1658138846 TRUE 53 153727297
## 30 6667177868 1658140739 FALSE 53 1170923497
## 31 6667234541 1658143841 TRUE 53 1173255449
## 32 6667305314 1658147179 TRUE 53 1173255449
## 33 6667411276 1658151435 TRUE 53 153727297
## 34 6668150103 1658193432 TRUE 53 202464276
## 35 6668150769 1658193513 FALSE 53 368668788
## 36 6668154335 1658193855 TRUE 53 249651648
## 37 6668214992 1658199924 TRUE 53 254710104
## 38 6668488644 1658223069 TRUE 53 886742476
## 39 6669118010 1658252002 FALSE 53 851178649
## 40 6669394757 1658268012 FALSE 53 116080282
## 41 6669455743 1658273257 FALSE 53 173971950
## 42 6669611769 1658289953 FALSE 53 202984325
## 43 6669704190 1658298434 FALSE 53 236322966
## 44 6669833934 1658308212 TRUE 53 360648679
## 45 6669894503 1658312143 FALSE 53 327800297
## 46 6669951485 1658315451 FALSE 53 218367241
## 47 6669958120 1658315961 FALSE 53 1045494300
## 48 6669999318 1658318413 TRUE 53 34505203
## 49 6670287920 1658331155 FALSE 53 177411785
## 50 6670482012 1658340005 TRUE 53 34505203
## 51 6670489387 1658340043 FALSE 53 206379758
## 52 6670781535 1658358674 TRUE 53 125455547
## 53 6671029299 1658383899 FALSE 53 399465090
## 54 6671588490 1658415604 TRUE 53 177411785
## 55 6671680767 1658419338 TRUE 53 177411785
## 56 6671827493 1658425735 TRUE 53 439345730
## 57 6671860500 1658427423 TRUE 53 106863163
## 58 6671949473 1658432182 TRUE 53 34505203
## 59 6672334898 1658466306 TRUE 53 254710104
## 60 6672361043 1658468671 FALSE 53 110142194
## 61 6672429701 1658474342 FALSE 53 104504914
## 62 6672487457 1658478590 TRUE 53 458287006
## 63 6672643482 1658488211 FALSE 53 458287006
## 64 6672675633 1658491198 FALSE 53 103910993
## 65 6672954942 1658502006 TRUE 53 311360822
## 66 6673081589 1658506714 FALSE 53 97658618
## 67 6673720237 1658549075 FALSE 53 171337923
## 68 6673734891 1658550038 FALSE 53 171337923
## 69 6673880837 1658559912 TRUE 53 1049956230
## 70 6673883278 1658559988 FALSE 53 101240880
## 71 6673952558 1658564159 FALSE 53 177061558
## 72 6674003905 1658566608 TRUE 53 254710104
## 73 6674108208 1658572065 FALSE 53 146372433
## 74 6674151935 1658574386 FALSE 53 94439767
## 75 6674445805 1658586499 FALSE 53 34505203
## 76 6674752354 1658598769 TRUE 53 1352080796
## 77 6674755794 1658599100 FALSE 53 94738847
## 78 6674843794 1658603146 TRUE 53 172099728
## 79 6674859295 1658604114 TRUE 53 94738847
## 80 6675282069 1658635714 FALSE 53 250544263
## 81 6675316855 1658638257 FALSE 53 209165399
## 82 6675382529 1658643230 TRUE 53 110142194
## 83 6675511055 1658649423 FALSE 53 247018516
## 84 6675654131 1658656280 TRUE 53 458287006
## 85 6675690081 1658658166 TRUE 53 101240880
## 86 6675726202 1658660380 TRUE 53 118134220
## 87 6676013166 1658671587 FALSE 53 130801103
## 88 6676023312 1658671904 FALSE 53 224981103
## 89 6676306088 1658683321 TRUE 53 172099728
## 90 6676358917 1658685847 TRUE 53 1414344010
## 91 6676488286 1658692615 TRUE 53 118134220
## 92 6676495794 1658693040 TRUE 53 202464276
## 93 6676555616 1658696390 TRUE 53 131303632
## 94 6676612110 1658700154 FALSE 53 41637292
## 95 6676724543 1658710926 FALSE 53 206379758
## 96 6676767202 1658715326 FALSE 53 96189126
## 97 6676953324 1658732736 TRUE 53 375507918
## 98 6677121365 1658743996 FALSE 53 1201927095
## 99 6678322841 1658815578 TRUE 53 1045494300
## 100 6678463629 1658826027 TRUE 53 156328257
## 101 6678511020 1658828928 TRUE 53 1102532597
## 102 6679342655 1658866858 TRUE 53 339235645
## 103 6679573632 1658887212 TRUE 53 339235645
## 104 6679657576 1658894888 TRUE 53 181339278
## 105 6679802770 1658907260 TRUE 53 130271765
## 106 6679903505 1658913978 TRUE 53 101240880
## 107 6680017014 1658920526 FALSE 53 285770518
## 108 6680145522 1658926655 TRUE 53 27178898
## 109 6680665928 1658948874 TRUE 53 41637292
## 110 6680878849 1658962972 FALSE 53 191362875
## 111 6680993748 1658975129 FALSE 53 94004717
## 112 6681062357 1658981652 FALSE 53 198631761
## 113 6681156803 1658990009 FALSE 53 164948665
## 114 6681230217 1658995283 FALSE 53 250547284
## 115 6681277222 1658998409 FALSE 53 177061558
## 116 6681396813 1659005681 FALSE 53 1125296112
## 117 6681478657 1659009834 FALSE 53 250547284
## 118 6681513765 1659011683 TRUE 53 152455523
## 119 6681867862 1659025385 TRUE 53 85367644
## 120 6681976191 1659030407 TRUE 53 98482878
## 121 6682066040 1659035217 TRUE 53 41637292
## 122 6682250660 1659044952 TRUE 53 126212866
## 123 6682277285 1659046216 TRUE 53 202464276
## 124 6682300306 1659047994 TRUE 53 191362875
## 125 6682481806 1659066192 TRUE 53 108452107
## 126 6682492162 1659067359 FALSE 53 353424606
## 127 6682513240 1659069183 FALSE 53 191362875
## 128 6682553675 1659072704 TRUE 53 1263892079
## 129 6682582314 1659075095 TRUE 53 177061558
## 130 6682601132 1659076596 FALSE 53 254710104
## 131 6682735589 1659085662 FALSE 53 345509021
## 132 6682792018 1659089174 TRUE 53 853928083
## 133 6682841003 1659092421 TRUE 53 399885807
## 134 6682857406 1659092750 TRUE 53 177061558
## 135 6682944177 1659096969 FALSE 53 287966898
## 136 6682960303 1659097801 FALSE 53 177411785
## 137 6683081104 1659102476 TRUE 53 177411785
## 138 6683177335 1659106632 FALSE 53 97590558
## 139 6683374759 1659114015 FALSE 53 177411785
## 140 6683769334 1659134442 TRUE 53 126212866
## 141 6683806784 1659138256 TRUE 53 126212866
## 142 6683976570 1659152412 TRUE 53 392277500
## 143 6684012176 1659155203 FALSE 53 86822085
## 144 6684155347 1659164477 TRUE 53 99573176
## 145 6684418241 1659178291 TRUE 53 162641196
## 146 6685073736 1659204694 TRUE 53 194521913
## 147 6685285522 1659215590 TRUE 53 191290281
## 148 6685452046 1659228631 FALSE 53 97658618
## 149 6685493723 1659232558 FALSE 53 126212866
## 150 6685545862 1659236652 FALSE 53 108452107
## 151 6685585153 1659239647 TRUE 53 247018516
## 152 6685749437 1659250478 TRUE 53 171337923
## 153 6685766100 1659251102 FALSE 53 176616331
## 154 6685893938 1659257496 FALSE 53 171337923
## 155 6686015749 1659263220 TRUE 53 458287006
## 156 6686253875 1659272730 FALSE 53 1247302157
## 157 6687078267 1659310579 FALSE 53 194521913
## 158 6687280868 1659330212 TRUE 53 162641196
## 159 6687357426 1659336758 FALSE 53 101240880
## 160 6687371700 1659337808 TRUE 53 113874152
## 161 6687466269 1659344230 TRUE 53 104504914
## 162 6687780002 1659359538 FALSE 53 355168766
## 163 6687956200 1659366013 TRUE 53 97590558
## 164 6688230252 1659376799 TRUE 53 177411785
## 165 6688311075 1659380419 TRUE 53 177411785
## 166 6688750995 1659414127 FALSE 53 209165399
## 167 6688926370 1659428275 TRUE 53 236322966
## 168 6689092149 1659438187 FALSE 53 104504914
## 169 6689161876 1659441607 FALSE 53 97590558
## 170 6689257142 1659445538 TRUE 53 171097887
## 171 6689476142 1659453490 TRUE 53 171097887
## 172 6690270489 1659506663 FALSE 53 162126721
## 173 6690566627 1659525943 TRUE 53 359068880
## 174 6690777541 1659535109 TRUE 53 177411785
## 175 6690857847 1659537994 TRUE 53 1050127319
## 176 6690905555 1659539714 FALSE 53 1050127319
## 177 6691619784 1659588724 FALSE 53 1200963940
## 178 6691624668 1659589154 TRUE 53 104504914
## 179 6691811669 1659603583 FALSE 53 1247011173
## 180 6691866371 1659607073 TRUE 53 219950601
## 181 6691953594 1659612048 FALSE 53 181339278
## 182 6692323200 1659627816 TRUE 53 387624608
## 183 6692728978 1659650902 TRUE 53 96189126
## 184 6692938319 1659672202 FALSE 53 1133046148
## 185 6692942529 1659672614 TRUE 53 1232094007
## 186 6692978872 1659675650 FALSE 53 460481806
## 187 6693057839 1659682484 FALSE 53 1119908658
## 188 6693722540 1659715425 TRUE 53 146744261
## 189 6693777829 1659717891 FALSE 53 237527926
## 190 6693934550 1659725523 FALSE 53 237527926
## 191 6694523000 1659769569 TRUE 53 110142194
## 192 6694583595 1659773285 TRUE 53 1068676453
## 193 6694689276 1659778583 FALSE 53 1220632105
## 194 6694863732 1659787175 TRUE 53 345509021
## 195 6696003659 1659850638 FALSE 53 138885864
## 196 6696012766 1659851203 TRUE 53 138885864
## 197 6696170524 1659859859 TRUE 53 27178898
## 198 6696706878 1659882037 TRUE 53 1251319262
## leaguename gold_per_min net_worth
## 1 Titus Colosseum Cup 569 15770
## 2 Destiny league 549 15317
## 3 Ultras Dota Pro League 654 16985
## 4 Titus Colosseum Cup 472 10399
## 5 Dota 2 Champions League 2021-2022 Season 894 40098
## 6 Thunderpick Bitcoin Series 593 21647
## 7 Thunderpick Bitcoin Series 442 11580
## 8 Ultras Dota Pro League 434 9842
## 9 Dota 2 Champions League 2021-2022 Season 701 26720
## 10 Destiny league 752 24276
## 11 Asia Battle Arena 810 22700
## 12 Destiny league 483 15927
## 13 Meister Provider Dota 2 Indonesia 2022 676 10746
## 14 Asia Battle Arena 902 41321
## 15 Asia Battle Arena 721 18818
## 16 TopClans2022 Summer Invitional 512 14369
## 17 Destiny league 412 12732
## 18 Thunderpick Bitcoin Series 848 34584
## 19 Thunderpick Bitcoin Series 559 15112
## 20 Khaz Modan Cup Series Season 3 290 6108
## 21 Asia Battle Arena 485 9739
## 22 Asia Battle Arena 655 13060
## 23 Khaz Modan Cup Series Season 3 453 11319
## 24 Titus Colosseum Cup 689 26345
## 25 TopClans2022 Summer Invitional 646 22515
## 26 Meister Provider Dota 2 Indonesia 2022 560 10202
## 27 Titus Colosseum Cup 643 13318
## 28 ESL One Malaysia 2022 Qualifiers powered by Intel 677 21088
## 29 ESL One Malaysia 2022 Qualifiers powered by Intel 785 29306
## 30 Destiny league 588 16788
## 31 Destiny league 672 18876
## 32 Destiny league 613 16366
## 33 ESL One Malaysia 2022 Qualifiers powered by Intel 808 27482
## 34 ESL One Malaysia 2022 Qualifiers powered by Intel 785 33771
## 35 ESL One Malaysia 2022 Qualifiers powered by Intel 442 8841
## 36 ESL One Malaysia 2022 Qualifiers powered by Intel 723 26798
## 37 Titus Colosseum Cup 580 13682
## 38 Khaz Modan Cup Series Season 3 805 27183
## 39 Dota 2 Champions League 2021-2022 Season 626 19381
## 40 TodayPay Dota2 Invitational Season 1 480 12344
## 41 ESL One Malaysia 2022 Qualifiers powered by Intel 603 21299
## 42 Destiny league 533 16905
## 43 Titus Colosseum Cup 455 8771
## 44 Asian Tigers 3 683 22392
## 45 ESL One Malaysia 2022 Qualifiers powered by Intel 777 29024
## 46 ESL One Malaysia 2022 Qualifiers powered by Intel 625 23730
## 47 Destiny league 578 20852
## 48 Riyadh Masters by Gamers8 623 17109
## 49 Riyadh Masters by Gamers8 624 21470
## 50 Riyadh Masters by Gamers8 650 22308
## 51 TodayPay Dota2 Invitational Season 1 559 16363
## 52 ESL One Malaysia 2022 Qualifiers powered by Intel 820 17797
## 53 Khaz Modan Cup Series Season 3 680 18160
## 54 Riyadh Masters by Gamers8 699 18374
## 55 Riyadh Masters by Gamers8 605 14825
## 56 ESL One Malaysia 2022 Qualifiers powered by Intel 773 21653
## 57 Riyadh Masters by Gamers8 821 29931
## 58 Riyadh Masters by Gamers8 543 15297
## 59 Asia Battle Arena 630 16061
## 60 Asia Battle Arena 584 11481
## 61 Asian Tigers 3 714 27986
## 62 Asian Tigers 3 756 22428
## 63 Asian Tigers 3 487 10988
## 64 Riyadh Masters by Gamers8 441 11678
## 65 Riyadh Masters by Gamers8 856 39130
## 66 Riyadh Masters by Gamers8 560 13953
## 67 Asian Tigers 3 128 7808
## 68 Asian Tigers 3 442 13145
## 69 Destiny league 477 17071
## 70 Khaz Modan Cup Series Season 3 493 11153
## 71 Khaz Modan Cup Series Season 3 764 26141
## 72 Titus Colosseum Cup 725 21680
## 73 Asia Battle Arena 672 23782
## 74 European Pro League Season 2 539 17176
## 75 Riyadh Masters by Gamers8 533 15389
## 76 Ultras Dota Pro League 415 8209
## 77 Riyadh Masters by Gamers8 659 28956
## 78 ESL One Malaysia 2022 Qualifiers powered by Intel 748 22375
## 79 Riyadh Masters by Gamers8 617 19469
## 80 Asia Battle Arena 488 10869
## 81 Asia Battle Arena 554 13675
## 82 Asian Tigers 3 807 28348
## 83 Asia Battle Arena 552 13818
## 84 Asian Tigers 3 829 34622
## 85 Khaz Modan Cup Series Season 3 581 13529
## 86 Riyadh Masters by Gamers8 592 12822
## 87 European Pro League Season 2 592 22001
## 88 YouTube SIVVIT - Big Russian Show 577 18847
## 89 ESL One Malaysia 2022 Qualifiers powered by Intel 815 27546
## 90 Ultras Dota Pro League 696 21007
## 91 Riyadh Masters by Gamers8 639 19413
## 92 TodayPay Dota2 Invitational Season 1 948 32455
## 93 TodayPay Dota2 Invitational Season 1 729 27526
## 94 Thunderpick Bitcoin Series 629 23854
## 95 ESL One Malaysia 2022 Qualifiers powered by Intel 657 19874
## 96 ESL One Malaysia 2022 Qualifiers powered by Intel 573 19265
## 97 ESL One Malaysia 2022 Qualifiers powered by Intel 813 25736
## 98 Khaz Modan Cup Series Season 3 583 16306
## 99 Destiny league 644 21768
## 100 ESL One Malaysia 2022 Qualifiers powered by Intel 678 23821
## 101 Khaz Modan Cup Series Season 3 671 13065
## 102 ESL One Malaysia 2022 Qualifiers powered by Intel 818 21247
## 103 ESL One Malaysia 2022 Qualifiers powered by Intel 759 22575
## 104 Destiny league 802 25566
## 105 Victorious Cup 756 17329
## 106 Khaz Modan Cup Series Season 3 616 18119
## 107 Khaz Modan Cup Series Season 3 694 21558
## 108 Dota 2 Champions League 2021-2022 Season 624 17438
## 109 TodayPay Dota2 Invitational Season 1 641 18588
## 110 TodayPay Dota2 Invitational Season 1 632 21294
## 111 Thunderpick Bitcoin Series 832 25164
## 112 Victorious Cup 828 31286
## 113 ESL One Malaysia 2022 Qualifiers powered by Intel 938 37910
## 114 ESL One Malaysia 2022 Qualifiers powered by Intel 521 15167
## 115 Khaz Modan Cup Series Season 3 698 24601
## 116 Asian Tigers 3 651 21659
## 117 ESL One Malaysia 2022 Qualifiers powered by Intel 780 29327
## 118 ESL One Malaysia 2022 Qualifiers powered by Intel 701 25954
## 119 Thunderpick Bitcoin Series 842 32565
## 120 Соревнования по электронным играм «Е-динство» 558 8796
## 121 TodayPay Dota2 Invitational Season 1 495 6596
## 122 ESL One Malaysia 2022 Qualifiers powered by Intel 594 12938
## 123 TodayPay Dota2 Invitational Season 1 663 16437
## 124 ESL One Malaysia 2022 Qualifiers powered by Intel 714 32550
## 125 ESL One Malaysia 2022 Qualifiers powered by Intel 692 13663
## 126 Asian Tigers 3 777 30697
## 127 ESL One Malaysia 2022 Qualifiers powered by Intel 451 11790
## 128 Victorious Cup 915 23816
## 129 Khaz Modan Cup Series Season 3 668 11438
## 130 ESL One Malaysia 2022 Qualifiers powered by Intel 332 7461
## 131 European Pro League Season 2 712 23657
## 132 Victorious Cup 801 20299
## 133 ESL One Malaysia 2022 Qualifiers powered by Intel 660 28349
## 134 Asia Battle Arena 838 34085
## 135 Asia Battle Arena 833 38657
## 136 ESL One Malaysia 2022 Qualifiers powered by Intel 729 26248
## 137 ESL One Malaysia 2022 Qualifiers powered by Intel 856 37695
## 138 ESL One Malaysia 2022 Qualifiers powered by Intel 540 19767
## 139 ESL One Malaysia 2022 Qualifiers powered by Intel 610 20269
## 140 ESL One Malaysia 2022 Qualifiers powered by Intel 638 19220
## 141 ESL One Malaysia 2022 Qualifiers powered by Intel 621 18086
## 142 Titus Colosseum Cup Season 2 727 23482
## 143 ESL One Malaysia 2022 Qualifiers powered by Intel 887 33534
## 144 Titus Colosseum Cup Season 2 374 7934
## 145 Asian Tigers 3 764 31745
## 146 TodayPay Dota2 Invitational Season 1 631 18752
## 147 TodayPay Dota2 Invitational Season 1 689 20105
## 148 ESL One Malaysia 2022 Qualifiers powered by Intel 618 19472
## 149 ESL One Malaysia 2022 Qualifiers powered by Intel 542 17019
## 150 ESL One Malaysia 2022 Qualifiers powered by Intel 596 18836
## 151 Titus Colosseum Cup Season 2 672 14803
## 152 Asian Tigers 3 561 15597
## 153 Victorious Cup 871 42703
## 154 Asian Tigers 3 492 10583
## 155 Asian Tigers 3 586 11254
## 156 Ultras Dota Pro League 456 9801
## 157 TodayPay Dota2 Invitational Season 1 641 19453
## 158 Asian Tigers 3 648 15604
## 159 Khaz Modan Cup Series Season 3 655 26640
## 160 Asian Tigers 3 563 15107
## 161 Asian Tigers 3 592 23429
## 162 Asia Battle Arena 824 35087
## 163 Dota 2 Champions League 2022-2023 Season 645 18321
## 164 Dota 2 Champions League 2022-2023 Season 775 24101
## 165 Dota 2 Champions League 2022-2023 Season 932 41313
## 166 Titus Colosseum Cup Season 2 513 14353
## 167 Asia Battle Arena 751 28538
## 168 Khaz Modan Cup Series Season 3 475 15255
## 169 Dota 2 Champions League 2022-2023 Season 570 15176
## 170 Dota 2 Champions League 2022-2023 Season 598 13113
## 171 Dota 2 Champions League 2022-2023 Season 678 16985
## 172 MInd Cup 877 31353
## 173 MInd Cup 831 23908
## 174 Dota 2 Champions League 2022-2023 Season 924 41743
## 175 Ultras Dota Pro League 134 7790
## 176 Ultras Dota Pro League 624 19285
## 177 Destiny league 664 19179
## 178 Asian Tigers 3 525 18069
## 179 Destiny league 591 15327
## 180 MInd Cup 919 41418
## 181 Destiny league 640 18230
## 182 Summer wave 777 22294
## 183 TodayPay Dota2 Invitational Season 1 516 9095
## 184 Destiny league 120 6875
## 185 Destiny league 659 19467
## 186 Titus Colosseum Cup Season 2 432 8934
## 187 Destiny league 667 19039
## 188 Predator Energy League 706 18046
## 189 Summer wave 722 24314
## 190 Summer wave 876 41076
## 191 Asia Battle Arena 792 16988
## 192 Destiny league 518 13922
## 193 Destiny league 431 12950
## 194 European Pro League Season 2 838 30875
## 195 Titus Colosseum Cup Season 2 129 7910
## 196 Titus Colosseum Cup Season 2 567 11645
## 197 European Pro League Season 2 612 15363
## 198 Ultras Dota Pro League 760 27049
## gold kills tower_damage duration lane lane_role
## 1 1140 3 1221 1754 1 3
## 2 892 3 1559 1794 1 3
## 3 3210 10 19823 1547 3 3
## 4 449 4 1264 1319 1 1
## 5 9648 10 28469 2938 3 1
## 6 6622 5 10388 2435 3 1
## 7 2130 0 11727 1846 1 1
## 8 1192 0 6657 1315 3 3
## 9 4570 9 18541 2332 1 3
## 10 2576 13 9039 2263 3 3
## 11 2700 10 7885 1701 3 1
## 12 2377 1 18953 1957 3 3
## 13 1641 11 9049 934 3 1
## 14 3596 16 11169 3349 1 1
## 15 2243 22 4646 1533 3 1
## 16 1144 4 2475 1891 1 3
## 17 4407 0 8942 2002 3 3
## 18 5309 18 14206 2977 3 3
## 19 2687 4 16908 1879 1 3
## 20 958 2 0 1297 5 4
## 21 1364 4 645 1215 3 3
## 22 3310 9 16466 1182 1 1
## 23 269 1 446 1613 1 1
## 24 2020 10 23219 2473 3 3
## 25 1215 5 5258 2200 1 1
## 26 2447 3 2572 1040 3 1
## 27 1968 4 14306 1363 3 1
## 28 1338 8 8893 1938 3 3
## 29 1281 8 23614 2310 3 1
## 30 1063 5 3822 1838 3 1
## 31 651 11 9172 1672 1 1
## 32 2316 6 8085 1641 3 1
## 33 3307 6 14326 2083 3 1
## 34 1721 7 16460 2207 1 1
## 35 716 2 1474 1240 1 1
## 36 1273 9 15706 2278 1 1
## 37 1082 4 8754 1373 1 1
## 38 4008 14 13960 2142 3 1
## 39 1831 3 7552 1989 1 1
## 40 444 1 292 1706 1 1
## 41 3074 3 6085 2230 1 1
## 42 905 2 10574 2045 3 3
## 43 521 8 714 1352 3 1
## 44 3417 5 10921 2063 3 1
## 45 424 8 10553 2947 3 1
## 46 1605 4 855 2425 1 1
## 47 1277 8 12741 2516 2 2
## 48 3959 9 9364 1740 1 3
## 49 2970 2 14037 2417 1 3
## 50 3333 8 8823 2088 3 1
## 51 1238 3 2867 1851 1 1
## 52 1372 13 19135 1279 3 1
## 53 2310 6 5307 1686 3 1
## 54 3619 9 19038 1795 1 3
## 55 1320 3 22129 1464 1 3
## 56 1628 11 13320 1692 2 2
## 57 3881 11 16538 2190 2 2
## 58 2167 4 15882 1717 1 3
## 59 4486 8 5379 1614 1 1
## 60 1 4 1336 1527 1 1
## 61 581 4 16584 2752 1 3
## 62 1653 12 16618 1893 1 1
## 63 1183 1 2294 1420 1 1
## 64 578 2 3675 1735 1 1
## 65 5955 15 17614 3023 3 1
## 66 1428 2 6477 1710 1 1
## 67 7708 0 0 3312 2 2
## 68 365 1 5661 1890 3 3
## 69 2346 2 6867 2246 4 4
## 70 1278 3 5387 1436 3 3
## 71 1891 11 5526 2140 3 1
## 72 2830 11 10106 1801 1 1
## 73 957 7 13573 2647 3 1
## 74 976 2 2289 2203 3 1
## 75 389 3 8324 2066 1 3
## 76 2084 2 12088 1370 3 1
## 77 3881 8 8553 3038 3 3
## 78 3875 7 14898 1840 3 1
## 79 2039 6 16105 2172 3 3
## 80 569 6 2906 1610 1 3
## 81 1945 2 6836 1528 3 1
## 82 2148 7 8907 2252 3 1
## 83 1743 1 7569 1586 1 3
## 84 3297 12 22697 3626 3 1
## 85 2599 4 10267 1392 3 3
## 86 3747 8 9246 1278 3 3
## 87 2921 7 10944 2419 1 3
## 88 1647 5 4707 2108 1 1
## 89 4646 9 15996 2063 1 1
## 90 2907 1 13848 1882 3 1
## 91 2038 6 21849 1824 1 3
## 92 5705 20 18030 2054 1 1
## 93 3001 9 15313 2421 2 2
## 94 5229 4 15903 2505 3 3
## 95 349 6 2622 1909 3 1
## 96 1165 3 5936 2182 1 1
## 97 2436 13 19047 1962 1 1
## 98 456 5 3334 1997 1 1
## 99 4268 4 17180 2157 2 2
## 100 2671 7 12074 2156 3 3
## 101 1340 9 3208 1112 3 1
## 102 1747 17 21430 1651 2 2
## 103 1275 12 18738 1819 1 1
## 104 2741 8 10292 1963 3 1
## 105 2079 12 12953 1382 3 1
## 106 1369 7 11799 1818 1 3
## 107 1033 2 4663 2162 3 1
## 108 1438 9 9004 1760 1 3
## 109 1108 10 7872 1850 3 3
## 110 394 2 7872 2373 3 3
## 111 589 8 8530 2606 1 1
## 112 4386 11 16196 2404 1 3
## 113 1885 9 15057 3504 1 1
## 114 317 2 3777 1893 1 1
## 115 1126 8 6679 2563 3 1
## 116 929 11 5158 2305 3 1
## 117 1302 5 11033 2633 1 1
## 118 2204 9 13424 2271 1 1
## 119 8615 11 16681 2405 3 1
## 120 1596 8 1858 909 3 3
## 121 746 5 1492 800 3 3
## 122 713 12 6280 1336 3 1
## 123 1262 9 4386 1464 1 1
## 124 5445 2 25778 2869 1 3
## 125 1113 10 9585 1173 1 1
## 126 1722 5 8765 2484 3 1
## 127 440 2 3163 2032 1 1
## 128 641 20 14435 1596 3 1
## 129 988 5 3828 1029 3 1
## 130 361 2 582 1359 1 1
## 131 482 11 7827 2092 3 1
## 132 1474 8 18974 1534 1 1
## 133 899 6 27940 2647 2 2
## 134 5235 12 27195 2690 3 1
## 135 4957 8 13468 3477 3 1
## 136 1498 8 6838 2558 3 3
## 137 7820 3 15473 2722 1 3
## 138 1217 0 7462 2249 3 3
## 139 1589 2 5696 2365 3 1
## 140 1945 8 9261 1859 1 3
## 141 2086 4 20104 1782 3 3
## 142 457 7 15069 2264 3 1
## 143 2359 10 14573 2954 1 3
## 144 1284 2 3822 1569 3 1
## 145 945 16 8973 2646 1 3
## 146 3502 5 9250 1834 1 3
## 147 2080 2 15605 1780 3 1
## 148 1872 3 2206 1923 1 1
## 149 1519 2 5704 2054 3 3
## 150 261 8 2933 2251 1 1
## 151 2053 9 3602 1311 3 3
## 152 392 7 9377 1737 3 3
## 153 3328 12 15210 3667 2 2
## 154 2233 6 1619 1414 3 3
## 155 1179 3 8821 1134 2 2
## 156 271 4 757 1451 3 3
## 157 3 7 4454 2452 1 3
## 158 1329 12 5930 1491 3 1
## 159 160 8 9129 2696 3 3
## 160 3357 7 3297 1676 1 3
## 161 5054 5 11500 2578 1 3
## 162 1512 12 9796 3144 1 1
## 163 3096 7 14920 1701 3 3
## 164 1576 4 14679 1896 3 1
## 165 6813 20 24147 2753 1 1
## 166 828 3 1670 1906 3 1
## 167 5113 10 16526 2272 3 1
## 168 575 5 6119 2178 1 3
## 169 726 5 4034 1671 1 3
## 170 963 6 7417 1285 3 1
## 171 2765 9 9096 1486 3 1
## 172 1853 9 18608 2650 3 1
## 173 3333 15 15104 1737 1 1
## 174 7768 10 22566 2924 3 3
## 175 7690 0 0 3157 2 2
## 176 3135 5 3411 1982 3 1
## 177 504 2 6142 1845 2 2
## 178 2089 3 4920 2154 1 3
## 179 552 5 2447 1616 2 2
## 180 4268 14 22774 3095 1 1
## 181 1930 11 4915 1789 3 1
## 182 769 13 13574 1793 1 1
## 183 1845 3 3897 1013 3 1
## 184 6775 0 0 3085 2 2
## 185 1192 1 12197 1851 2 2
## 186 1809 6 1608 1378 3 3
## 187 539 4 6142 1782 2 2
## 188 3421 13 10925 1574 1 1
## 189 2189 6 12682 2140 3 1
## 190 2451 15 15030 3254 1 1
## 191 913 12 7430 1278 1 1
## 192 1522 1 10554 1636 2 2
## 193 875 2 8594 2040 3 3
## 194 2100 10 22006 2577 3 1
## 195 7810 0 0 3345 2 2
## 196 1815 6 6639 1214 3 3
## 197 4888 4 14661 1643 3 3
## 198 2274 7 19911 2301 1 1#npData$account_id <- as.factor(npData$account_id)
#npData$match_id <- as.factor(npData$match_id)
# gold <- "gold" %in% charvec
# gold_per_min <- "gold_per_min" %in% charvec
# kills <- "kills" %in% charvec
# tower_damage <- "tower_damage" %in% charvec
# duration <- "duration" %in% charvec
# lane <- "lane" %in% charvec
# lane_role <- "lane_role" %in% charvec
# net_worth <- "net_worth" %in% charvecEDA
varname <- "gold"
varHist <- ggplot(npData, aes(x=gold)) +
theme_bw() + #Set classic bw plot theme
geom_histogram(color="black", fill = "#34495E", alpha = 0.8, binwidth = 100) + #Color options, binwidth set to 100 shares
labs(x = "Win Or Lose", y = "Count", title = paste0("Counts of ", varname))
varHist
boxPlot <- ggplot(npData, aes(x=win,y=gold), fill = win) +
geom_hline(yintercept = median(npData$gold), size = 0.8) + #Add line for overall median shares
geom_point(size = 0.8) + #Add points
geom_boxplot(lwd = 0.5, width = 0.5, outlier.size = 0.8, alpha = 0.7) + #Create boxplot
xlab("") + ylab("Amount of Gold") + #Label axis
#theme(legend.position = "none") + #Remove legend
ggtitle(paste0("Wins vs ", varname))
#Display plot
boxPlot
ggplot(npData, aes(x = gold, y = net_worth, color = win, fill = win) ) +
geom_jitter() +
scale_color_manual(values=c("#999999", "#E69F00", "#56B4E9"))+
scale_fill_manual(values=c("#999999", "#E69F00", "#56B4E9"))+
labs(title = paste0("Networth over Gold"))+
theme_classic()
# Numerical Summaries
npData## # A tibble: 198 × 14
## match_id start_time win hero_id account_id leaguename gold_per_min
## <dbl> <dbl> <lgl> <dbl> <dbl> <chr> <dbl>
## 1 6653637943 1657339909 FALSE 53 250544263 Titus Colosseum … 569
## 2 6653911377 1657357190 FALSE 53 76082922 Destiny league 549
## 3 6654090150 1657366555 TRUE 53 1312740399 Ultras Dota Pro … 654
## 4 6656721191 1657512120 FALSE 53 209165399 Titus Colosseum … 472
## 5 6657130147 1657540852 TRUE 53 132851371 Dota 2 Champions… 894
## 6 6657972166 1657586272 TRUE 53 59752811 Thunderpick Bitc… 593
## 7 6658023411 1657591653 FALSE 53 85367644 Thunderpick Bitc… 442
## 8 6658141348 1657602904 FALSE 53 1312740399 Ultras Dota Pro … 434
## 9 6658904045 1657643886 TRUE 53 155370747 Dota 2 Champions… 701
## 10 6659582275 1657695291 TRUE 53 412910858 Destiny league 752
## # ℹ 188 more rows
## # ℹ 7 more variables: net_worth <dbl>, gold <dbl>, kills <dbl>,
## # tower_damage <dbl>, duration <dbl>, lane <dbl>, lane_role <dbl>playerMatches <- npData %>% select(win,account_id,match_id)
playerMatches## # A tibble: 198 × 3
## win account_id match_id
## <lgl> <dbl> <dbl>
## 1 FALSE 250544263 6653637943
## 2 FALSE 76082922 6653911377
## 3 TRUE 1312740399 6654090150
## 4 FALSE 209165399 6656721191
## 5 TRUE 132851371 6657130147
## 6 TRUE 59752811 6657972166
## 7 FALSE 85367644 6658023411
## 8 FALSE 1312740399 6658141348
## 9 TRUE 155370747 6658904045
## 10 TRUE 412910858 6659582275
## # ℹ 188 more rowsplayerWins <- select(playerMatches, -match_id)
#Total Wins by a Player
playerWins <- playerWins %>% group_by(account_id) %>% summarise(TotalWins = sum(win==1))
playerWins <- as.data.table(playerWins)
#Total Matches by a Player
playerMatches <- as.data.table(playerMatches)
playerMatches[, .(rowCount =.N), by=account_id]## account_id rowCount
## 1: 250544263 2
## 2: 76082922 1
## 3: 1312740399 2
## 4: 209165399 3
## 5: 132851371 1
## ---
## 116: 237527926 2
## 117: 1068676453 1
## 118: 1220632105 1
## 119: 138885864 2
## 120: 1251319262 1playerMatches <- as.data.table(playerMatches)
sumMatches <- playerMatches[, .(totalMatches =.N), by=account_id]
sumMatches <- playerMatches[sumMatches, on =.(account_id =account_id)]
sumMatches## win account_id match_id totalMatches
## 1: FALSE 250544263 6653637943 2
## 2: FALSE 250544263 6675282069 2
## 3: FALSE 76082922 6653911377 1
## 4: TRUE 1312740399 6654090150 2
## 5: FALSE 1312740399 6658141348 2
## ---
## 194: TRUE 1068676453 6694583595 1
## 195: FALSE 1220632105 6694689276 1
## 196: FALSE 138885864 6696003659 2
## 197: TRUE 138885864 6696012766 2
## 198: TRUE 1251319262 6696706878 1#Player Frame comprising of the player matches and wins
playerFrame <- sumMatches[playerWins, on =.(account_id =account_id)]
playerFrame <- playerFrame %>% mutate(win_perc = TotalWins/totalMatches) %>% arrange(desc(totalMatches))
playerFrame## win account_id match_id totalMatches TotalWins win_perc
## 1: FALSE 177411785 6670287920 9 6 0.6666667
## 2: TRUE 177411785 6671588490 9 6 0.6666667
## 3: TRUE 177411785 6671680767 9 6 0.6666667
## 4: FALSE 177411785 6682960303 9 6 0.6666667
## 5: TRUE 177411785 6683081104 9 6 0.6666667
## ---
## 194: TRUE 1251319262 6696706878 1 1 1.0000000
## 195: TRUE 1261414529 6664870055 1 1 1.0000000
## 196: TRUE 1263892079 6682553675 1 1 1.0000000
## 197: TRUE 1352080796 6674752354 1 1 1.0000000
## 198: TRUE 1414344010 6676358917 1 1 1.0000000#League Matches
leagueMatches <- npData %>% select(win,account_id,match_id,leaguename)
leagueMatches <- as.data.table(leagueMatches)
sumLeagueMatches <- leagueMatches[,.(totalLeagueMatches =.N), by=leaguename]
leagueMatches## win account_id match_id leaguename
## 1: FALSE 250544263 6653637943 Titus Colosseum Cup
## 2: FALSE 76082922 6653911377 Destiny league
## 3: TRUE 1312740399 6654090150 Ultras Dota Pro League
## 4: FALSE 209165399 6656721191 Titus Colosseum Cup
## 5: TRUE 132851371 6657130147 Dota 2 Champions League 2021-2022 Season
## ---
## 194: TRUE 345509021 6694863732 European Pro League Season 2
## 195: FALSE 138885864 6696003659 Titus Colosseum Cup Season 2
## 196: TRUE 138885864 6696012766 Titus Colosseum Cup Season 2
## 197: TRUE 27178898 6696170524 European Pro League Season 2
## 198: TRUE 1251319262 6696706878 Ultras Dota Pro League#League Wins
leagueWins <- leagueMatches %>% group_by(leaguename) %>% summarise(leagueTotalWins = sum(win==1))
leagueWins## # A tibble: 22 × 2
## leaguename leagueTotalWins
## <chr> <int>
## 1 Asia Battle Arena 8
## 2 Asian Tigers 3 11
## 3 Destiny league 10
## 4 Dota 2 Champions League 2021-2022 Season 3
## 5 Dota 2 Champions League 2022-2023 Season 6
## 6 ESL One Malaysia 2022 Qualifiers powered by Intel 21
## 7 European Pro League Season 2 2
## 8 Khaz Modan Cup Series Season 3 5
## 9 MInd Cup 2
## 10 Meister Provider Dota 2 Indonesia 2022 2
## # ℹ 12 more rowsleagueWins <- as.data.table(leagueWins)
#League Frame
leagueFrame <- sumLeagueMatches[leagueWins, on =.(leaguename =leaguename)]
leagueFrame <- leagueFrame %>% mutate(win_perc = leagueTotalWins/totalLeagueMatches) %>% arrange(desc(totalLeagueMatches))
#Numerical
avgTable <- table(avgKills=round(mean(npData$kills),digits=2),avgDur=round(mean(npData$duration),digits=2),avgGPM=round(mean(npData$gold_per_min),digits=2),avgNetWorth = round(mean(npData$net_worth),digits=2), avgTowerDmg=round(mean(npData$tower_damage),digits=2))
avgTable## , , avgGPM = 644.11, avgNetWorth = 20283.04, avgTowerDmg = 10066.26
##
## avgDur
## avgKills 2011.8
## 6.75 1kable(avgTable)| avgKills | avgDur | avgGPM | avgNetWorth | avgTowerDmg | Freq |
|---|---|---|---|---|---|
| 6.75 | 2011.8 | 644.11 | 20283.04 | 10066.26 | 1 |
#Categorical
#Create new categorical variables to classify game as high/normal/low GPM, networth, tower damage, kills,duration
table(npData$win,npData$lane,npData$lane_role)## , , = 1
##
##
## 1 2 3 4 5
## FALSE 28 0 21 0 0
## TRUE 23 0 38 0 0
##
## , , = 2
##
##
## 1 2 3 4 5
## FALSE 0 8 0 0 0
## TRUE 0 10 0 0 0
##
## , , = 3
##
##
## 1 2 3 4 5
## FALSE 14 0 16 0 0
## TRUE 17 0 21 0 0
##
## , , = 4
##
##
## 1 2 3 4 5
## FALSE 0 0 0 0 1
## TRUE 0 0 0 1 0#note that lane role determined by networth
categoryTable <- table(win=npData$win,lane=npData$lane,lanerole=npData$lane_role)
kable(categoryTable)| win | lane | lanerole | Freq |
|---|---|---|---|
| FALSE | 1 | 1 | 28 |
| TRUE | 1 | 1 | 23 |
| FALSE | 2 | 1 | 0 |
| TRUE | 2 | 1 | 0 |
| FALSE | 3 | 1 | 21 |
| TRUE | 3 | 1 | 38 |
| FALSE | 4 | 1 | 0 |
| TRUE | 4 | 1 | 0 |
| FALSE | 5 | 1 | 0 |
| TRUE | 5 | 1 | 0 |
| FALSE | 1 | 2 | 0 |
| TRUE | 1 | 2 | 0 |
| FALSE | 2 | 2 | 8 |
| TRUE | 2 | 2 | 10 |
| FALSE | 3 | 2 | 0 |
| TRUE | 3 | 2 | 0 |
| FALSE | 4 | 2 | 0 |
| TRUE | 4 | 2 | 0 |
| FALSE | 5 | 2 | 0 |
| TRUE | 5 | 2 | 0 |
| FALSE | 1 | 3 | 14 |
| TRUE | 1 | 3 | 17 |
| FALSE | 2 | 3 | 0 |
| TRUE | 2 | 3 | 0 |
| FALSE | 3 | 3 | 16 |
| TRUE | 3 | 3 | 21 |
| FALSE | 4 | 3 | 0 |
| TRUE | 4 | 3 | 0 |
| FALSE | 5 | 3 | 0 |
| TRUE | 5 | 3 | 0 |
| FALSE | 1 | 4 | 0 |
| TRUE | 1 | 4 | 0 |
| FALSE | 2 | 4 | 0 |
| TRUE | 2 | 4 | 0 |
| FALSE | 3 | 4 | 0 |
| TRUE | 3 | 4 | 0 |
| FALSE | 4 | 4 | 0 |
| TRUE | 4 | 4 | 1 |
| FALSE | 5 | 4 | 1 |
| TRUE | 5 | 4 | 0 |
#Modeling Page
Model Fitting
#Glm Fit # Change p = user input for proportion of data # Change var used to user input
#dataModel <- data.frame(npData)
#dataModel$win <- as.factor(dataModel$win)
if ("account_id" %in% colnames(dataModel)){
dataModel <- subset(dataModel, select = -account_id)
}
if ("match_id" %in% colnames(dataModel)){
dataModel <- subset(dataModel, select = -match_id)
}
if ("hero_id" %in% colnames(dataModel)){
dataModel <- subset(dataModel, select = -hero_id)
}
if ("leaguename" %in% colnames(dataModel)){
dataModel <- subset(dataModel, select = -leaguename)
}
if ("start_time" %in% colnames(dataModel)) {
dataModel <- subset(dataModel, select = -start_time)
}
dataModel## win gold_per_min net_worth gold kills tower_damage duration lane
## 1 FALSE 569 15770 1140 3 1221 1754 1
## 2 FALSE 549 15317 892 3 1559 1794 1
## 3 TRUE 654 16985 3210 10 19823 1547 3
## 4 FALSE 472 10399 449 4 1264 1319 1
## 5 TRUE 894 40098 9648 10 28469 2938 3
## 6 TRUE 593 21647 6622 5 10388 2435 3
## 7 FALSE 442 11580 2130 0 11727 1846 1
## 8 FALSE 434 9842 1192 0 6657 1315 3
## 9 TRUE 701 26720 4570 9 18541 2332 1
## 10 TRUE 752 24276 2576 13 9039 2263 3
## 11 TRUE 810 22700 2700 10 7885 1701 3
## 12 TRUE 483 15927 2377 1 18953 1957 3
## 13 TRUE 676 10746 1641 11 9049 934 3
## 14 FALSE 902 41321 3596 16 11169 3349 1
## 15 TRUE 721 18818 2243 22 4646 1533 3
## 16 FALSE 512 14369 1144 4 2475 1891 1
## 17 TRUE 412 12732 4407 0 8942 2002 3
## 18 FALSE 848 34584 5309 18 14206 2977 3
## 19 TRUE 559 15112 2687 4 16908 1879 1
## 20 FALSE 290 6108 958 2 0 1297 5
## 21 TRUE 485 9739 1364 4 645 1215 3
## 22 TRUE 655 13060 3310 9 16466 1182 1
## 23 FALSE 453 11319 269 1 446 1613 1
## 24 TRUE 689 26345 2020 10 23219 2473 3
## 25 FALSE 646 22515 1215 5 5258 2200 1
## 26 TRUE 560 10202 2447 3 2572 1040 3
## 27 TRUE 643 13318 1968 4 14306 1363 3
## 28 TRUE 677 21088 1338 8 8893 1938 3
## 29 TRUE 785 29306 1281 8 23614 2310 3
## 30 FALSE 588 16788 1063 5 3822 1838 3
## 31 TRUE 672 18876 651 11 9172 1672 1
## 32 TRUE 613 16366 2316 6 8085 1641 3
## 33 TRUE 808 27482 3307 6 14326 2083 3
## 34 TRUE 785 33771 1721 7 16460 2207 1
## 35 FALSE 442 8841 716 2 1474 1240 1
## 36 TRUE 723 26798 1273 9 15706 2278 1
## 37 TRUE 580 13682 1082 4 8754 1373 1
## 38 TRUE 805 27183 4008 14 13960 2142 3
## 39 FALSE 626 19381 1831 3 7552 1989 1
## 40 FALSE 480 12344 444 1 292 1706 1
## 41 FALSE 603 21299 3074 3 6085 2230 1
## 42 FALSE 533 16905 905 2 10574 2045 3
## 43 FALSE 455 8771 521 8 714 1352 3
## 44 TRUE 683 22392 3417 5 10921 2063 3
## 45 FALSE 777 29024 424 8 10553 2947 3
## 46 FALSE 625 23730 1605 4 855 2425 1
## 47 FALSE 578 20852 1277 8 12741 2516 2
## 48 TRUE 623 17109 3959 9 9364 1740 1
## 49 FALSE 624 21470 2970 2 14037 2417 1
## 50 TRUE 650 22308 3333 8 8823 2088 3
## 51 FALSE 559 16363 1238 3 2867 1851 1
## 52 TRUE 820 17797 1372 13 19135 1279 3
## 53 FALSE 680 18160 2310 6 5307 1686 3
## 54 TRUE 699 18374 3619 9 19038 1795 1
## 55 TRUE 605 14825 1320 3 22129 1464 1
## 56 TRUE 773 21653 1628 11 13320 1692 2
## 57 TRUE 821 29931 3881 11 16538 2190 2
## 58 TRUE 543 15297 2167 4 15882 1717 1
## 59 TRUE 630 16061 4486 8 5379 1614 1
## 60 FALSE 584 11481 1 4 1336 1527 1
## 61 FALSE 714 27986 581 4 16584 2752 1
## 62 TRUE 756 22428 1653 12 16618 1893 1
## 63 FALSE 487 10988 1183 1 2294 1420 1
## 64 FALSE 441 11678 578 2 3675 1735 1
## 65 TRUE 856 39130 5955 15 17614 3023 3
## 66 FALSE 560 13953 1428 2 6477 1710 1
## 67 FALSE 128 7808 7708 0 0 3312 2
## 68 FALSE 442 13145 365 1 5661 1890 3
## 69 TRUE 477 17071 2346 2 6867 2246 4
## 70 FALSE 493 11153 1278 3 5387 1436 3
## 71 FALSE 764 26141 1891 11 5526 2140 3
## 72 TRUE 725 21680 2830 11 10106 1801 1
## 73 FALSE 672 23782 957 7 13573 2647 3
## 74 FALSE 539 17176 976 2 2289 2203 3
## 75 FALSE 533 15389 389 3 8324 2066 1
## 76 TRUE 415 8209 2084 2 12088 1370 3
## 77 FALSE 659 28956 3881 8 8553 3038 3
## 78 TRUE 748 22375 3875 7 14898 1840 3
## 79 TRUE 617 19469 2039 6 16105 2172 3
## 80 FALSE 488 10869 569 6 2906 1610 1
## 81 FALSE 554 13675 1945 2 6836 1528 3
## 82 TRUE 807 28348 2148 7 8907 2252 3
## 83 FALSE 552 13818 1743 1 7569 1586 1
## 84 TRUE 829 34622 3297 12 22697 3626 3
## 85 TRUE 581 13529 2599 4 10267 1392 3
## 86 TRUE 592 12822 3747 8 9246 1278 3
## 87 FALSE 592 22001 2921 7 10944 2419 1
## 88 FALSE 577 18847 1647 5 4707 2108 1
## 89 TRUE 815 27546 4646 9 15996 2063 1
## 90 TRUE 696 21007 2907 1 13848 1882 3
## 91 TRUE 639 19413 2038 6 21849 1824 1
## 92 TRUE 948 32455 5705 20 18030 2054 1
## 93 TRUE 729 27526 3001 9 15313 2421 2
## 94 FALSE 629 23854 5229 4 15903 2505 3
## 95 FALSE 657 19874 349 6 2622 1909 3
## 96 FALSE 573 19265 1165 3 5936 2182 1
## 97 TRUE 813 25736 2436 13 19047 1962 1
## 98 FALSE 583 16306 456 5 3334 1997 1
## 99 TRUE 644 21768 4268 4 17180 2157 2
## 100 TRUE 678 23821 2671 7 12074 2156 3
## 101 TRUE 671 13065 1340 9 3208 1112 3
## 102 TRUE 818 21247 1747 17 21430 1651 2
## 103 TRUE 759 22575 1275 12 18738 1819 1
## 104 TRUE 802 25566 2741 8 10292 1963 3
## 105 TRUE 756 17329 2079 12 12953 1382 3
## 106 TRUE 616 18119 1369 7 11799 1818 1
## 107 FALSE 694 21558 1033 2 4663 2162 3
## 108 TRUE 624 17438 1438 9 9004 1760 1
## 109 TRUE 641 18588 1108 10 7872 1850 3
## 110 FALSE 632 21294 394 2 7872 2373 3
## 111 FALSE 832 25164 589 8 8530 2606 1
## 112 FALSE 828 31286 4386 11 16196 2404 1
## 113 FALSE 938 37910 1885 9 15057 3504 1
## 114 FALSE 521 15167 317 2 3777 1893 1
## 115 FALSE 698 24601 1126 8 6679 2563 3
## 116 FALSE 651 21659 929 11 5158 2305 3
## 117 FALSE 780 29327 1302 5 11033 2633 1
## 118 TRUE 701 25954 2204 9 13424 2271 1
## 119 TRUE 842 32565 8615 11 16681 2405 3
## 120 TRUE 558 8796 1596 8 1858 909 3
## 121 TRUE 495 6596 746 5 1492 800 3
## 122 TRUE 594 12938 713 12 6280 1336 3
## 123 TRUE 663 16437 1262 9 4386 1464 1
## 124 TRUE 714 32550 5445 2 25778 2869 1
## 125 TRUE 692 13663 1113 10 9585 1173 1
## 126 FALSE 777 30697 1722 5 8765 2484 3
## 127 FALSE 451 11790 440 2 3163 2032 1
## 128 TRUE 915 23816 641 20 14435 1596 3
## 129 TRUE 668 11438 988 5 3828 1029 3
## 130 FALSE 332 7461 361 2 582 1359 1
## 131 FALSE 712 23657 482 11 7827 2092 3
## 132 TRUE 801 20299 1474 8 18974 1534 1
## 133 TRUE 660 28349 899 6 27940 2647 2
## 134 TRUE 838 34085 5235 12 27195 2690 3
## 135 FALSE 833 38657 4957 8 13468 3477 3
## 136 FALSE 729 26248 1498 8 6838 2558 3
## 137 TRUE 856 37695 7820 3 15473 2722 1
## 138 FALSE 540 19767 1217 0 7462 2249 3
## 139 FALSE 610 20269 1589 2 5696 2365 3
## 140 TRUE 638 19220 1945 8 9261 1859 1
## 141 TRUE 621 18086 2086 4 20104 1782 3
## 142 TRUE 727 23482 457 7 15069 2264 3
## 143 FALSE 887 33534 2359 10 14573 2954 1
## 144 TRUE 374 7934 1284 2 3822 1569 3
## 145 TRUE 764 31745 945 16 8973 2646 1
## 146 TRUE 631 18752 3502 5 9250 1834 1
## 147 TRUE 689 20105 2080 2 15605 1780 3
## 148 FALSE 618 19472 1872 3 2206 1923 1
## 149 FALSE 542 17019 1519 2 5704 2054 3
## 150 FALSE 596 18836 261 8 2933 2251 1
## 151 TRUE 672 14803 2053 9 3602 1311 3
## 152 TRUE 561 15597 392 7 9377 1737 3
## 153 FALSE 871 42703 3328 12 15210 3667 2
## 154 FALSE 492 10583 2233 6 1619 1414 3
## 155 TRUE 586 11254 1179 3 8821 1134 2
## 156 FALSE 456 9801 271 4 757 1451 3
## 157 FALSE 641 19453 3 7 4454 2452 1
## 158 TRUE 648 15604 1329 12 5930 1491 3
## 159 FALSE 655 26640 160 8 9129 2696 3
## 160 TRUE 563 15107 3357 7 3297 1676 1
## 161 TRUE 592 23429 5054 5 11500 2578 1
## 162 FALSE 824 35087 1512 12 9796 3144 1
## 163 TRUE 645 18321 3096 7 14920 1701 3
## 164 TRUE 775 24101 1576 4 14679 1896 3
## 165 TRUE 932 41313 6813 20 24147 2753 1
## 166 FALSE 513 14353 828 3 1670 1906 3
## 167 TRUE 751 28538 5113 10 16526 2272 3
## 168 FALSE 475 15255 575 5 6119 2178 1
## 169 FALSE 570 15176 726 5 4034 1671 1
## 170 TRUE 598 13113 963 6 7417 1285 3
## 171 TRUE 678 16985 2765 9 9096 1486 3
## 172 FALSE 877 31353 1853 9 18608 2650 3
## 173 TRUE 831 23908 3333 15 15104 1737 1
## 174 TRUE 924 41743 7768 10 22566 2924 3
## 175 TRUE 134 7790 7690 0 0 3157 2
## 176 FALSE 624 19285 3135 5 3411 1982 3
## 177 FALSE 664 19179 504 2 6142 1845 2
## 178 TRUE 525 18069 2089 3 4920 2154 1
## 179 FALSE 591 15327 552 5 2447 1616 2
## 180 TRUE 919 41418 4268 14 22774 3095 1
## 181 FALSE 640 18230 1930 11 4915 1789 3
## 182 TRUE 777 22294 769 13 13574 1793 1
## 183 TRUE 516 9095 1845 3 3897 1013 3
## 184 FALSE 120 6875 6775 0 0 3085 2
## 185 TRUE 659 19467 1192 1 12197 1851 2
## 186 FALSE 432 8934 1809 6 1608 1378 3
## 187 FALSE 667 19039 539 4 6142 1782 2
## 188 TRUE 706 18046 3421 13 10925 1574 1
## 189 FALSE 722 24314 2189 6 12682 2140 3
## 190 FALSE 876 41076 2451 15 15030 3254 1
## 191 TRUE 792 16988 913 12 7430 1278 1
## 192 TRUE 518 13922 1522 1 10554 1636 2
## 193 FALSE 431 12950 875 2 8594 2040 3
## 194 TRUE 838 30875 2100 10 22006 2577 3
## 195 FALSE 129 7910 7810 0 0 3345 2
## 196 TRUE 567 11645 1815 6 6639 1214 3
## 197 TRUE 612 15363 4888 4 14661 1643 3
## 198 TRUE 760 27049 2274 7 19911 2301 1
## lane_role
## 1 3
## 2 3
## 3 3
## 4 1
## 5 1
## 6 1
## 7 1
## 8 3
## 9 3
## 10 3
## 11 1
## 12 3
## 13 1
## 14 1
## 15 1
## 16 3
## 17 3
## 18 3
## 19 3
## 20 4
## 21 3
## 22 1
## 23 1
## 24 3
## 25 1
## 26 1
## 27 1
## 28 3
## 29 1
## 30 1
## 31 1
## 32 1
## 33 1
## 34 1
## 35 1
## 36 1
## 37 1
## 38 1
## 39 1
## 40 1
## 41 1
## 42 3
## 43 1
## 44 1
## 45 1
## 46 1
## 47 2
## 48 3
## 49 3
## 50 1
## 51 1
## 52 1
## 53 1
## 54 3
## 55 3
## 56 2
## 57 2
## 58 3
## 59 1
## 60 1
## 61 3
## 62 1
## 63 1
## 64 1
## 65 1
## 66 1
## 67 2
## 68 3
## 69 4
## 70 3
## 71 1
## 72 1
## 73 1
## 74 1
## 75 3
## 76 1
## 77 3
## 78 1
## 79 3
## 80 3
## 81 1
## 82 1
## 83 3
## 84 1
## 85 3
## 86 3
## 87 3
## 88 1
## 89 1
## 90 1
## 91 3
## 92 1
## 93 2
## 94 3
## 95 1
## 96 1
## 97 1
## 98 1
## 99 2
## 100 3
## 101 1
## 102 2
## 103 1
## 104 1
## 105 1
## 106 3
## 107 1
## 108 3
## 109 3
## 110 3
## 111 1
## 112 3
## 113 1
## 114 1
## 115 1
## 116 1
## 117 1
## 118 1
## 119 1
## 120 3
## 121 3
## 122 1
## 123 1
## 124 3
## 125 1
## 126 1
## 127 1
## 128 1
## 129 1
## 130 1
## 131 1
## 132 1
## 133 2
## 134 1
## 135 1
## 136 3
## 137 3
## 138 3
## 139 1
## 140 3
## 141 3
## 142 1
## 143 3
## 144 1
## 145 3
## 146 3
## 147 1
## 148 1
## 149 3
## 150 1
## 151 3
## 152 3
## 153 2
## 154 3
## 155 2
## 156 3
## 157 3
## 158 1
## 159 3
## 160 3
## 161 3
## 162 1
## 163 3
## 164 1
## 165 1
## 166 1
## 167 1
## 168 3
## 169 3
## 170 1
## 171 1
## 172 1
## 173 1
## 174 3
## 175 2
## 176 1
## 177 2
## 178 3
## 179 2
## 180 1
## 181 1
## 182 1
## 183 1
## 184 2
## 185 2
## 186 3
## 187 2
## 188 1
## 189 1
## 190 1
## 191 1
## 192 2
## 193 3
## 194 1
## 195 2
## 196 3
## 197 3
## 198 1# class(dataModel$win)
# class(dataModel$gold_per_min)
# class(dataModel$net_worth)
# class(dataModel$gold)
# class(dataModel$kills)
# class(dataModel$tower_damage)
# class(dataModel$duration)
# class(dataModel$lane)
# class(dataModel$lane_role)keepVars <- c("gold_per_min","kills")
keepVars <- c(keepVars, "win")
dataModel <- dataModel[, names(dataModel) %in% keepVars]
dataModel## win gold_per_min kills
## 1 FALSE 569 3
## 2 FALSE 549 3
## 3 TRUE 654 10
## 4 FALSE 472 4
## 5 TRUE 894 10
## 6 TRUE 593 5
## 7 FALSE 442 0
## 8 FALSE 434 0
## 9 TRUE 701 9
## 10 TRUE 752 13
## 11 TRUE 810 10
## 12 TRUE 483 1
## 13 TRUE 676 11
## 14 FALSE 902 16
## 15 TRUE 721 22
## 16 FALSE 512 4
## 17 TRUE 412 0
## 18 FALSE 848 18
## 19 TRUE 559 4
## 20 FALSE 290 2
## 21 TRUE 485 4
## 22 TRUE 655 9
## 23 FALSE 453 1
## 24 TRUE 689 10
## 25 FALSE 646 5
## 26 TRUE 560 3
## 27 TRUE 643 4
## 28 TRUE 677 8
## 29 TRUE 785 8
## 30 FALSE 588 5
## 31 TRUE 672 11
## 32 TRUE 613 6
## 33 TRUE 808 6
## 34 TRUE 785 7
## 35 FALSE 442 2
## 36 TRUE 723 9
## 37 TRUE 580 4
## 38 TRUE 805 14
## 39 FALSE 626 3
## 40 FALSE 480 1
## 41 FALSE 603 3
## 42 FALSE 533 2
## 43 FALSE 455 8
## 44 TRUE 683 5
## 45 FALSE 777 8
## 46 FALSE 625 4
## 47 FALSE 578 8
## 48 TRUE 623 9
## 49 FALSE 624 2
## 50 TRUE 650 8
## 51 FALSE 559 3
## 52 TRUE 820 13
## 53 FALSE 680 6
## 54 TRUE 699 9
## 55 TRUE 605 3
## 56 TRUE 773 11
## 57 TRUE 821 11
## 58 TRUE 543 4
## 59 TRUE 630 8
## 60 FALSE 584 4
## 61 FALSE 714 4
## 62 TRUE 756 12
## 63 FALSE 487 1
## 64 FALSE 441 2
## 65 TRUE 856 15
## 66 FALSE 560 2
## 67 FALSE 128 0
## 68 FALSE 442 1
## 69 TRUE 477 2
## 70 FALSE 493 3
## 71 FALSE 764 11
## 72 TRUE 725 11
## 73 FALSE 672 7
## 74 FALSE 539 2
## 75 FALSE 533 3
## 76 TRUE 415 2
## 77 FALSE 659 8
## 78 TRUE 748 7
## 79 TRUE 617 6
## 80 FALSE 488 6
## 81 FALSE 554 2
## 82 TRUE 807 7
## 83 FALSE 552 1
## 84 TRUE 829 12
## 85 TRUE 581 4
## 86 TRUE 592 8
## 87 FALSE 592 7
## 88 FALSE 577 5
## 89 TRUE 815 9
## 90 TRUE 696 1
## 91 TRUE 639 6
## 92 TRUE 948 20
## 93 TRUE 729 9
## 94 FALSE 629 4
## 95 FALSE 657 6
## 96 FALSE 573 3
## 97 TRUE 813 13
## 98 FALSE 583 5
## 99 TRUE 644 4
## 100 TRUE 678 7
## 101 TRUE 671 9
## 102 TRUE 818 17
## 103 TRUE 759 12
## 104 TRUE 802 8
## 105 TRUE 756 12
## 106 TRUE 616 7
## 107 FALSE 694 2
## 108 TRUE 624 9
## 109 TRUE 641 10
## 110 FALSE 632 2
## 111 FALSE 832 8
## 112 FALSE 828 11
## 113 FALSE 938 9
## 114 FALSE 521 2
## 115 FALSE 698 8
## 116 FALSE 651 11
## 117 FALSE 780 5
## 118 TRUE 701 9
## 119 TRUE 842 11
## 120 TRUE 558 8
## 121 TRUE 495 5
## 122 TRUE 594 12
## 123 TRUE 663 9
## 124 TRUE 714 2
## 125 TRUE 692 10
## 126 FALSE 777 5
## 127 FALSE 451 2
## 128 TRUE 915 20
## 129 TRUE 668 5
## 130 FALSE 332 2
## 131 FALSE 712 11
## 132 TRUE 801 8
## 133 TRUE 660 6
## 134 TRUE 838 12
## 135 FALSE 833 8
## 136 FALSE 729 8
## 137 TRUE 856 3
## 138 FALSE 540 0
## 139 FALSE 610 2
## 140 TRUE 638 8
## 141 TRUE 621 4
## 142 TRUE 727 7
## 143 FALSE 887 10
## 144 TRUE 374 2
## 145 TRUE 764 16
## 146 TRUE 631 5
## 147 TRUE 689 2
## 148 FALSE 618 3
## 149 FALSE 542 2
## 150 FALSE 596 8
## 151 TRUE 672 9
## 152 TRUE 561 7
## 153 FALSE 871 12
## 154 FALSE 492 6
## 155 TRUE 586 3
## 156 FALSE 456 4
## 157 FALSE 641 7
## 158 TRUE 648 12
## 159 FALSE 655 8
## 160 TRUE 563 7
## 161 TRUE 592 5
## 162 FALSE 824 12
## 163 TRUE 645 7
## 164 TRUE 775 4
## 165 TRUE 932 20
## 166 FALSE 513 3
## 167 TRUE 751 10
## 168 FALSE 475 5
## 169 FALSE 570 5
## 170 TRUE 598 6
## 171 TRUE 678 9
## 172 FALSE 877 9
## 173 TRUE 831 15
## 174 TRUE 924 10
## 175 TRUE 134 0
## 176 FALSE 624 5
## 177 FALSE 664 2
## 178 TRUE 525 3
## 179 FALSE 591 5
## 180 TRUE 919 14
## 181 FALSE 640 11
## 182 TRUE 777 13
## 183 TRUE 516 3
## 184 FALSE 120 0
## 185 TRUE 659 1
## 186 FALSE 432 6
## 187 FALSE 667 4
## 188 TRUE 706 13
## 189 FALSE 722 6
## 190 FALSE 876 15
## 191 TRUE 792 12
## 192 TRUE 518 1
## 193 FALSE 431 2
## 194 TRUE 838 10
## 195 FALSE 129 0
## 196 TRUE 567 6
## 197 TRUE 612 4
## 198 TRUE 760 7dataSubset <- dataModel[, names(dataModel) %in% keepVars]dataIndex <- createDataPartition(dataModel$win, p = 0.75, list = FALSE)
class(dataIndex)## [1] "matrix" "array"dataTrain <- dataModel[dataIndex, ]
dataTest <- dataModel[-dataIndex, ]glmFit <- train(win ~ ., data = dataTrain,
method = "glm",
family = "binomial",
preProcess = c("center", "scale"),
trControl = trainControl(method = "cv", number = 10))
predictions <- predict(glmFit, dataTrain)
devia<-summary(glmFit)$deviance
coef <-summary(glmFit)$coefficients
predictions## [1] FALSE TRUE TRUE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE TRUE
## [13] FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE TRUE FALSE
## [25] TRUE TRUE TRUE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE TRUE
## [37] TRUE FALSE TRUE TRUE FALSE TRUE TRUE TRUE FALSE TRUE FALSE FALSE
## [49] TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE
## [61] TRUE TRUE TRUE FALSE TRUE FALSE TRUE TRUE TRUE FALSE TRUE FALSE
## [73] FALSE TRUE TRUE TRUE TRUE FALSE TRUE FALSE TRUE TRUE FALSE TRUE
## [85] TRUE FALSE TRUE TRUE TRUE TRUE TRUE FALSE FALSE FALSE TRUE FALSE
## [97] TRUE TRUE TRUE TRUE TRUE FALSE FALSE FALSE TRUE TRUE FALSE FALSE
## [109] FALSE FALSE TRUE TRUE TRUE FALSE FALSE TRUE TRUE TRUE TRUE TRUE
## [121] FALSE TRUE FALSE TRUE TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE
## [133] TRUE TRUE TRUE FALSE FALSE TRUE FALSE TRUE TRUE TRUE TRUE FALSE
## [145] FALSE TRUE FALSE TRUE FALSE
## Levels: FALSE TRUEdevia## [1] 183.5477coef## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.29374756 0.1799485 1.6323981 0.10259566
## gold_per_min -0.09331071 0.2594704 -0.3596198 0.71913145
## kills 0.93775334 0.2940190 3.1894313 0.00142553confusionMatrix(data = dataTest$win, reference = predict(glmFit, newdata = dataTest))## Confusion Matrix and Statistics
##
## Reference
## Prediction FALSE TRUE
## FALSE 11 11
## TRUE 9 18
##
## Accuracy : 0.5918
## 95% CI : (0.4421, 0.73)
## No Information Rate : 0.5918
## P-Value [Acc > NIR] : 0.5611
##
## Kappa : 0.1681
##
## Mcnemar's Test P-Value : 0.8231
##
## Sensitivity : 0.5500
## Specificity : 0.6207
## Pos Pred Value : 0.5000
## Neg Pred Value : 0.6667
## Prevalence : 0.4082
## Detection Rate : 0.2245
## Detection Prevalence : 0.4490
## Balanced Accuracy : 0.5853
##
## 'Positive' Class : FALSE
## Classification Tree
Change . to user input
classTreeFit <- tree(win ~ ., data = dataTrain) # The '.' means all variables to be used as explanatory variables
summary(classTreeFit)##
## Classification tree:
## tree(formula = win ~ ., data = dataTrain)
## Number of terminal nodes: 9
## Residual mean deviance: 0.9625 = 134.7 / 140
## Misclassification error rate: 0.2617 = 39 / 149class(classTreeFit)## [1] "tree"Confusion Matrix of Accuracy
df <- data.frame(matrix(rnorm(16), nrow = 4))
colnames(df) <- paste0("Col", 1:4)
rownames(df) <- paste0("Row", 1:4)###Pruning
pruneFit <- cv.tree(classTreeFit, FUN = prune.misclass)
summary(pruneFit)## Length Class Mode
## size 6 -none- numeric
## dev 6 -none- numeric
## k 6 -none- numeric
## method 1 -none- characterplot(pruneFit$size, pruneFit$dev, type = "b")
pruneFit## $size
## [1] 9 8 5 3 2 1
##
## $dev
## [1] 54 56 56 55 47 68
##
## $k
## [1] -Inf 0.0000000 0.6666667 1.0000000 3.0000000 20.0000000
##
## $method
## [1] "misclass"
##
## attr(,"class")
## [1] "prune" "tree.sequence"dfPruneFit <- cbind(size=pruneFit$size,dev=pruneFit$dev)
dfPruneFit <- data.frame(dfPruneFit)
dfPruneFit <- dfPruneFit %>% group_by(size)%>%arrange(size)%>%arrange(dev)
dfPruneFit## # A tibble: 6 × 2
## # Groups: size [6]
## size dev
## <dbl> <dbl>
## 1 2 47
## 2 9 54
## 3 3 55
## 4 5 56
## 5 8 56
## 6 1 68bestVal <- dfPruneFit$size[1]
bestVal## [1] 2Pruned Predictions
pruneFitFinal <- prune.misclass(classTreeFit, best = bestVal)
summary(pruneFitFinal)##
## Classification tree:
## snip.tree(tree = classTreeFit, nodes = 3:2)
## Variables actually used in tree construction:
## [1] "kills"
## Number of terminal nodes: 2
## Residual mean deviance: 1.232 = 181.1 / 147
## Misclassification error rate: 0.3087 = 46 / 149fullPred <- predict(classTreeFit, dplyr::select(dataTest, -"win"), type = "class")
prunePred <- predict(pruneFitFinal, dplyr::select(dataTest, -"win"), type = "class")
prunePred## [1] FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE TRUE TRUE
## [13] FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE TRUE TRUE TRUE FALSE
## [25] FALSE FALSE TRUE TRUE TRUE TRUE TRUE TRUE TRUE FALSE TRUE FALSE
## [37] TRUE TRUE FALSE TRUE TRUE TRUE FALSE FALSE TRUE FALSE FALSE FALSE
## [49] TRUE
## Levels: FALSE TRUEComparison Full Fit
cm <- confusionMatrix(fullPred,dataTest$win)
cmFull <- cm$table
cmFull## Reference
## Prediction FALSE TRUE
## FALSE 12 18
## TRUE 10 9accFull <- cm$overall[1]
list(cmFull,accFull)## [[1]]
## Reference
## Prediction FALSE TRUE
## FALSE 12 18
## TRUE 10 9
##
## [[2]]
## Accuracy
## 0.4285714Comparison Pruned Fit
pruneTbl <- table(data.frame(prunePred, dataTest[, "win"]))
kable(pruneTbl)| FALSE | TRUE | |
|---|---|---|
| FALSE | 13 | 11 |
| TRUE | 9 | 16 |
accPrune<-sum(diag(pruneTbl)/sum(pruneTbl))
print(accPrune)## [1] 0.5918367Random Forest
Change the . on 254 to user input
Here we change our mtry to be ncol(diamondsTrain)/3 since this example was regression
trainRFModel <- train(win ~ ., data = dataTrain,
method = "rf",
trControl = trainControl(method = "repeatedcv", number = 5, repeats = 3),
tuneGrid = data.frame(mtry = sqrt(ncol(dataTrain) - 1)))
trainConMat <- confusionMatrix(trainRFModel, newdata = dataTest)
#testConMat <- confusionMatrix(data = dataTest$win, reference = predict(trainRFModel, newdata = dataTest))
trainConMat## Cross-Validated (5 fold, repeated 3 times) Confusion Matrix
##
## (entries are percentual average cell counts across resamples)
##
## Reference
## Prediction FALSE TRUE
## FALSE 25.5 18.3
## TRUE 18.8 37.4
##
## Accuracy (average) : 0.6286#testConMat
summary(trainRFModel)## Length Class Mode
## call 4 -none- call
## type 1 -none- character
## predicted 149 factor numeric
## err.rate 1500 -none- numeric
## confusion 6 -none- numeric
## votes 298 matrix numeric
## oob.times 149 -none- numeric
## classes 2 -none- character
## importance 2 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 149 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 2 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## param 0 -none- listpredict(trainRFModel,newdata=dataTest)## [1] FALSE TRUE FALSE TRUE TRUE TRUE TRUE FALSE FALSE TRUE FALSE FALSE
## [13] TRUE FALSE TRUE FALSE FALSE TRUE FALSE TRUE TRUE TRUE TRUE TRUE
## [25] TRUE FALSE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE FALSE
## [37] TRUE TRUE FALSE TRUE TRUE TRUE FALSE FALSE TRUE TRUE TRUE FALSE
## [49] TRUE
## Levels: FALSE TRUE#Classification Using mlr3
#https://mlr3book.mlr-org.com/basics.html#learners
#npTask <- tsk("npData")
data("npData", package = "datasets")
str(npData)## spc_tbl_ [198 × 14] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
## $ match_id : num [1:198] 6.65e+09 6.65e+09 6.65e+09 6.66e+09 6.66e+09 ...
## $ start_time : num [1:198] 1.66e+09 1.66e+09 1.66e+09 1.66e+09 1.66e+09 ...
## $ win : logi [1:198] FALSE FALSE TRUE FALSE TRUE TRUE ...
## $ hero_id : num [1:198] 53 53 53 53 53 53 53 53 53 53 ...
## $ account_id : num [1:198] 2.51e+08 7.61e+07 1.31e+09 2.09e+08 1.33e+08 ...
## $ leaguename : chr [1:198] "Titus Colosseum Cup" "Destiny league" "Ultras Dota Pro League" "Titus Colosseum Cup" ...
## $ gold_per_min: num [1:198] 569 549 654 472 894 593 442 434 701 752 ...
## $ net_worth : num [1:198] 15770 15317 16985 10399 40098 ...
## $ gold : num [1:198] 1140 892 3210 449 9648 ...
## $ kills : num [1:198] 3 3 10 4 10 5 0 0 9 13 ...
## $ tower_damage: num [1:198] 1221 1559 19823 1264 28469 ...
## $ duration : num [1:198] 1754 1794 1547 1319 2938 ...
## $ lane : num [1:198] 1 1 3 1 3 3 1 3 1 3 ...
## $ lane_role : num [1:198] 3 3 3 1 1 1 1 3 3 3 ...
## - attr(*, "spec")=
## .. cols(
## .. match_id = col_double(),
## .. start_time = col_double(),
## .. win = col_logical(),
## .. hero_id = col_double(),
## .. account_id = col_double(),
## .. leaguename = col_character(),
## .. gold_per_min = col_double(),
## .. net_worth = col_double(),
## .. gold = col_double(),
## .. kills = col_double(),
## .. tower_damage = col_double(),
## .. duration = col_double(),
## .. lane = col_double(),
## .. lane_role = col_double()
## .. )
## - attr(*, "problems")=<externalptr>npSubset = subset(npData, select = c("win","gold_per_min","net_worth","duration"))
taskNP = as_task_classif(npSubset, target = "win", id = "wins", positive = "TRUE") #We can set the variable of interest here
taskNP## <TaskClassif:wins> (198 x 4)
## * Target: win
## * Properties: twoclass
## * Features (3):
## - dbl (3): duration, gold_per_min, net_worthautoplot(taskNP, type = "pairs")
#Checking Elements of TaskNP
c(taskNP$nrow, taskNP$ncol) #Get size of data object## [1] 198 4list(taskNP$feature_names, taskNP$target_names) #we can see which variables are features vs targets from our earlier as_task_clasif## [[1]]
## [1] "duration" "gold_per_min" "net_worth"
##
## [[2]]
## [1] "win"head(taskNP$row_ids) #Pull indexes of the task## [1] 1 2 3 4 5 6taskNP$data() #Look at data with this## win duration gold_per_min net_worth
## 1: FALSE 1754 569 15770
## 2: FALSE 1794 549 15317
## 3: TRUE 1547 654 16985
## 4: FALSE 1319 472 10399
## 5: TRUE 2938 894 40098
## ---
## 194: TRUE 2577 838 30875
## 195: FALSE 3345 129 7910
## 196: TRUE 1214 567 11645
## 197: TRUE 1643 612 15363
## 198: TRUE 2301 760 27049taskNP$data(rows = c(1, 3, 10), cols = "duration") #Pull specific data with this## duration
## 1: 1754
## 2: 1547
## 3: 2263summary(as.data.table(taskNP)) #Pull everything out## win duration gold_per_min net_worth
## TRUE :110 Min. : 800 Min. :120.0 Min. : 6108
## FALSE: 88 1st Qu.:1614 1st Qu.:560.0 1st Qu.:14357
## Median :1908 Median :644.5 Median :19109
## Mean :2012 Mean :644.1 Mean :20283
## 3rd Qu.:2326 3rd Qu.:755.0 3rd Qu.:24529
## Max. :3667 Max. :948.0 Max. :42703Task Mutators with mlr3
taskNPSmall = taskNP$clone() # use clone method to create copy
taskNPSmall$select(c("gold_per_min", "duration")) # keep only these features
taskNPSmall$filter(2:4) # keep only these rows
taskNPSmall$data()## win duration gold_per_min
## 1: FALSE 1794 549
## 2: TRUE 1547 654
## 3: FALSE 1319 472#Lets make it even smaller
taskNPSmaller = taskNPSmall$clone()
taskNPSmaller$filter(2) #keep only the 2nd row
taskNPSmaller$data()## win duration gold_per_min
## 1: FALSE 1794 549#Create a learner
classifLrn <- lrn("classif.rpart")
#Partition our Data (default uses 33/67)
splits = partition(taskNP)
splits## $train
## [1] 1 4 7 8 14 20 25 30 42 47 49 51 53 60 61 64 66 67
## [19] 68 73 74 75 77 80 81 87 95 96 98 107 110 111 112 113 114 115
## [37] 116 117 130 131 136 150 153 154 157 159 162 166 169 172 176 181 184 186
## [55] 187 189 190 193 195 3 9 11 12 13 15 19 21 27 28 31 32 33
## [73] 34 36 38 48 52 54 55 56 57 58 59 72 76 78 82 85 89 91
## [91] 92 97 100 102 105 106 109 118 119 120 121 122 123 124 125 128 132 133
## [109] 134 137 144 145 146 147 151 155 158 160 161 167 171 173 174 175 178 180
## [127] 182 183 185 188 191 192 197
##
## $test
## [1] 2 16 18 23 35 39 40 41 43 45 46 63 70 71 83 88 94 126 127
## [20] 135 138 139 143 148 149 156 168 177 179 5 6 10 17 22 24 26 29 37
## [39] 44 50 62 65 69 79 84 86 90 93 99 101 103 104 108 129 140 141 142
## [58] 152 163 164 165 170 194 196 198classifLrn$train(taskNP, splits$train)
classifLrn$model## n= 133
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 133 59 TRUE (0.55639098 0.44360902)
## 2) duration< 1884.5 67 19 TRUE (0.71641791 0.28358209)
## 4) gold_per_min>=591 39 3 TRUE (0.92307692 0.07692308) *
## 5) gold_per_min< 591 28 12 FALSE (0.42857143 0.57142857)
## 10) duration< 1317 7 2 TRUE (0.71428571 0.28571429) *
## 11) duration>=1317 21 7 FALSE (0.33333333 0.66666667) *
## 3) duration>=1884.5 66 26 FALSE (0.39393939 0.60606061)
## 6) net_worth>=25450 32 12 TRUE (0.62500000 0.37500000)
## 12) duration< 2368 12 0 TRUE (1.00000000 0.00000000) *
## 13) duration>=2368 20 8 FALSE (0.40000000 0.60000000)
## 26) net_worth>=31549 12 5 TRUE (0.58333333 0.41666667) *
## 27) net_worth< 31549 8 1 FALSE (0.12500000 0.87500000) *
## 7) net_worth< 25450 34 6 FALSE (0.17647059 0.82352941) *classifLrn$param_set #to view hyper parameters, current settings are stored in the value column## <ParamSet>
## id class lower upper nlevels default value
## 1: cp ParamDbl 0 1 Inf 0.01
## 2: keep_model ParamLgl NA NA 2 FALSE
## 3: maxcompete ParamInt 0 Inf Inf 4
## 4: maxdepth ParamInt 1 30 30 30
## 5: maxsurrogate ParamInt 0 Inf Inf 5
## 6: minbucket ParamInt 1 Inf Inf <NoDefault[3]>
## 7: minsplit ParamInt 1 Inf Inf 20
## 8: surrogatestyle ParamInt 0 1 2 0
## 9: usesurrogate ParamInt 0 2 3 2
## 10: xval ParamInt 0 Inf Inf 10 0#classifLrn$param_set$set_values(minsplit = 10) if you want to set the parameters
#classifLrn = lrn("classif.rpart", minsplit =10) hyperparameters can be done at creation of learner
#Predicting
predsClass = classifLrn$predict(taskNP,splits$test)
predsClass #truth is what actually happened, response is the prediction## <PredictionClassif> for 65 observations:
## row_ids truth response
## 2 FALSE FALSE
## 16 FALSE FALSE
## 18 FALSE TRUE
## ---
## 194 TRUE FALSE
## 196 TRUE TRUE
## 198 TRUE TRUEas.data.table(predsClass)## row_ids truth response
## 1: 2 FALSE FALSE
## 2: 16 FALSE FALSE
## 3: 18 FALSE TRUE
## 4: 23 FALSE FALSE
## 5: 35 FALSE TRUE
## 6: 39 FALSE FALSE
## 7: 40 FALSE FALSE
## 8: 41 FALSE FALSE
## 9: 43 FALSE FALSE
## 10: 45 FALSE FALSE
## 11: 46 FALSE FALSE
## 12: 63 FALSE FALSE
## 13: 70 FALSE FALSE
## 14: 71 FALSE TRUE
## 15: 83 FALSE FALSE
## 16: 88 FALSE FALSE
## 17: 94 FALSE FALSE
## 18: 126 FALSE FALSE
## 19: 127 FALSE FALSE
## 20: 135 FALSE TRUE
## 21: 138 FALSE FALSE
## 22: 139 FALSE FALSE
## 23: 143 FALSE TRUE
## 24: 148 FALSE FALSE
## 25: 149 FALSE FALSE
## 26: 156 FALSE FALSE
## 27: 168 FALSE FALSE
## 28: 177 FALSE TRUE
## 29: 179 FALSE TRUE
## 30: 5 TRUE TRUE
## 31: 6 TRUE FALSE
## 32: 10 TRUE FALSE
## 33: 17 TRUE FALSE
## 34: 22 TRUE TRUE
## 35: 24 TRUE FALSE
## 36: 26 TRUE TRUE
## 37: 29 TRUE TRUE
## 38: 37 TRUE FALSE
## 39: 44 TRUE FALSE
## 40: 50 TRUE FALSE
## 41: 62 TRUE FALSE
## 42: 65 TRUE TRUE
## 43: 69 TRUE FALSE
## 44: 79 TRUE FALSE
## 45: 84 TRUE TRUE
## 46: 86 TRUE TRUE
## 47: 90 TRUE TRUE
## 48: 93 TRUE FALSE
## 49: 99 TRUE FALSE
## 50: 101 TRUE TRUE
## 51: 103 TRUE TRUE
## 52: 104 TRUE TRUE
## 53: 108 TRUE TRUE
## 54: 129 TRUE TRUE
## 55: 140 TRUE TRUE
## 56: 141 TRUE TRUE
## 57: 142 TRUE FALSE
## 58: 152 TRUE FALSE
## 59: 163 TRUE TRUE
## 60: 164 TRUE FALSE
## 61: 165 TRUE TRUE
## 62: 170 TRUE TRUE
## 63: 194 TRUE FALSE
## 64: 196 TRUE TRUE
## 65: 198 TRUE TRUE
## row_ids truth response#This learner's model can then be applied to other data
newPreds = classifLrn$predict_newdata(npData)
newPreds## <PredictionClassif> for 198 observations:
## row_ids truth response
## 1 FALSE FALSE
## 2 FALSE FALSE
## 3 TRUE TRUE
## ---
## 196 TRUE TRUE
## 197 TRUE TRUE
## 198 TRUE TRUEautoplot(newPreds)
#Evaluating Accuracy of MLR3 Model
#First we create the measure
msrAcc = msr("classif.acc")
msrAUC = msr("classif.auc")
msrFP = msr("classif.fp")
msrFPR = msr("classif.fpr")
msrFN = msr("classif.fn")
msrFNR = msr("classif.fnr")
measures = c(msrAcc,msrAUC,msrFP,msrFPR,msrFN,msrFNR)
predsClass$score(measures)## classif.acc classif.auc classif.fp classif.fpr classif.fn classif.fnr
## 0.6461538 NaN 7.0000000 0.2413793 16.0000000 0.4444444predsClass$confusion## truth
## response TRUE FALSE
## TRUE 20 7
## FALSE 16 22#Thresholding Can adjust the threshold of log odds at which it would be predicted as a win as below. Recall that if the log odds are above .5, it will be predicted a win. Sometimes you may want to adjust this for accuracy.
#predsClass$set_threshold(??)
predsClass$confusion## truth
## response TRUE FALSE
## TRUE 20 7
## FALSE 16 22#kNN
npSubset = subset(npData, select = c("win","gold_per_min","net_worth","duration"))
taskNP = as_task_classif(npSubset, target = "win", id = "wins", positive = "TRUE")
classifLrn <- lrn("classif.rpart")
#Partition our Data (default uses 33/67)
splits = partition(taskNP)
splits## $train
## [1] 1 2 4 7 16 18 20 23 25 35 39 40 43 46 47 51 60 61
## [19] 63 66 67 70 71 73 75 77 87 88 94 95 96 98 107 110 112 116
## [37] 126 131 135 136 138 143 148 149 150 153 154 156 157 159 168 172 179 181
## [55] 186 187 190 193 195 5 6 11 15 21 24 26 27 28 29 32 33 38
## [73] 44 50 52 54 55 58 59 69 72 76 85 90 91 92 93 97 100 101
## [91] 102 103 104 105 106 118 120 121 122 125 128 129 132 133 134 137 140 141
## [109] 142 145 146 147 151 152 155 158 160 163 165 170 173 175 178 180 182 183
## [127] 188 191 192 194 196 197 198
##
## $test
## [1] 8 14 30 41 42 45 49 53 64 68 74 80 81 83 111 113 114 115 117
## [20] 127 130 139 162 166 169 176 177 184 189 3 9 10 12 13 17 19 22 31
## [39] 34 36 37 48 56 57 62 65 78 79 82 84 86 89 99 108 109 119 123
## [58] 124 144 161 164 167 171 174 185classifLrn$train(taskNP, splits$train)
classifLrn$model## n= 133
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 133 59 TRUE (0.5563910 0.4436090)
## 2) gold_per_min>=592.5 86 28 TRUE (0.6744186 0.3255814)
## 4) duration< 2090 46 5 TRUE (0.8913043 0.1086957) *
## 5) duration>=2090 40 17 FALSE (0.4250000 0.5750000)
## 10) gold_per_min>=659.5 30 14 TRUE (0.5333333 0.4666667)
## 20) duration< 2845.5 23 9 TRUE (0.6086957 0.3913043)
## 40) net_worth>=26296.5 15 4 TRUE (0.7333333 0.2666667) *
## 41) net_worth< 26296.5 8 3 FALSE (0.3750000 0.6250000) *
## 21) duration>=2845.5 7 2 FALSE (0.2857143 0.7142857) *
## 11) gold_per_min< 659.5 10 1 FALSE (0.1000000 0.9000000) *
## 3) gold_per_min< 592.5 47 16 FALSE (0.3404255 0.6595745)
## 6) duration< 1227.5 7 0 TRUE (1.0000000 0.0000000) *
## 7) duration>=1227.5 40 9 FALSE (0.2250000 0.7750000)
## 14) net_worth>=13239.5 22 7 FALSE (0.3181818 0.6818182)
## 28) duration< 1745.5 7 2 TRUE (0.7142857 0.2857143) *
## 29) duration>=1745.5 15 2 FALSE (0.1333333 0.8666667) *
## 15) net_worth< 13239.5 18 2 FALSE (0.1111111 0.8888889) *#Predicting data on our testing split
predsClass = classifLrn$predict(taskNP,splits$test)
predsClass## <PredictionClassif> for 65 observations:
## row_ids truth response
## 8 FALSE FALSE
## 14 FALSE FALSE
## 30 FALSE FALSE
## ---
## 171 TRUE TRUE
## 174 TRUE FALSE
## 185 TRUE TRUEmsrAcc = msr("classif.acc")
msrAUC = msr("classif.auc")
msrFP = msr("classif.fp")
msrFPR = msr("classif.fpr")
msrFN = msr("classif.fn")
msrFNR = msr("classif.fnr")
measures = c(msrAcc,msrAUC,msrFP,msrFPR,msrFN,msrFNR)
predsClass$score(measures)## classif.acc classif.auc classif.fp classif.fpr classif.fn classif.fnr
## 0.6923077 NaN 7.0000000 0.2413793 13.0000000 0.3611111predsClass$confusion## truth
## response TRUE FALSE
## TRUE 23 7
## FALSE 13 22#######
knnLrn <- lrn("classif.kknn")
taskNPKnn = as_task_classif(npSubset, target = "win", id = "wins", positive = "TRUE", backend =npSubset)
splitsKNN = partition(taskNPKnn)
knnLrn$train(taskNPKnn, splitsKNN$train)
knnLrn$model## $formula
## win ~ .
## NULL
##
## $data
## win duration gold_per_min net_worth
## 1: FALSE 1754 569 15770
## 2: FALSE 1794 549 15317
## 3: FALSE 1846 442 11580
## 4: FALSE 1315 434 9842
## 5: FALSE 3349 902 41321
## ---
## 129: TRUE 1851 659 19467
## 130: TRUE 1278 792 16988
## 131: TRUE 1636 518 13922
## 132: TRUE 2577 838 30875
## 133: TRUE 1214 567 11645
##
## $pv
## $pv$k
## [1] 7
##
##
## $kknn
## NULLknnPreds <- knnLrn$predict(taskNPKnn,splits$test)
#Measuring Accuracy
confMat <- knnPreds$confusion
confMat## truth
## response TRUE FALSE
## TRUE 27 3
## FALSE 9 26#Creating the measures
msrAcc = msr("classif.acc")
msrFPR = msr("classif.fpr")
msrFNR = msr("classif.fnr")
measures = c(msrAcc,msrFPR,msrFNR)
knnPreds$score(measures)## classif.acc classif.fpr classif.fnr
## 0.8153846 0.1034483 0.2500000#kNN Caret
knnNP <- train(win ~ ., data = dataTrain, method = "knn",
trControl = trainControl(method = "cv", number = 10),
tuneGrid = expand.grid(k = 1:10)
)
knnPreds <- predict(knnNP, newdata = dataTest)
knnAcc <- confusionMatrix(knnPreds, dataTest$win)
knnAcc$table## Reference
## Prediction FALSE TRUE
## FALSE 9 11
## TRUE 13 16kNN Caret F1
truePositives <- knnAcc$table[2, 2]
falsePositives <- knnAcc$table[2, 1]
falseNegatives <- knnAcc$table[1, 2]
precision <- truePositives / (truePositives + falsePositives)
recall <- truePositives / (truePositives + falseNegatives)
F1_score <- 2 * (precision * recall) / (precision + recall)Prediction Page Using the GLM
gold_per_min,net_worth,gold,kills,tower_damage,duration,lane, lane_role
dataModel
predictions <- predict(glmFit, data = c(input\(gpm,input\)networth,input\(gold,input\)kills,input\(tower_damage,input\)duration,input\(lane,input\)lane_role))
dataIndex <- createDataPartition(dataModel$win, p = 0.75, list = FALSE)
dataTrain <- dataModel[dataIndex, ]
dataTest <- dataModel[-dataIndex, ]
glmFit <- train(win ~ ., data = dataTrain,
method = "glm",
family = "binomial",
preProcess = c("center", "scale"),
trControl = trainControl(method = "cv", number = 10))
#Change to be reactive to user data
userData <- data.frame(gold_per_min=600,net_worth=17000,gold=400,kills=11,tower_damage=9000,duration=2500,lane=3,lane_role=3)
userPred<-predict(glmFit, newdata= userData)
userPred## [1] TRUE
## Levels: FALSE TRUE