H2O Workshop
Introduction to H2O
Load packages and data
Import datasets
## Parsed with column specification:
## cols(
## .default = col_double(),
## SPACEID = col_character()
## )## See spec(...) for full column specifications.##
## H2O is not running yet, starting it now...
##
## Note: In case of errors look at the following log files:
## C:\Users\gabri\AppData\Local\Temp\Rtmp4qbCXm/h2o_gabri_started_from_r.out
## C:\Users\gabri\AppData\Local\Temp\Rtmp4qbCXm/h2o_gabri_started_from_r.err
##
##
## Starting H2O JVM and connecting: Connection successful!
##
## R is connected to the H2O cluster:
## H2O cluster uptime: 5 seconds 395 milliseconds
## H2O cluster timezone: Europe/Paris
## H2O data parsing timezone: UTC
## H2O cluster version: 3.22.1.1
## H2O cluster version age: 6 months and 17 days !!!
## H2O cluster name: H2O_started_from_R_gabri_hoy297
## H2O cluster total nodes: 1
## H2O cluster total memory: 1.74 GB
## H2O cluster total cores: 8
## H2O cluster allowed cores: 8
## H2O cluster healthy: TRUE
## H2O Connection ip: localhost
## H2O Connection port: 54321
## H2O Connection proxy: NA
## H2O Internal Security: FALSE
## H2O API Extensions: Algos, AutoML, Core V3, Core V4
## R Version: R version 3.6.0 (2019-04-26)## Warning in h2o.clusterInfo():
## Your H2O cluster version is too old (6 months and 17 days)!
## Please download and install the latest version from http://h2o.ai/download/## Connection successful!
##
## R is connected to the H2O cluster:
## H2O cluster uptime: 5 seconds 600 milliseconds
## H2O cluster timezone: Europe/Paris
## H2O data parsing timezone: UTC
## H2O cluster version: 3.22.1.1
## H2O cluster version age: 6 months and 17 days !!!
## H2O cluster name: H2O_started_from_R_gabri_hoy297
## H2O cluster total nodes: 1
## H2O cluster total memory: 1.74 GB
## H2O cluster total cores: 8
## H2O cluster allowed cores: 8
## H2O cluster healthy: TRUE
## H2O Connection ip: localhost
## H2O Connection port: 54321
## H2O Connection proxy: NA
## H2O Internal Security: FALSE
## H2O API Extensions: Algos, AutoML, Core V3, Core V4
## R Version: R version 3.6.0 (2019-04-26)## Warning in h2o.clusterInfo():
## Your H2O cluster version is too old (6 months and 17 days)!
## Please download and install the latest version from http://h2o.ai/download/##
|
| | 0%
|
|=================================================================| 100%##
|
| | 0%
|
|=================================================================| 100%## [1] "WAP001" "WAP002" "WAP003"
## [4] "WAP004" "WAP005" "WAP006"
## [7] "WAP007" "WAP008" "WAP009"
## [10] "WAP010" "WAP011" "WAP012"
## [13] "WAP013" "WAP014" "WAP015"
## [16] "WAP016" "WAP017" "WAP018"
## [19] "WAP019" "WAP020" "WAP021"
## [22] "WAP022" "WAP023" "WAP024"
## [25] "WAP025" "WAP026" "WAP027"
## [28] "WAP028" "WAP029" "WAP030"
## [31] "WAP031" "WAP032" "WAP033"
## [34] "WAP034" "WAP035" "WAP036"
## [37] "WAP037" "WAP038" "WAP039"
## [40] "WAP040" "WAP041" "WAP042"
## [43] "WAP043" "WAP044" "WAP045"
## [46] "WAP046" "WAP047" "WAP048"
## [49] "WAP049" "WAP050" "WAP051"
## [52] "WAP052" "WAP053" "WAP054"
## [55] "WAP055" "WAP056" "WAP057"
## [58] "WAP058" "WAP059" "WAP060"
## [61] "WAP061" "WAP062" "WAP063"
## [64] "WAP064" "WAP065" "WAP066"
## [67] "WAP067" "WAP068" "WAP069"
## [70] "WAP070" "WAP071" "WAP072"
## [73] "WAP073" "WAP074" "WAP075"
## [76] "WAP076" "WAP077" "WAP078"
## [79] "WAP079" "WAP080" "WAP081"
## [82] "WAP082" "WAP083" "WAP084"
## [85] "WAP085" "WAP086" "WAP087"
## [88] "WAP088" "WAP089" "WAP090"
## [91] "WAP091" "WAP092" "WAP093"
## [94] "WAP094" "WAP095" "WAP096"
## [97] "WAP097" "WAP098" "WAP099"
## [100] "WAP100" "WAP101" "WAP102"
## [103] "WAP103" "WAP104" "WAP105"
## [106] "WAP106" "WAP107" "WAP108"
## [109] "WAP109" "WAP110" "WAP111"
## [112] "WAP112" "WAP113" "WAP114"
## [115] "WAP115" "WAP116" "WAP117"
## [118] "WAP118" "WAP119" "WAP120"
## [121] "WAP121" "WAP122" "WAP123"
## [124] "WAP124" "WAP125" "WAP126"
## [127] "WAP127" "WAP128" "WAP129"
## [130] "WAP130" "WAP131" "WAP132"
## [133] "WAP133" "WAP134" "WAP135"
## [136] "WAP136" "WAP137" "WAP138"
## [139] "WAP139" "WAP140" "WAP141"
## [142] "WAP142" "WAP143" "WAP144"
## [145] "WAP145" "WAP146" "WAP147"
## [148] "WAP148" "WAP149" "WAP150"
## [151] "WAP151" "WAP152" "WAP153"
## [154] "WAP154" "WAP155" "WAP156"
## [157] "WAP157" "WAP158" "WAP159"
## [160] "WAP160" "WAP161" "WAP162"
## [163] "WAP163" "WAP164" "WAP165"
## [166] "WAP166" "WAP167" "WAP168"
## [169] "WAP169" "WAP170" "WAP171"
## [172] "WAP172" "WAP173" "WAP174"
## [175] "WAP175" "WAP176" "WAP177"
## [178] "WAP178" "WAP179" "WAP180"
## [181] "WAP181" "WAP182" "WAP183"
## [184] "WAP184" "WAP185" "WAP186"
## [187] "WAP187" "WAP188" "WAP189"
## [190] "WAP190" "WAP191" "WAP192"
## [193] "WAP193" "WAP194" "WAP195"
## [196] "WAP196" "WAP197" "WAP198"
## [199] "WAP199" "WAP200" "WAP201"
## [202] "WAP202" "WAP203" "WAP204"
## [205] "WAP205" "WAP206" "WAP207"
## [208] "WAP208" "WAP209" "WAP210"
## [211] "WAP211" "WAP212" "WAP213"
## [214] "WAP214" "WAP215" "WAP216"
## [217] "WAP217" "WAP218" "WAP219"
## [220] "WAP220" "WAP221" "WAP222"
## [223] "WAP223" "WAP224" "WAP225"
## [226] "WAP226" "WAP227" "WAP228"
## [229] "WAP229" "WAP230" "WAP231"
## [232] "WAP232" "WAP233" "WAP234"
## [235] "WAP235" "WAP236" "WAP237"
## [238] "WAP238" "WAP239" "WAP240"
## [241] "WAP241" "WAP242" "WAP243"
## [244] "WAP244" "WAP245" "WAP246"
## [247] "WAP247" "WAP248" "WAP249"
## [250] "WAP250" "WAP251" "WAP252"
## [253] "WAP253" "WAP254" "WAP255"
## [256] "WAP256" "WAP257" "WAP258"
## [259] "WAP259" "WAP260" "WAP261"
## [262] "WAP262" "WAP263" "WAP264"
## [265] "WAP265" "WAP266" "WAP267"
## [268] "WAP268" "WAP269" "WAP270"
## [271] "WAP271" "WAP272" "WAP273"
## [274] "WAP274" "WAP275" "WAP276"
## [277] "WAP277" "WAP278" "WAP279"
## [280] "WAP280" "WAP281" "WAP282"
## [283] "WAP283" "WAP284" "WAP285"
## [286] "WAP286" "WAP287" "WAP288"
## [289] "WAP289" "WAP290" "WAP291"
## [292] "WAP292" "WAP293" "WAP294"
## [295] "WAP295" "WAP296" "WAP297"
## [298] "WAP298" "WAP299" "WAP300"
## [301] "WAP301" "WAP302" "WAP303"
## [304] "WAP304" "WAP305" "WAP306"
## [307] "WAP307" "WAP308" "WAP309"
## [310] "WAP310" "WAP311" "WAP312"
## [313] "WAP313" "WAP314" "WAP315"
## [316] "WAP316" "WAP317" "WAP318"
## [319] "WAP319" "WAP320" "WAP321"
## [322] "WAP322" "WAP323" "WAP324"
## [325] "WAP325" "WAP326" "WAP327"
## [328] "WAP328" "WAP329" "WAP330"
## [331] "WAP331" "WAP332" "WAP333"
## [334] "WAP334" "WAP335" "WAP336"
## [337] "WAP337" "WAP338" "WAP339"
## [340] "WAP340" "WAP341" "WAP342"
## [343] "WAP343" "WAP344" "WAP345"
## [346] "WAP346" "WAP347" "WAP348"
## [349] "WAP349" "WAP350" "WAP351"
## [352] "WAP352" "WAP353" "WAP354"
## [355] "WAP355" "WAP356" "WAP357"
## [358] "WAP358" "WAP359" "WAP360"
## [361] "WAP361" "WAP362" "WAP363"
## [364] "WAP364" "WAP365" "WAP366"
## [367] "WAP367" "WAP368" "WAP369"
## [370] "WAP370" "WAP371" "WAP372"
## [373] "WAP373" "WAP374" "WAP375"
## [376] "WAP376" "WAP377" "WAP378"
## [379] "WAP379" "WAP380" "WAP381"
## [382] "WAP382" "WAP383" "WAP384"
## [385] "WAP385" "WAP386" "WAP387"
## [388] "WAP388" "WAP389" "WAP390"
## [391] "WAP391" "WAP392" "WAP393"
## [394] "WAP394" "WAP395" "WAP396"
## [397] "WAP397" "WAP398" "WAP399"
## [400] "WAP400" "WAP401" "WAP402"
## [403] "WAP403" "WAP404" "WAP405"
## [406] "WAP406" "WAP407" "WAP408"
## [409] "WAP409" "WAP410" "WAP411"
## [412] "WAP412" "WAP413" "WAP414"
## [415] "WAP415" "WAP416" "WAP417"
## [418] "WAP418" "WAP419" "WAP420"
## [421] "WAP421" "WAP422" "WAP423"
## [424] "WAP424" "WAP425" "WAP426"
## [427] "WAP427" "WAP428" "WAP429"
## [430] "WAP430" "WAP431" "WAP432"
## [433] "WAP433" "WAP434" "WAP435"
## [436] "WAP436" "WAP437" "WAP438"
## [439] "WAP439" "WAP440" "WAP441"
## [442] "WAP442" "WAP443" "WAP444"
## [445] "WAP445" "WAP446" "WAP447"
## [448] "WAP448" "WAP449" "WAP450"
## [451] "WAP451" "WAP452" "WAP453"
## [454] "WAP454" "WAP455" "WAP456"
## [457] "WAP457" "WAP458" "WAP459"
## [460] "WAP460" "WAP461" "WAP462"
## [463] "WAP463" "WAP464" "WAP465"
## [466] "WAP466" "WAP467" "WAP468"
## [469] "WAP469" "WAP470" "WAP471"
## [472] "WAP472" "WAP473" "WAP474"
## [475] "WAP475" "WAP476" "WAP477"
## [478] "WAP478" "WAP479" "WAP480"
## [481] "WAP481" "WAP482" "WAP483"
## [484] "WAP484" "WAP485" "WAP486"
## [487] "WAP487" "WAP488" "WAP489"
## [490] "WAP490" "WAP491" "WAP492"
## [493] "WAP493" "WAP494" "WAP495"
## [496] "WAP496" "WAP497" "WAP498"
## [499] "WAP499" "WAP500" "WAP501"
## [502] "WAP502" "WAP503" "WAP504"
## [505] "WAP505" "WAP506" "WAP507"
## [508] "WAP508" "WAP509" "WAP510"
## [511] "WAP511" "WAP512" "WAP513"
## [514] "WAP514" "WAP515" "WAP516"
## [517] "WAP517" "WAP518" "WAP519"
## [520] "WAP520" "LONGITUDE" "LATITUDE"
## [523] "FLOOR" "BUILDINGID" "SPACEID"
## [526] "RELATIVEPOSITION" "USERID" "PHONEID"
## [529] "TIMESTAMP"#dependent variable (Lat)
y.dep <- 369
#independent variables (WAPS)
x.indep <- c(1:367)
regression.model <- h2o.glm( y = y.dep, x = x.indep, training_frame = train.h2o, family = "gaussian")## Warning in .h2o.startModelJob(algo, params, h2oRestApiVersion): Dropping bad and constant columns: [WAP004, WAP246, WAP301, WAP247, WAP244, WAP365, WAP003, WAP245, WAP242, WAP243, WAP240, WAP241, WAP360, WAP160, WAP307, WAP349, WAP226, WAP303, WAP227, WAP304, WAP158, WAP159, WAP333, WAP254, WAP353, WAP152, WAP296, WAP095, WAP293, WAP093, WAP094, WAP092, WAP217, WAP239, WAP215, WAP238].##
|
| | 0%
|
|=== | 4%
|
|=================================================================| 100%## H2ORegressionMetrics: glm
## ** Reported on training data. **
##
## MSE: 188.9417
## RMSE: 13.74561
## MAE: 3.268599
## RMSLE: NaN
## Mean Residual Deviance : 188.9417
## R^2 : 0.741519
## Null Deviance :14573335
## Null D.o.F. :19936
## Residual Deviance :3766930
## Residual D.o.F. :19779
## AIC :161395.3##
|
| | 0%
|
|=================================================================| 100%## RMSE Rsquared MAE
## 4.864805e+06 3.579288e-02 4.864805e+06#Random Forest
system.time(
rforest.model <- h2o.randomForest(y=y.dep, x=x.indep, training_frame = train.h2o,
ntrees = 1000, mtries = 3, max_depth = 4, seed = 1122))## Warning in .h2o.startModelJob(algo, params, h2oRestApiVersion): Dropping bad and constant columns: [WAP004, WAP246, WAP301, WAP247, WAP244, WAP365, WAP003, WAP245, WAP242, WAP243, WAP240, WAP241, WAP360, WAP160, WAP307, WAP349, WAP226, WAP303, WAP227, WAP304, WAP158, WAP159, WAP333, WAP254, WAP353, WAP152, WAP296, WAP095, WAP293, WAP093, WAP094, WAP092, WAP217, WAP239, WAP215, WAP238].##
|
| | 0%
|
|== | 4%
|
|======= | 11%
|
|============== | 21%
|
|====================== | 34%
|
|====================================== | 58%
|
|====================================================== | 82%
|
|=================================================================| 100%## user system elapsed
## 0.20 0.00 7.49## H2ORegressionMetrics: drf
## ** Reported on training data. **
## ** Metrics reported on Out-Of-Bag training samples **
##
## MSE: 411.7117
## RMSE: 20.29068
## MAE: 6.07995
## RMSLE: NaN
## Mean Residual Deviance : 411.7117## Variable Importances:
## variable relative_importance scaled_importance percentage
## 1 WAP313 216489056.000000 1.000000 0.070431
## 2 WAP260 191283072.000000 0.883569 0.062231
## 3 WAP258 183884448.000000 0.849394 0.059824
## 4 WAP315 155841712.000000 0.719860 0.050701
## 5 WAP261 145931696.000000 0.674083 0.047476
##
## ---
## variable relative_importance scaled_importance percentage
## 326 WAP352 0.000000 0.000000 0.000000
## 327 WAP354 0.000000 0.000000 0.000000
## 328 WAP357 0.000000 0.000000 0.000000
## 329 WAP358 0.000000 0.000000 0.000000
## 330 WAP361 0.000000 0.000000 0.000000
## 331 WAP363 0.000000 0.000000 0.000000#making predictions on unseen data
system.time(predict.rforest <- as.data.frame(h2o.predict(rforest.model, test.h2o)))##
|
| | 0%
|
|=================================================================| 100%## user system elapsed
## 0.06 0.00 1.17## RMSE Rsquared MAE
## 4.864806e+06 5.058331e-02 4.864806e+06#GBM
system.time(
gbm.model <- h2o.gbm(y=y.dep, x=x.indep, training_frame = train.h2o, ntrees = 1000, max_depth = 4, learn_rate = 0.01, seed = 1122)
)## Warning in .h2o.startModelJob(algo, params, h2oRestApiVersion): Dropping bad and constant columns: [WAP004, WAP246, WAP301, WAP247, WAP244, WAP365, WAP003, WAP245, WAP242, WAP243, WAP240, WAP241, WAP360, WAP160, WAP307, WAP349, WAP226, WAP303, WAP227, WAP304, WAP158, WAP159, WAP333, WAP254, WAP353, WAP152, WAP296, WAP095, WAP293, WAP093, WAP094, WAP092, WAP217, WAP239, WAP215, WAP238].##
|
| | 0%
|
|= | 1%
|
|== | 2%
|
|== | 4%
|
|=== | 5%
|
|==== | 7%
|
|===== | 8%
|
|====== | 10%
|
|======= | 11%
|
|======== | 13%
|
|========= | 14%
|
|========== | 16%
|
|=========== | 17%
|
|============ | 19%
|
|============= | 20%
|
|============== | 21%
|
|=============== | 22%
|
|=============== | 24%
|
|================ | 25%
|
|================= | 26%
|
|================== | 28%
|
|=================== | 29%
|
|==================== | 31%
|
|===================== | 32%
|
|====================== | 33%
|
|====================== | 34%
|
|======================= | 36%
|
|======================== | 37%
|
|========================= | 38%
|
|========================== | 40%
|
|=========================== | 41%
|
|============================ | 42%
|
|============================ | 44%
|
|============================= | 45%
|
|============================== | 46%
|
|=============================== | 48%
|
|================================ | 49%
|
|================================= | 50%
|
|================================== | 52%
|
|=================================== | 53%
|
|=================================== | 54%
|
|==================================== | 56%
|
|===================================== | 57%
|
|====================================== | 58%
|
|======================================= | 60%
|
|======================================== | 61%
|
|======================================== | 62%
|
|========================================= | 64%
|
|========================================== | 65%
|
|=========================================== | 66%
|
|============================================ | 68%
|
|============================================= | 69%
|
|============================================== | 70%
|
|=============================================== | 72%
|
|=============================================== | 73%
|
|================================================ | 74%
|
|================================================= | 76%
|
|================================================== | 77%
|
|=================================================== | 78%
|
|==================================================== | 80%
|
|===================================================== | 81%
|
|===================================================== | 82%
|
|====================================================== | 84%
|
|======================================================= | 85%
|
|======================================================== | 86%
|
|========================================================= | 88%
|
|========================================================== | 89%
|
|=========================================================== | 90%
|
|============================================================ | 92%
|
|============================================================ | 93%
|
|============================================================= | 94%
|
|============================================================== | 96%
|
|=============================================================== | 97%
|
|================================================================ | 99%
|
|=================================================================| 100%## user system elapsed
## 1.06 0.04 81.80## H2ORegressionMetrics: gbm
## ** Reported on training data. **
##
## MSE: 69.70412
## RMSE: 8.3489
## MAE: 1.227049
## RMSLE: NaN
## Mean Residual Deviance : 69.70412##
|
| | 0%
|
|=================================================================| 100%## RMSE Rsquared MAE
## 4.864805e+06 2.763601e-02 4.864805e+06
Comment Box is loading comments…