## 분석 목적 : Deep Learning 을 이용한 무인 자동차 핵심 기술별 논문 분류 분석
## 본 Page는 R을 이용한 Deep Learning 연습용 페이지입니다.
## 위 분석 중 일부의 분석을 가공하여 공개합니다.
## 본 Page 설명
## Data : 핵심 기술별 특허 정보 한국 DB from WIPS
## 분석 : 특허 DB의 초록을 Deep Learning을 이용하여 텍스트 기반 특허 기술 분류 모형 학습
## 실제 최종 분석에는 특허 DB 중 미국 DB를 사용하였습니다.
## 한국 DATA는 기술 별 수가 적어, 결과 해석에 주의할 필요가 있습니다.
## 배포 및 수정, 재배포를 삼가주세요.
## 2015.04.28 by 김형준(Hyung-jun, Kim)
## soeque1@gmail.com
## http://soeque1.github.io/r_slide/
rm(list=ls())
save_dir <- "/Users/kimhyungjun/Dropbox/h2o/prac/"
options(repos='http://cran.nexr.com')
install_lib <- function(x){
for( i in x ){
# require returns TRUE invisibly if it was able to load package
if( ! require( i , character.only = TRUE ) ){
# If package was not able to be loaded then re-install
install.packages( i , dependencies = TRUE )
# Load package after installing
library( i , character.only = T)
} else {
library( i , character.only = T)
}
}
}
suppressMessages(install_lib(c("readxl",
"dplyr", "stringr",
"tm", "lsa",
"KoNLP",
"h2o")))
save_dir <- "/Users/kimhyungjun/Dropbox/h2o/prac/"
data_list <- list.files(paste(save_dir,"data/wips",sep=""))
data_list_kr <- data_list[grep("kr",data_list)]
data <- data.frame()
for (i in 1:length(data_list_kr))
{
data_temp <- read_excel(paste(save_dir,"data/wips/",data_list_kr[i],sep=""))
data_temp <- cbind(rep(substr(data_list_kr[i],1,1),nrow(data_temp)),data_temp)
colnames(data_temp)[1] <- "기술"
data <- rbind(data,data_temp)
rm(data_temp)
}
dt_kr <- data[,c("기술", "Original IPC Main", "출원일","발명의 명칭", "요약", "국가코드")]
dim(dt_kr)
[1] 126 6
dt_text <- dt_kr %>% dplyr::select(요약) %>% .[[1]]
DT <- sapply(dt_text, extractNoun, USE.NAMES = F) %>%
sapply(function(x) paste(x, collapse = ' ')) %>%
as.data.frame
tdm <- TermDocumentMatrix(Corpus(DataframeSource(DT)),
control = list(
removeNumbers = TRUE,
wordLengths = c(2,Inf),
removePunctuation = TRUE,
weighting = function(x)
weightSMART(x, spec = "nnn")))
tdm <- as.matrix(tdm)
print(tdm[1:5,1:5])
Docs Terms 1 2 3 4 5 가능 0 0 1 0 0 가능성 0 0 0 0 0 가능하다동일 0 0 0 0 0 가변 0 0 0 0 0 가변하는 0 0 0 0 0
tdm <- lw_logtf(tdm) * gw_entropy(tdm)
print(tdm[1:5,1:5])
Docs Terms 1 2 3 4 5 가능 0 0 1.126579 0 0 가능성 0 0 0.000000 0 0 가능하다동일 0 0 0.000000 0 0 가변 0 0 0.000000 0 0 가변하는 0 0 0.000000 0 0
rm(DT)
actual_y <- dt_kr%>%dplyr::select(기술)%>%.[[1]]
save_data <- data.frame(cbind(t(tdm),actual_y))
write.table(save_data, paste(save_dir,'data/patent_kr.csv', sep=""),
row.names=F, col.names=F)
## Deep Learning
h2oServer <- h2o.init(nthreads=-1, max_mem_size = "6g")
Successfully connected to http://127.0.0.1:54321
R is connected to H2O cluster:
H2O cluster uptime: 1 hours 2 minutes
H2O cluster version: 2.8.4.4
H2O cluster name: H2O_started_from_R
H2O cluster total nodes: 1
H2O cluster total memory: 5.33 GB
H2O cluster total cores: 4
H2O cluster allowed cores: 4
H2O cluster healthy: TRUE
data_hex <- h2o.importFile(h2oServer, path = paste(save_dir,"data/patent_kr.csv", sep=""))
|======================================================================| 100%
random <- h2o.runif(data_hex, seed = 654321)
train.hex <- h2o.assign(data_hex[random <= .8,], "train.hex")
test.hex <- h2o.assign(data_hex[random > .8,], "test.hex")
label_t <- test.hex %>% as.data.frame %>% select(ncol(test.hex)) %>% table
label_t
test.hex %>% as.data.frame %>% select(ncol(test.hex)) A B C E H 4 2 23 2 1
label_t[label_t == max(label_t)] / sum(label_t)
C 0.71875
my.dl <- h2o.deeplearning(x = 1:(ncol(train.hex)-1), y = ncol(train.hex), data=train.hex, validation=test.hex,
variable_importances=T,
activation = "RectifierWithDropout",
input_dropout_ratio = 0.25,
hidden_dropout_ratios = c(0.5,0.5,0.5),
adaptive_rate = T,
balance_classes = T,
train_samples_per_iteration = 1500,
hidden = c(250,250,250),
epochs = 15)
|======================================================================| 100%
my.dl
IP Address: 127.0.0.1
Port : 54321
Parsed Data Key: train.hex
Deep Learning Model Key: DeepLearning_b1d4668517cab735bd3788d3a204df6
Training classification error: 0.01542416
Validation classification error: 0.15625
Confusion matrix:
Reported on test.hex
Predicted
Actual A B C D E G H Error
A 2 0 1 0 1 0 0 0.50000
B 1 1 0 0 0 0 0 0.50000
C 0 0 23 0 0 0 0 0.00000
D 0 0 0 0 0 0 0 NaN
E 1 0 0 0 1 0 0 0.50000
G 0 0 0 0 0 0 0 NaN
H 1 0 0 0 0 0 0 1.00000
Totals 5 1 24 0 2 0 0 0.15625
Hit Ratios for Multi-class Classification:
k hit_ratios
1 1 0.84375
2 2 0.84375
3 3 0.87500
4 4 0.96875
5 5 1.00000
6 6 1.00000
7 7 1.00000
Relative Variable Importance:
C966 C1129 C590 C326 C1002 C496 C686 C64
1 1 0.8975929 0.8479354 0.846368 0.8334404 0.8257619 0.821659 0.8195269
C1307 C509 C618 C922 C388 C1032 C399
1 0.8156956 0.8146315 0.8145919 0.8120914 0.8095592 0.8094196 0.8089094
C1130 C740 C1217 C1105 C1257 C467 C1025
1 0.8085309 0.8078029 0.8077086 0.8062583 0.8059309 0.8057524 0.8045306
C271 C135 C111 C42 C947 C954 C199
1 0.8035234 0.8022107 0.7999194 0.7996116 0.7982485 0.7972937 0.7971659
C319 C497 C517 C1194 C534 C905 C958 C1096
1 0.7971356 0.79634 0.796011 0.7956779 0.7952835 0.7949308 0.7945616 0.7931651
C265 C1023 C1174 C33 C369 C884 C1276
1 0.7927092 0.7920221 0.7913429 0.7905443 0.7903202 0.7902501 0.7901686
C588 C1005 C1326 C1065 C365 C463 C802
1 0.7892069 0.7882137 0.7879453 0.7878744 0.7875574 0.7873529 0.7871828
C362 C376 C769 C1067 C546 C735 C622 C301
1 0.7857954 0.7856193 0.78552 0.7852402 0.7847852 0.7847664 0.7835196 0.7833849
C829 C115 C1195 C744 C970 C1223 C1293
1 0.7828824 0.7826813 0.7823161 0.7822815 0.7822725 0.7822048 0.7821126
C828 C430 C344 C37 C1154 C290 C140
1 0.7816629 0.7814871 0.7814563 0.7814389 0.7809651 0.7809343 0.7798126
C702 C1116 C1298 C907 C426 C628 C725
1 0.7796668 0.7795824 0.7792722 0.7791799 0.7791016 0.7789949 0.7789022
C1134 C383 C572 C996 C474 C1314 C471
1 0.7784212 0.7784176 0.7783199 0.7782393 0.7781489 0.7780302 0.7779759
C291 C1224 C823 C1173 C929 C1331 C670
1 0.7777364 0.7776245 0.7773763 0.7772556 0.7771392 0.7769874 0.7765389
C1074 C1121 C765 C84 C127 C594 C770
1 0.7760179 0.7759684 0.7754998 0.7748432 0.7748062 0.7747998 0.774744
C323 C487 C1271 C334 C387 C293 C2
1 0.7744636 0.7744355 0.7741257 0.7734792 0.7731457 0.7729978 0.7729579
C730 C1259 C1080 C345 C407 C87 C847
1 0.7724191 0.7723939 0.7723885 0.7723081 0.7721801 0.7720025 0.7718574
C356 C1034 C894 C745 C143 C977 C840
1 0.7713206 0.7711317 0.7711133 0.7703581 0.7702123 0.7701865 0.7698066
C1016 C794 C1296 C891 C835 C1141 C1047 C1272
1 0.7697496 0.7697167 0.7694979 0.7694064 0.7690371 0.7690339 0.76896 0.7689471
C909 C116 C1160 C482 C714 C1013 C232
1 0.7689156 0.7688924 0.7688897 0.7686064 0.7685998 0.767898 0.7675303
C734 C1055 C796 C134 C359 C1052 C1020
1 0.7674907 0.7670408 0.7668033 0.7663577 0.7663179 0.7662599 0.7660069
C81 C825 C1309 C842 C309 C221 C928
1 0.7659751 0.7659576 0.7658365 0.7657751 0.7657126 0.7656459 0.7655602
C1012 C24 C215 C133 C1069 C706 C935 C1198
1 0.7654943 0.7653605 0.7653115 0.765112 0.7649566 0.7648462 0.764743 0.7647045
C55 C52 C1035 C1339 C375 C1006 C107
1 0.7646913 0.7645694 0.7642381 0.764168 0.7641654 0.7638252 0.7637324
C67 C473 C410 C1327 C773 C984 C1205 C921
1 0.7636613 0.7635288 0.7632498 0.7629954 0.7629081 0.76278 0.7627078 0.7626934
C855 C677 C522 C138 C630 C1254 C357
1 0.7626634 0.7626478 0.7623096 0.7622596 0.7622589 0.7618416 0.7616351
C812 C1022 C421 C94 C1114 C480 C1297
1 0.7615276 0.7614833 0.7611583 0.761064 0.7610591 0.7609112 0.7607001
C885 C687 C533 C1136 C151 C14 C1010
1 0.7606875 0.7601503 0.7600196 0.7600086 0.7598212 0.7595975 0.7594127
C41 C1021 C1079 C614 C685 C395 C1182
1 0.7594103 0.7594041 0.7593878 0.7592459 0.7592105 0.7592056 0.7591121
C1249 C1315 C715 C900 C1027 C1142 C1189
1 0.7587667 0.7587557 0.7584937 0.7584094 0.758297 0.7582086 0.7580647
C914 C523 C122 C672 C461 C1329 C203
1 0.7579346 0.7578264 0.7577965 0.7577749 0.7576365 0.7574875 0.7574754
C666 C1292 C844 C705 C181 C455 C913
1 0.7573696 0.7572414 0.7571458 0.7569963 0.7568834 0.7568464 0.756492
C61 C165 C50 C694 C1294 C137 C780
1 0.7563937 0.7562829 0.7562515 0.7559683 0.7557278 0.7557101 0.7556475
C1239 C435 C20 C1336 C1235 C485 C1113
1 0.7556031 0.7554152 0.7553359 0.7552998 0.7552724 0.7551503 0.7551097
C650 C394 C279 C234 C521 C852 C554
1 0.7550901 0.7549536 0.7549089 0.7548334 0.754686 0.7546334 0.7543668
C157 C749 C611 C1072 C417 C254 C128
1 0.7543433 0.7541392 0.7540181 0.7539256 0.7539119 0.7539011 0.7538765
C305 C826 C1054 C1246 C576 C846 C633
1 0.7538201 0.7537107 0.7537042 0.7536645 0.7536554 0.7535312 0.7533575
C1031 C736 C889 C610 C36 C989 C1001
1 0.7532858 0.7532716 0.7532635 0.7531562 0.7527947 0.7527768 0.7526592
C1081 C29 C1221 C1150 C378 C296 C131
1 0.7526432 0.7526221 0.7524357 0.7522174 0.7521967 0.7519562 0.7519069
C350 C727 C207 C1318 C916 C1026 C940
1 0.7518244 0.7516847 0.7514795 0.7514294 0.751395 0.7513676 0.7512986
C807 C105 C488 C447 C1091 C1190 C915
1 0.7512678 0.7511446 0.7509872 0.7509811 0.7508925 0.7507546 0.7506416
C321 C240 C1050 C90 C324 C903 C1218
1 0.7505885 0.7505074 0.7504052 0.7503611 0.7502369 0.7501564 0.7499986
C1168 C43 C374 C848 C371 C578 C404
1 0.7499676 0.7499409 0.7498293 0.749774 0.7496913 0.7495782 0.7495402
C1028 C713 C563 C1311 C830 C728 C269
1 0.7495103 0.7494926 0.749487 0.7494801 0.7494037 0.7493975 0.7493664
C668 C446 C819 C144 C196 C104 C587
1 0.7488936 0.748737 0.7484987 0.7484856 0.7484807 0.7484469 0.7483365
C729 C272 C208 C986 C1186 C760 C580
1 0.7483054 0.7481934 0.7481154 0.7481051 0.7478497 0.7477873 0.7475753
C346 C662 C856 C717 C448 C136 C1222
1 0.747523 0.7474669 0.7474087 0.7473926 0.7473556 0.7473185 0.7471813
C689 C1178 C742 C206 C883 C1125 C1062
1 0.7471206 0.7469949 0.7469006 0.7467601 0.746563 0.7465131 0.7464874
C283 C1163 C1039 C567 C92 C1206 C544
1 0.7463614 0.7463488 0.7462604 0.7462167 0.7462034 0.7460668 0.7458935
C1011 C1131 C758 C236 C616 C1058 C408
1 0.7458351 0.7457771 0.7456776 0.7456732 0.7456213 0.7454973 0.7454961
C385 C1043 C179 C955 C813 C857 C145
1 0.7454814 0.7453001 0.7451011 0.7450859 0.7449064 0.7448703 0.7448248
C753 C843 C739 C1036 C1341 C1090 C890 C259
1 0.7447634 0.74475 0.7447041 0.7446395 0.744613 0.7445288 0.7444885 0.7444531
C1044 C464 C527 C267 C398 C329 C581 C861
1 0.7443773 0.7442853 0.74422 0.7441939 0.7441821 0.7440794 0.7440608 0.7439879
C1045 C333 C505 C257 C386 C1145 C89 C316
1 0.7439396 0.7438539 0.7436836 0.74361 0.7435558 0.743533 0.7435106 0.7433932
C601 C512 C542 C936 C458 C339 C816
1 0.7433888 0.7433121 0.743299 0.7432885 0.7432319 0.7430946 0.7428714
C95 C893 C1323 C146 C681 C700 C1004 C639
1 0.7428358 0.7428229 0.742514 0.7424703 0.742461 0.7424514 0.7424095 0.7423982
C997 C99 C552 C304 C432 C793 C332
1 0.7423424 0.7423131 0.7421758 0.7421699 0.7420009 0.7419685 0.7419454
C420 C615 C172 C918 C320 C470 C63 C1082
1 0.7418606 0.7418558 0.7415906 0.7415719 0.741483 0.741244 0.7412315 0.7412179
C1092 C1172 C1345 C1197 C462 C1165 C1322
1 0.7410469 0.740999 0.7408158 0.7408131 0.7408046 0.7407908 0.7407196
C906 C575 C337 C555 C370 C800 C253 C476
1 0.7407006 0.7406963 0.7404777 0.740415 0.740382 0.7403559 0.7403405 0.7400334
C54 C506 C845 C183 C431 C1071 C434
1 0.7399247 0.7398568 0.7397155 0.739683 0.7394299 0.7393245 0.7392812
C25 C1000 C1184 C1304 C612 C406 C1015
1 0.7392372 0.7390486 0.7388765 0.7388513 0.7388461 0.7387329 0.7386037
C1210 C1287 C35 C814 C295 C4 C1302
1 0.7385069 0.7384486 0.7384242 0.7383907 0.7383103 0.7381767 0.7381145
C294 C427 C479 C1120 C310 C787 C27
1 0.7379762 0.7379045 0.7377451 0.7377428 0.7376915 0.737655 0.7375934
C874 C680 C556 C106 C591 C557 C1179
1 0.7374082 0.7373557 0.737317 0.7372713 0.7372595 0.7372367 0.7372044
C48 C440 C1237 C1305 C441 C632 C876
1 0.7369888 0.7369506 0.7369368 0.7369273 0.7367464 0.7366913 0.736545
C472 C1338 C454 C1009 C879 C31 C520
1 0.7364756 0.7364675 0.7364429 0.7364427 0.7363729 0.7363645 0.7363268
C1207 C962 C246 C676 C1075 C289 C963
1 0.7363051 0.7362438 0.736175 0.7361508 0.7361286 0.7360806 0.7360426
C185 C1057 C465 C948 C1187 C311 C75
1 0.7360024 0.7359349 0.7358131 0.7358015 0.7357221 0.7356907 0.7355947
C652 C312 C892 C260 C239 C709 C288
1 0.7354209 0.7353572 0.7352065 0.7351296 0.7350991 0.735075 0.7350131
C466 C302 C949 C771 C274 C1146 C841
1 0.7349718 0.7349485 0.734744 0.7345638 0.7344201 0.7343992 0.7343854
C1203 C721 C1029 C817 C1283 C862 C661 C911
1 0.7343469 0.7342613 0.734135 0.7341335 0.734081 0.7340664 0.7340214 0.7340019
C1101 C561 C79 C1332 C1215 C451 C565 C1280
1 0.7339368 0.733878 0.7338329 0.7336548 0.7336519 0.733615 0.7335549 0.7334427
C1244 C56 C193 C508 C860 C411 C1007
1 0.7334313 0.7333252 0.7332497 0.7330612 0.7330523 0.7329103 0.7328154
C1196 C895 C397 C988 C737 C16 C297 C978
1 0.7327754 0.7326934 0.7326428 0.7326143 0.7325304 0.7325301 0.7325003 0.73223
C985 C851 C530 C189 C62 C192 C638 C423
1 0.7318763 0.7318415 0.7316546 0.7316399 0.731589 0.7315839 0.731581 0.7315204
C453 C202 C97 C859 C10 C888 C761
1 0.7313812 0.7312518 0.7311867 0.7311819 0.7310823 0.7310578 0.7310466
C1328 C640 C1192 C1260 C1213 C822 C139
1 0.7309218 0.7308531 0.7307048 0.7306598 0.7305744 0.7305577 0.7305316
C945 C1078 C946 C646 C1126 C262 C1041
1 0.7305005 0.7304934 0.730439 0.7304055 0.7303486 0.7303273 0.7302767
C704 C1299 C1231 C1330 C166 C722 C930
1 0.7302715 0.7302477 0.7302068 0.7301922 0.7300835 0.7299992 0.7299532
C194 C368 C102 C1306 C912 C317 C526
1 0.7297704 0.7296972 0.7296708 0.7295907 0.7295895 0.7295792 0.7295696
C1188 C969 C1127 C1251 C1175 C1171 C537 C854
1 0.7294126 0.7293707 0.7293646 0.729287 0.729272 0.7291312 0.7290774 0.7289547
C560 C1286 C110 C1076 C1112 C1132 C1137 C278
1 0.7289056 0.7288549 0.728703 0.7286866 0.7286117 0.728604 0.7285699 0.7284995
C245 C201 C490 C690 C536 C1284 C1335 C460
1 0.728353 0.7283059 0.728281 0.7282014 0.7281757 0.7281398 0.7280523 0.7280511
C821 C792 C636 C241 C114 C991 C1344
1 0.7280304 0.7279805 0.7279483 0.7277631 0.7277523 0.727735 0.7277259
C1111 C983 C518 C1270 C968 C227 C524
1 0.7276319 0.7274838 0.727406 0.7272971 0.7271523 0.7270928 0.7270865
C91 C1140 C1149 C1084 C188 C1083 C973
1 0.7270833 0.7270638 0.7268338 0.7267823 0.7267478 0.7267203 0.7266377
C1099 C938 C276 C1204 C515 C168 C1085
1 0.7265508 0.7264624 0.7263944 0.7263526 0.7262682 0.726178 0.7261704
C547 C777 C51 C352 C990 C452 C266
1 0.7260259 0.7258962 0.7258841 0.7258764 0.7257587 0.7257494 0.7257239
C1110 C732 C77 C1153 C703 C498 C1048 C130
1 0.7255935 0.7255872 0.7255705 0.7254947 0.725378 0.7252613 0.7252464 0.725246
C1320 C901 C209 C1264 C1 C45 C643
1 0.7252022 0.7251865 0.7250204 0.7249654 0.724896 0.7248847 0.7248449
C49 C173 C824 C70 C149 C1252 C1266
1 0.7248425 0.7247788 0.7247357 0.7247025 0.7246799 0.7246764 0.7246438
C235 C788 C865 C707 C898 C1030 C1324
1 0.7246324 0.7246209 0.7246115 0.7245578 0.724535 0.7244304 0.7242343
C1042 C1159 C180 C1040 C693 C327 C1321
1 0.7239729 0.7238959 0.7238867 0.7233977 0.723367 0.7233623 0.7233247
C281 C1093 C853 C784 C992 C868 C818
1 0.7231582 0.7231197 0.7230434 0.7229472 0.7229041 0.7228208 0.7227647
C1295 C1046 C1301 C597 C229 C1214 C917
1 0.7227563 0.7227126 0.7227007 0.7226906 0.7226747 0.7226316 0.7225092
C1242 C867 C1277 C342 C872 C1097 C964 C1063
1 0.7224655 0.7224415 0.7222546 0.7221573 0.722146 0.722094 0.7219863 0.7219596
C934 C726 C325 C69 C1337 C927 C176
1 0.7218661 0.7218481 0.7218267 0.7217745 0.721683 0.7216432 0.7216396
C535 C519 C212 C1258 C258 C401 C999
1 0.7214982 0.7213431 0.7213139 0.7212271 0.7211434 0.7211424 0.7211385
C609 C1226 C161 C103 C264 C993 C1232
1 0.7209745 0.7209169 0.7208147 0.7207948 0.7207845 0.7207643 0.7207557
C604 C47 C1247 C413 C804 C256 C228
1 0.7207462 0.7207387 0.7207313 0.7206826 0.7204869 0.7204686 0.7204651
C340 C255 C529 C125 C965 C422 C887
1 0.7204415 0.7202761 0.7202601 0.7202058 0.7201258 0.720008 0.7198415
C790 C216 C489 C1290 C88 C223 C11
1 0.7198257 0.7198125 0.7197592 0.7197108 0.7197048 0.7195349 0.7194908
C1281 C459 C230 C389 C513 C762 C1157
1 0.7194812 0.7194719 0.7194481 0.7194405 0.7194187 0.7193191 0.7192875
C658 C1180 C827 C108 C347 C1278 C213
1 0.7188737 0.7187835 0.7187383 0.7187253 0.7186469 0.7185432 0.7185239
C66 C763 C1170 C994 C602 C226 C925
1 0.7185237 0.7184799 0.7183663 0.7183149 0.7183145 0.7183065 0.7182727
C390 C647 C292 C1191 C328 C478 C785
1 0.7182472 0.7182034 0.7181351 0.7181304 0.7181035 0.7180343 0.7179406
C1164 C871 C869 C723 C952 C1267 C156
1 0.7178933 0.7178248 0.7177798 0.7177551 0.717751 0.7176986 0.7176813
C218 C1008 C910 C129 C83 C23 C445
1 0.7175977 0.7175719 0.7174172 0.7171728 0.7171105 0.7170805 0.7170284
C486 C607 C834 C864 C897 C1056 C718
1 0.7170102 0.7169808 0.7166972 0.7164092 0.7163653 0.7163365 0.7163192
C1139 C412 C1274 C920 C637 C225 C775 C808
1 0.7161423 0.716037 0.7159036 0.7157712 0.715755 0.7152205 0.7152143 0.7150871
C562 C820 C338 C908 C998 C155 C810
1 0.715073 0.7150667 0.7149462 0.7149453 0.7148497 0.7148433 0.7147877
C1185 C499 C664 C475 C237 C416 C381
1 0.7146954 0.7145163 0.714487 0.7144051 0.7142384 0.7142098 0.7142019
C158 C944 C1268 C981 C273 C331 C653
1 0.7141619 0.7141538 0.7140766 0.7140674 0.7139791 0.7137869 0.7137728
C1167 C280 C1166 C442 C396 C353 C1176
1 0.7137065 0.7136951 0.7136383 0.7136252 0.7135556 0.7135518 0.7135348
C882 C1156 C300 C469 C1201 C571 C1343 C669
1 0.713501 0.7134683 0.7133731 0.713302 0.7131832 0.7128011 0.7127619 0.7127088
C566 C655 C943 C214 C1162 C682 C1177
1 0.7126492 0.7125909 0.7125428 0.7124574 0.7123262 0.7122599 0.7121975
C164 C516 C501 C1094 C1225 C211 C772 C873
1 0.7121759 0.712017 0.711811 0.7116925 0.7116609 0.7116311 0.7115285 0.7111539
C3 C1250 C1333 C733 C961 C1123 C1233
1 0.7111214 0.7109287 0.7106769 0.7106579 0.7106286 0.7106099 0.7105436
C1135 C44 C550 C341 C645 C141 C583 C159
1 0.7105151 0.7104236 0.710413 0.7102728 0.710192 0.7101766 0.7100138 0.7099958
C1143 C960 C74 C778 C382 C1086 C167
1 0.7099392 0.7098044 0.7097037 0.7096261 0.7096105 0.709577 0.7094892
C355 C1109 C979 C698 C880 C875 C449
1 0.7094579 0.7093945 0.7092651 0.7092171 0.708971 0.7087178 0.7082565
C711 C789 C1325 C720 C1169 C148 C831
1 0.7081269 0.7081051 0.7077674 0.7076669 0.7072759 0.7072732 0.7070149
C924 C275 C858 C330 C532 C343 C424
1 0.7069317 0.706922 0.7069212 0.7068166 0.7067605 0.7067465 0.7067207
C716 C72 C1152 C1183 C153 C178 C1238
1 0.7066548 0.7065266 0.7065162 0.7061791 0.7058851 0.7058771 0.7056066
C58 C372 C781 C392 C1241 C987 C549
1 0.7055554 0.7054521 0.7054408 0.7052159 0.7050218 0.7049958 0.7049114
C608 C363 C1181 C678 C598 C774 C494
1 0.7048945 0.7046925 0.7045758 0.7045578 0.7044086 0.704217 0.7039853
C361 C750 C953 C832 C1342 C1253 C673 C1104
1 0.7039447 0.7038245 0.7037021 0.703631 0.703606 0.7035153 0.7034556 0.7033704
C1256 C623 C1227 C1103 C481 C200 C53
1 0.7033493 0.7033337 0.7033017 0.7033001 0.703262 0.7030302 0.7029667
C939 C1151 C811 C1211 C799 C656 C649
1 0.7029475 0.7028952 0.7028711 0.702709 0.7025641 0.7024982 0.7024234
C7 C1208 C282 C222 C170 C243 C150
1 0.7023728 0.7022805 0.7022539 0.7021303 0.7021204 0.7019117 0.7018275
C806 C175 C162 C198 C160 C1124 C1269
1 0.7016392 0.7016337 0.7014179 0.7013676 0.701191 0.7011765 0.7010739
C169 C249 C695 C1245 C1019 C982 C428
1 0.7010584 0.7009816 0.7007717 0.7005556 0.7003686 0.7003679 0.7003365
C1340 C642 C9 C429 C191 C1024 C231
1 0.7002858 0.7002758 0.7002059 0.7001151 0.7000586 0.6997445 0.699601
C629 C839 C746 C250 C1229 C86 C112 C456
1 0.6995186 0.699373 0.6992595 0.699205 0.6991216 0.6991202 0.6990664 0.6990303
C30 C298 C373 C675 C805 C1313 C187 C731
1 0.699007 0.6989816 0.6988905 0.6986963 0.698683 0.6986476 0.6985942 0.6984091
C699 C1220 C1308 C100 C195 C923 C59
1 0.6982429 0.6981468 0.6979891 0.6978096 0.6977614 0.6976922 0.6976057
C641 C621 C224 C657 C815 C1202 C660
1 0.6975431 0.6975341 0.6975082 0.6971343 0.6971101 0.6970732 0.6970658
C26 C1234 C468 C951 C665 C776 C118
1 0.6966581 0.6965873 0.6965725 0.6964759 0.6960695 0.696057 0.6957356
C1312 C995 C625 C438 C950 C1003 C380
1 0.6956233 0.6954372 0.6953285 0.6951095 0.6949871 0.6949301 0.6948329
C360 C152 C13 C942 C605 C620 C500
1 0.6948285 0.6947499 0.6946539 0.6944026 0.6941826 0.6941012 0.693973
C1064 C1087 C1077 C177 C1100 C1334 C277
1 0.6939357 0.6938794 0.6938523 0.6937278 0.6933864 0.6931501 0.6931373
C783 C1118 C1193 C551 C959 C303 C142
1 0.6930872 0.6930514 0.6928251 0.6926186 0.6925067 0.692349 0.6923158
C976 C425 C558 C197 C1147 C5 C606
1 0.6923003 0.6921511 0.6917399 0.6917113 0.6916144 0.6915811 0.6915093
C564 C541 C1230 C457 C691 C759 C896
1 0.6914742 0.6909213 0.6907975 0.6906986 0.690504 0.6904957 0.6904862
C32 C577 C654 C593 C1199 C510 C748
1 0.6903257 0.690174 0.6900639 0.6900535 0.6900114 0.6899878 0.6899688
C1346 C322 C15 C313 C252 C613 C634 C204
1 0.6898377 0.6895236 0.6892427 0.6891936 0.689148 0.688661 0.6885933 0.6881501
C1017 C743 C418 C46 C335 C377 C543
1 0.6881135 0.6880546 0.6879719 0.6879095 0.6879056 0.6876898 0.6875504
C957 C217 C393 C57 C450 C592 C1148 C174
1 0.6872115 0.687151 0.6871322 0.6870043 0.6868423 0.6868161 0.686631 0.6865577
C336 C307 C484 C553 C1289 C314 C754 C6
1 0.6862841 0.686172 0.6860717 0.6857949 0.685532 0.6853169 0.6852123 0.6851733
C803 C186 C1049 C1317 C741 C863 C491
1 0.6848819 0.6848219 0.6839699 0.6835398 0.6828781 0.6826866 0.6824793
C1122 C507 C1059 C1240 C626 C1138 C747
1 0.6823466 0.6822506 0.6815725 0.6815367 0.6813985 0.6803253 0.6799549
C21 C902 C1095 C931 C1088 C351 C559
1 0.6797686 0.6789448 0.6789384 0.6787121 0.678239 0.6778748 0.6775958
C1066 C1288 C437 C409 C1236 C1255 C71
1 0.6774592 0.6773075 0.6770167 0.6768748 0.6766678 0.6750192 0.6746039
C538 C569 C163 C248 C366 C932 C768 C1115
1 0.6742165 0.6734675 0.6726171 0.6716926 0.671208 0.671065 0.6705245 0.6704519
C1089 C502 C124 C967 C1282 C710 C40
1 0.6700866 0.6690288 0.6687425 0.6681562 0.667744 0.6676462 0.6672427
C833 C348 C242 C1038 C154 C132 C937
1 0.6663449 0.6649933 0.6638626 0.662448 0.6615145 0.6553058 0.6417201
my.dl@model[[5]]
1 - my.dl@model[[7]] ## ACC
Predicted Actual A B C D E G H Error A 2 0 1 0 1 0 0 0.50000 B 1 1 0 0 0 0 0 0.50000 C 0 0 23 0 0 0 0 0.00000 D 0 0 0 0 0 0 0 NaN E 1 0 0 0 1 0 0 0.50000 G 0 0 0 0 0 0 0 NaN H 1 0 0 0 0 0 0 1.00000 Totals 5 1 24 0 2 0 0 0.15625
[1] 0.84375
h2o.predict(my.dl, test.hex)%>%as.data.frame
predict A B C D E
1 A 8.910353e-01 7.370919e-02 4.151384e-04 2.001426e-02 1.469966e-02
2 C 6.575533e-02 4.006533e-03 6.922491e-01 1.994550e-03 2.357033e-01
3 A 9.916058e-01 3.679702e-05 4.273191e-03 1.494910e-03 2.100528e-03
4 E 8.663132e-04 2.727302e-04 1.563110e-02 2.535046e-05 9.809678e-01
5 B 6.141900e-03 8.492556e-01 7.596204e-02 8.158414e-04 6.774489e-02
6 A 4.749292e-01 3.193418e-02 3.677447e-01 6.163182e-02 5.019460e-02
7 C 1.811038e-01 3.085864e-06 8.183260e-01 2.069992e-05 5.096302e-04
8 C 5.756019e-03 2.522375e-03 9.888914e-01 1.087247e-03 1.734982e-03
9 C 5.930470e-02 3.130129e-05 9.375899e-01 3.312651e-04 2.671929e-03
10 C 1.825703e-03 4.023292e-06 9.747251e-01 1.472770e-05 2.341670e-02
11 C 1.510467e-01 9.470678e-06 8.479494e-01 2.856549e-04 6.926252e-04
12 C 8.396786e-04 9.822395e-07 9.987714e-01 2.709155e-06 3.833199e-04
13 C 1.825993e-02 7.534517e-07 9.756461e-01 1.462772e-04 5.910906e-03
14 C 1.339226e-02 4.830822e-05 9.745616e-01 5.765637e-05 1.158979e-02
15 C 1.749968e-03 3.559534e-08 9.951811e-01 2.040259e-05 2.941091e-03
16 C 4.439138e-03 8.681210e-05 9.923633e-01 5.276446e-04 2.578807e-03
17 C 1.426087e-03 1.415208e-03 9.861260e-01 3.098939e-03 7.889327e-03
18 C 2.582368e-02 6.033398e-06 9.713737e-01 7.237525e-05 2.713803e-03
19 C 1.280009e-01 1.309878e-06 8.604778e-01 1.401316e-05 1.134650e-02
20 C 4.581249e-04 3.253871e-06 9.955222e-01 1.269852e-06 4.000640e-03
21 C 3.556316e-01 4.930092e-04 5.866958e-01 5.074530e-02 6.051867e-03
22 C 4.196652e-03 2.281768e-03 9.871616e-01 4.027095e-04 5.698834e-03
23 C 7.468453e-04 8.635247e-06 9.991198e-01 2.139740e-06 1.183868e-04
24 C 1.859690e-03 1.281740e-04 9.570424e-01 4.032781e-02 4.311664e-04
25 C 3.659647e-04 2.134902e-04 9.760911e-01 5.589099e-06 2.302772e-02
26 C 8.512945e-04 4.381240e-06 9.776638e-01 1.415095e-05 2.142075e-02
27 C 4.565042e-04 2.823675e-06 9.994981e-01 7.607645e-06 2.595043e-05
28 C 3.962346e-01 4.569110e-06 5.495145e-01 1.112044e-03 4.062530e-02
29 C 2.624084e-03 3.717237e-06 9.965943e-01 2.394627e-04 5.298649e-04
30 E 2.945286e-06 1.138345e-07 3.548083e-06 4.594016e-08 9.999933e-01
31 A 9.566181e-01 1.836613e-04 3.475539e-02 5.484897e-03 2.308514e-03
32 A 9.819630e-01 3.784730e-08 1.782342e-02 1.398177e-04 9.965606e-06
G H
1 1.700104e-05 1.094636e-04
2 1.429311e-04 1.482188e-04
3 4.132097e-05 4.474512e-04
4 2.197825e-03 3.890582e-05
5 6.187477e-05 1.789151e-05
6 7.559906e-03 6.005612e-03
7 3.464324e-05 2.199473e-06
8 5.861047e-06 2.077966e-06
9 1.776990e-05 5.313415e-05
10 6.773403e-06 6.934447e-06
11 1.534041e-05 7.933132e-07
12 1.854687e-06 3.544071e-08
13 1.506946e-05 2.101103e-05
14 3.349420e-04 1.544023e-05
15 6.000408e-05 4.739714e-05
16 3.315862e-06 9.583165e-07
17 3.814938e-05 6.259146e-06
18 1.024018e-05 1.188076e-07
19 6.583014e-05 9.367479e-05
20 1.441582e-05 8.449995e-08
21 6.863641e-05 3.138040e-04
22 2.417184e-04 1.673150e-05
23 4.180110e-06 2.196789e-08
24 4.772300e-06 2.060103e-04
25 2.727209e-04 2.338081e-05
26 3.735673e-05 8.244993e-06
27 8.901210e-06 6.761185e-08
28 1.999702e-04 1.230907e-02
29 6.282980e-06 2.298442e-06
30 1.389598e-09 3.232724e-08
31 1.270388e-04 5.223532e-04
32 5.418841e-06 5.831452e-05
colnames(save_data)[str_replace_all(names(my.dl@model[[9]]), "C", "")%>%as.numeric%>%.[1:10]]
[1] "주차" "통신" "안전" "목적지" "지도" "센서" "운전자" "경고" [9] "lf" "속도"
my.rf <- h2o.randomForest(x = 1:(ncol(train.hex)-1), y = ncol(train.hex), data = train.hex, validation = test.hex,
type="fast",
importance=TRUE,
ntree=c(5),
depth=c(5,10))
|======================================================================| 100%
print(my.rf)
IP Address: 127.0.0.1
Port : 54321
Parsed Data Key: train.hex
Grid Search Model Key: GridSearch_854db4fb4e907d8477a4ddd9d52d429b
Summary
model_key ntrees max_depth nbins
1 SpeeDRF_8e1207d25a80c2ab78a9f42b87084dec 5 5 1024
2 SpeeDRF_a5396d0c751d639fd2bf68a9c0224c25 5 10 1024
prediction_error run_time
1 0.25 5893
2 0.34375 6311
my.rf@model[[1]]@model$confusion
Predicted Actual A B C D E G H Error A 1 0 2 0 1 0 0 0.75 B 0 0 2 0 0 0 0 1.00 C 0 0 23 0 0 0 0 0.00 D 0 0 0 0 0 0 0 NaN E 0 0 2 0 0 0 0 1.00 G 0 0 0 0 0 0 0 NaN H 0 0 1 0 0 0 0 1.00 Totals 1 0 30 0 1 0 0 0.25
print(1 - my.rf@sumtable[[1]]$prediction_error)
print(1 - my.rf@sumtable[[2]]$prediction_error)
[1] 0.75 [1] 0.65625
h2o.shutdown(h2oServer)
Are you sure you want to shutdown the H2O instance running at http://127.0.0.1:54321 (Y/N)?