Call Package
lapply(c("h2o","readxl","caret","dplyr","ggplot2","hrbrthemes"),
library,character.only=T)##
## ----------------------------------------------------------------------
##
## Your next step is to start H2O:
## > h2o.init()
##
## For H2O package documentation, ask for help:
## > ??h2o
##
## After starting H2O, you can use the Web UI at http://localhost:54321
## For more information visit https://docs.h2o.ai
##
## ----------------------------------------------------------------------##
## Attaching package: 'h2o'## The following objects are masked from 'package:stats':
##
## cor, sd, var## The following objects are masked from 'package:base':
##
## %*%, %in%, &&, ||, apply, as.factor, as.numeric, colnames,
## colnames<-, ifelse, is.character, is.factor, is.numeric, log,
## log10, log1p, log2, round, signif, trunc## Loading required package: ggplot2## Loading required package: lattice##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union## NOTE: Either Arial Narrow or Roboto Condensed fonts are required to use these themes.## Please use hrbrthemes::import_roboto_condensed() to install Roboto Condensed and## if Arial Narrow is not on your system, please see https://bit.ly/arialnarrow## [[1]]
## [1] "h2o" "stats" "graphics" "grDevices" "utils" "datasets"
## [7] "methods" "base"
##
## [[2]]
## [1] "readxl" "h2o" "stats" "graphics" "grDevices" "utils"
## [7] "datasets" "methods" "base"
##
## [[3]]
## [1] "caret" "lattice" "ggplot2" "readxl" "h2o" "stats"
## [7] "graphics" "grDevices" "utils" "datasets" "methods" "base"
##
## [[4]]
## [1] "dplyr" "caret" "lattice" "ggplot2" "readxl" "h2o"
## [7] "stats" "graphics" "grDevices" "utils" "datasets" "methods"
## [13] "base"
##
## [[5]]
## [1] "dplyr" "caret" "lattice" "ggplot2" "readxl" "h2o"
## [7] "stats" "graphics" "grDevices" "utils" "datasets" "methods"
## [13] "base"
##
## [[6]]
## [1] "hrbrthemes" "dplyr" "caret" "lattice" "ggplot2"
## [6] "readxl" "h2o" "stats" "graphics" "grDevices"
## [11] "utils" "datasets" "methods" "base"Input Data
datann <- data.frame(read_excel("D:/Users/Documents/TPM/05. NN Regression - SeoulBikeData.xlsx"))[,-1]Splitting Data
set.seed(123);i<-createDataPartition(datann$Snowfall..cm.,p=0.8,list=F)
tr<-datann[i,]
ts<-datann[-i,]Pre-process to H2O Model
h2o.init()## Connection successful!
##
## R is connected to the H2O cluster:
## H2O cluster uptime: 2 hours 3 minutes
## H2O cluster timezone: Asia/Bangkok
## H2O data parsing timezone: UTC
## H2O cluster version: 3.38.0.1
## H2O cluster version age: 5 months and 6 days !!!
## H2O cluster name: H2O_started_from_R_Fitra_eoy050
## H2O cluster total nodes: 1
## H2O cluster total memory: 0.72 GB
## H2O cluster total cores: 4
## H2O cluster allowed cores: 4
## H2O cluster healthy: TRUE
## H2O Connection ip: localhost
## H2O Connection port: 54321
## H2O Connection proxy: NA
## H2O Internal Security: FALSE
## R Version: R version 4.2.2 (2022-10-31 ucrt)## Warning in h2o.clusterInfo():
## Your H2O cluster version is too old (5 months and 6 days)!
## Please download and install the latest version from http://h2o.ai/download/trh<-as.h2o(tr)##
|
| | 0%
|
|======================================================================| 100%tsh<-as.h2o(ts)##
|
| | 0%
|
|======================================================================| 100%Model 1
dl<-h2o.deeplearning(x=names(trh)[-1],y=names(trh)[1],trh,
nfolds=12,epochs=115,seed=25,
keep_cross_validation_predictions=T,
activation="Rectifier");dl #hidden default (200,200)## Warning in .h2o.processResponseWarnings(res): Dropping bad and constant columns: [Holiday, Functioning.Day, Seasons].##
|
| | 0%
|
|= | 1%
|
|= | 2%
|
|== | 2%
|
|=== | 4%
|
|==== | 6%
|
|===== | 7%
|
|===== | 8%
|
|====== | 9%
|
|======= | 10%
|
|======== | 11%
|
|======== | 12%
|
|========= | 13%
|
|========== | 14%
|
|========== | 15%
|
|=========== | 16%
|
|============ | 17%
|
|============ | 18%
|
|============= | 18%
|
|============= | 19%
|
|============== | 20%
|
|=============== | 21%
|
|=============== | 22%
|
|================ | 23%
|
|================= | 24%
|
|================== | 25%
|
|================== | 26%
|
|=================== | 27%
|
|=================== | 28%
|
|==================== | 28%
|
|==================== | 29%
|
|===================== | 30%
|
|====================== | 31%
|
|======================= | 33%
|
|======================== | 34%
|
|======================== | 35%
|
|========================= | 35%
|
|========================= | 36%
|
|========================== | 37%
|
|=========================== | 38%
|
|=========================== | 39%
|
|============================ | 40%
|
|============================ | 41%
|
|============================= | 41%
|
|============================= | 42%
|
|============================== | 42%
|
|============================== | 43%
|
|============================== | 44%
|
|=============================== | 44%
|
|=============================== | 45%
|
|================================ | 45%
|
|================================ | 46%
|
|================================= | 47%
|
|================================= | 48%
|
|================================== | 48%
|
|================================== | 49%
|
|=================================== | 49%
|
|=================================== | 50%
|
|=================================== | 51%
|
|==================================== | 51%
|
|==================================== | 52%
|
|===================================== | 52%
|
|===================================== | 53%
|
|====================================== | 54%
|
|====================================== | 55%
|
|======================================= | 55%
|
|======================================= | 56%
|
|======================================== | 57%
|
|========================================= | 58%
|
|========================================= | 59%
|
|========================================== | 59%
|
|=========================================== | 61%
|
|============================================ | 62%
|
|============================================= | 64%
|
|============================================= | 65%
|
|============================================== | 65%
|
|============================================== | 66%
|
|=============================================== | 67%
|
|=============================================== | 68%
|
|================================================ | 68%
|
|================================================ | 69%
|
|================================================= | 69%
|
|================================================= | 70%
|
|================================================== | 71%
|
|================================================== | 72%
|
|=================================================== | 72%
|
|=================================================== | 73%
|
|==================================================== | 74%
|
|==================================================== | 75%
|
|===================================================== | 75%
|
|===================================================== | 76%
|
|====================================================== | 77%
|
|========================================================== | 83%
|
|=========================================================== | 84%
|
|============================================================ | 85%
|
|============================================================ | 86%
|
|============================================================= | 86%
|
|============================================================= | 87%
|
|============================================================= | 88%
|
|============================================================== | 88%
|
|============================================================== | 89%
|
|=============================================================== | 90%
|
|=============================================================== | 91%
|
|================================================================ | 91%
|
|================================================================ | 92%
|
|================================================================= | 92%
|
|================================================================= | 93%
|
|================================================================== | 94%
|
|================================================================== | 95%
|
|=================================================================== | 95%
|
|=================================================================== | 96%
|
|==================================================================== | 97%
|
|==================================================================== | 98%
|
|===================================================================== | 98%
|
|===================================================================== | 99%
|
|======================================================================| 99%
|
|======================================================================| 100%## Model Details:
## ==============
##
## H2ORegressionModel: deeplearning
## Model ID: DeepLearning_model_R_1677385886459_4
## Status of Neuron Layers: predicting Rented.Bike.Count, regression, gaussian distribution, Quadratic loss, 42.401 weights/biases, 506,0 KB, 840.960 training samples, mini-batch size 1
## layer units type dropout l1 l2 mean_rate rate_rms momentum
## 1 1 9 Input 0.00 % NA NA NA NA NA
## 2 2 200 Rectifier 0.00 % 0.000000 0.000000 0.005278 0.007651 0.000000
## 3 3 200 Rectifier 0.00 % 0.000000 0.000000 0.112685 0.141129 0.000000
## 4 4 1 Linear NA 0.000000 0.000000 0.002310 0.001660 0.000000
## mean_weight weight_rms mean_bias bias_rms
## 1 NA NA NA NA
## 2 -0.041115 0.254775 0.152359 0.234858
## 3 -0.055206 0.129769 0.776165 0.227415
## 4 -0.035732 0.159716 1.064322 0.000000
##
##
## H2ORegressionMetrics: deeplearning
## ** Reported on training data. **
## ** Metrics reported on full training frame **
##
## MSE: 97280.29
## RMSE: 311.8979
## MAE: 217.0225
## RMSLE: NaN
## Mean Residual Deviance : 97280.29
##
##
##
## H2ORegressionMetrics: deeplearning
## ** Reported on cross-validation data. **
## ** 12-fold cross-validation on training data (Metrics computed for combined holdout predictions) **
##
## MSE: 130254.2
## RMSE: 360.9074
## MAE: 239.3832
## RMSLE: NaN
## Mean Residual Deviance : 130254.2
##
##
## Cross-Validation Metrics Summary:
## mean sd cv_1_valid cv_2_valid
## mae 239.416500 17.039099 253.506180 224.759080
## mean_residual_deviance 130257.440000 14178.136000 146392.100000 125653.690000
## mse 130257.440000 14178.136000 146392.100000 125653.690000
## r2 0.686359 0.038817 0.641775 0.721182
## residual_deviance 130257.440000 14178.136000 146392.100000 125653.690000
## rmse 360.399840 20.073936 382.612180 354.476620
## rmsle NA 0.000000 NA NA
## cv_3_valid cv_4_valid cv_5_valid cv_6_valid
## mae 248.222530 236.064360 242.256730 210.644030
## mean_residual_deviance 141776.660000 133401.000000 138339.950000 107786.160000
## mse 141776.660000 133401.000000 138339.950000 107786.160000
## r2 0.646831 0.650237 0.697647 0.751430
## residual_deviance 141776.660000 133401.000000 138339.950000 107786.160000
## rmse 376.532400 365.241030 371.940800 328.308000
## rmsle NA NA NA NA
## cv_7_valid cv_8_valid cv_9_valid cv_10_valid
## mae 266.882800 249.794900 236.106800 239.309160
## mean_residual_deviance 149329.690000 130268.875000 129489.305000 123209.570000
## mse 149329.690000 130268.875000 129489.305000 123209.570000
## r2 0.643817 0.668878 0.719508 0.701442
## residual_deviance 149329.690000 130268.875000 129489.305000 123209.570000
## rmse 386.432000 360.927800 359.846220 351.012200
## rmsle NA NA NA NA
## cv_11_valid cv_12_valid
## mae 253.855870 211.595540
## mean_residual_deviance 135101.690000 102340.540000
## mse 135101.690000 102340.540000
## r2 0.661522 0.732035
## residual_deviance 135101.690000 102340.540000
## rmse 367.561800 319.907070
## rmsle NA NAh<-h2o.performance(dl);r<-h@metrics$r2;r## [1] 0.766766Model 2
dl1<-h2o.deeplearning(x=names(trh)[-1],y=names(trh)[1],trh,
nfolds=12,epochs=115,seed=25,
keep_cross_validation_predictions=T,
hidden = c(300,200,100),activation="Rectifier");dl1## Warning in .h2o.processResponseWarnings(res): Dropping bad and constant columns: [Holiday, Functioning.Day, Seasons].##
|
| | 0%
|
| | 1%
|
|= | 1%
|
|= | 2%
|
|== | 2%
|
|== | 3%
|
|=== | 4%
|
|=== | 5%
|
|==== | 5%
|
|==== | 6%
|
|===== | 7%
|
|===== | 8%
|
|====== | 8%
|
|====== | 9%
|
|======= | 9%
|
|======= | 10%
|
|======= | 11%
|
|======== | 11%
|
|======== | 12%
|
|========= | 12%
|
|========= | 13%
|
|========== | 14%
|
|========== | 15%
|
|=========== | 15%
|
|=========== | 16%
|
|============ | 17%
|
|============ | 18%
|
|============= | 18%
|
|============= | 19%
|
|============== | 19%
|
|============== | 20%
|
|============== | 21%
|
|=============== | 21%
|
|=============== | 22%
|
|================ | 22%
|
|================ | 23%
|
|================ | 24%
|
|================= | 24%
|
|================= | 25%
|
|================== | 25%
|
|================== | 26%
|
|=================== | 27%
|
|=================== | 28%
|
|==================== | 28%
|
|==================== | 29%
|
|===================== | 29%
|
|===================== | 30%
|
|========================== | 38%
|
|=========================== | 38%
|
|=========================== | 39%
|
|============================ | 39%
|
|============================ | 40%
|
|============================ | 41%
|
|============================= | 41%
|
|============================= | 42%
|
|============================== | 42%
|
|============================== | 43%
|
|============================== | 44%
|
|=============================== | 44%
|
|=============================== | 45%
|
|================================ | 45%
|
|================================ | 46%
|
|================================= | 47%
|
|================================= | 48%
|
|================================== | 48%
|
|================================== | 49%
|
|=================================== | 49%
|
|=================================== | 50%
|
|=================================== | 51%
|
|==================================== | 51%
|
|==================================== | 52%
|
|===================================== | 52%
|
|===================================== | 53%
|
|====================================== | 54%
|
|====================================== | 55%
|
|======================================= | 55%
|
|======================================= | 56%
|
|======================================== | 57%
|
|======================================== | 58%
|
|========================================= | 58%
|
|========================================= | 59%
|
|========================================== | 60%
|
|=========================================== | 61%
|
|=========================================== | 62%
|
|============================================ | 62%
|
|============================================ | 63%
|
|============================================= | 64%
|
|============================================= | 65%
|
|============================================== | 65%
|
|============================================== | 66%
|
|=============================================== | 67%
|
|================================================ | 68%
|
|================================================ | 69%
|
|================================================= | 69%
|
|================================================= | 70%
|
|================================================= | 71%
|
|================================================== | 71%
|
|================================================== | 72%
|
|=================================================== | 72%
|
|=================================================== | 73%
|
|=================================================== | 74%
|
|==================================================== | 74%
|
|==================================================== | 75%Error in .h2o.doSafeREST(h2oRestApiVersion = h2oRestApiVersion, urlSuffix = urlSuffix, :
## Unexpected CURL error: Timeout was reached: [localhost:54321] Resolving timed out after 11929 milliseconds
## [1] "Job request failed Unexpected CURL error: Timeout was reached: [localhost:54321] Resolving timed out after 11929 milliseconds, will retry after 3s."
##
|
|========================================================= | 81%
|
|========================================================= | 82%
|
|========================================================== | 82%
|
|========================================================== | 83%Error in .h2o.doSafeREST(h2oRestApiVersion = h2oRestApiVersion, urlSuffix = urlSuffix, :
## Unexpected CURL error: Timeout was reached: [localhost:54321] Connection time-out
## [1] "Job request failed Unexpected CURL error: Timeout was reached: [localhost:54321] Connection time-out, will retry after 3s."
##
|
|=========================================================== | 84%
|
|=========================================================== | 85%
|
|============================================================ | 85%
|
|============================================================ | 86%
|
|============================================================= | 86%
|
|============================================================= | 87%
|
|============================================================= | 88%
|
|============================================================== | 88%
|
|============================================================== | 89%
|
|=============================================================== | 89%
|
|=============================================================== | 90%
|
|=============================================================== | 91%
|
|================================================================ | 91%
|
|================================================================ | 92%
|
|================================================================= | 92%
|
|================================================================= | 93%
|
|================================================================= | 94%
|
|================================================================== | 94%
|
|================================================================== | 95%
|
|=================================================================== | 95%
|
|=================================================================== | 96%
|
|==================================================================== | 97%
|
|==================================================================== | 98%
|
|===================================================================== | 98%
|
|===================================================================== | 99%
|
|======================================================================| 99%
|
|======================================================================| 100%## Model Details:
## ==============
##
## H2ORegressionModel: deeplearning
## Model ID: DeepLearning_model_R_1677385886459_5
## Status of Neuron Layers: predicting Rented.Bike.Count, regression, gaussian distribution, Quadratic loss, 83.401 weights/biases, 989,5 KB, 813.601 training samples, mini-batch size 1
## layer units type dropout l1 l2 mean_rate rate_rms momentum
## 1 1 9 Input 0.00 % NA NA NA NA NA
## 2 2 300 Rectifier 0.00 % 0.000000 0.000000 0.028458 0.030978 0.000000
## 3 3 200 Rectifier 0.00 % 0.000000 0.000000 0.021209 0.027022 0.000000
## 4 4 100 Rectifier 0.00 % 0.000000 0.000000 0.026573 0.047228 0.000000
## 5 5 1 Linear NA 0.000000 0.000000 0.000689 0.000326 0.000000
## mean_weight weight_rms mean_bias bias_rms
## 1 NA NA NA NA
## 2 -0.029176 0.247557 0.361664 0.189008
## 3 -0.033128 0.104047 0.929238 0.061913
## 4 -0.041477 0.145876 0.783925 0.462913
## 5 -0.020998 0.177461 0.066837 0.000000
##
##
## H2ORegressionMetrics: deeplearning
## ** Reported on training data. **
## ** Metrics reported on full training frame **
##
## MSE: 67553.27
## RMSE: 259.9101
## MAE: 169.7145
## RMSLE: NaN
## Mean Residual Deviance : 67553.27
##
##
##
## H2ORegressionMetrics: deeplearning
## ** Reported on cross-validation data. **
## ** 12-fold cross-validation on training data (Metrics computed for combined holdout predictions) **
##
## MSE: 115450.5
## RMSE: 339.7801
## MAE: 217.0844
## RMSLE: NaN
## Mean Residual Deviance : 115450.5
##
##
## Cross-Validation Metrics Summary:
## mean sd cv_1_valid cv_2_valid
## mae 217.286240 12.821776 222.566120 223.050690
## mean_residual_deviance 115458.150000 11996.120000 115033.590000 128246.540000
## mse 115458.150000 11996.120000 115033.590000 128246.540000
## r2 0.722425 0.028207 0.718510 0.715428
## residual_deviance 115458.150000 11996.120000 115033.590000 128246.540000
## rmse 339.364000 17.793196 339.166020 358.115260
## rmsle NA 0.000000 NA NA
## cv_3_valid cv_4_valid cv_5_valid cv_6_valid
## mae 238.294390 214.660680 234.075600 195.019360
## mean_residual_deviance 126683.810000 119690.010000 134555.390000 95446.875000
## mse 126683.810000 119690.010000 134555.390000 95446.875000
## r2 0.684427 0.686186 0.705918 0.779886
## residual_deviance 126683.810000 119690.010000 134555.390000 95446.875000
## rmse 355.926700 345.962430 366.817900 308.944800
## rmsle NA NA NA NA
## cv_7_valid cv_8_valid cv_9_valid cv_10_valid
## mae 222.292560 220.790070 208.514070 198.378080
## mean_residual_deviance 121915.305000 103861.770000 118993.090000 98665.350000
## mse 121915.305000 103861.770000 118993.090000 98665.350000
## r2 0.709206 0.736000 0.742244 0.760917
## residual_deviance 121915.305000 103861.770000 118993.090000 98665.350000
## rmse 349.163730 322.275900 344.953770 314.110400
## rmsle NA NA NA NA
## cv_11_valid cv_12_valid
## mae 210.531510 219.261800
## mean_residual_deviance 112486.940000 109919.164000
## mse 112486.940000 109919.164000
## r2 0.718180 0.712192
## residual_deviance 112486.940000 109919.164000
## rmse 335.390720 331.540600
## rmsle NA NAh1<-h2o.performance(dl1);r1<-h1@metrics$r2;r1## [1] 0.8380379Model 3
dl2<-h2o.deeplearning(x=names(trh)[-1],y=names(trh)[1],trh,
nfolds=12,epochs=115,seed=25,
keep_cross_validation_predictions=T,
hidden = c(200,200,200),activation="Rectifier");dl2## Warning in .h2o.processResponseWarnings(res): Dropping bad and constant columns: [Holiday, Functioning.Day, Seasons].##
|
| | 0%
|
| | 1%
|
|= | 1%
|
|= | 2%
|
|== | 2%
|
|== | 3%
|
|=== | 4%
|
|=== | 5%
|
|==== | 5%
|
|==== | 6%
|
|===== | 6%
|
|===== | 7%
|
|===== | 8%
|
|====== | 8%
|
|====== | 9%
|
|======= | 9%
|
|======= | 10%
|
|======== | 11%
|
|======== | 12%
|
|========= | 13%
|
|========== | 14%
|
|========== | 15%
|
|=========== | 16%
|
|============ | 17%
|
|============ | 18%
|
|============= | 18%
|
|============= | 19%
|
|============== | 20%
|
|============== | 21%
|
|=============== | 21%
|
|=============== | 22%
|
|================ | 23%
|
|================= | 24%
|
|========================== | 37%
|
|=============================== | 44%
|
|=============================== | 45%
|
|================================ | 45%
|
|================================ | 46%
|
|================================= | 47%
|
|================================= | 48%
|
|================================== | 48%
|
|================================== | 49%
|
|=================================== | 50%
|
|==================================== | 51%
|
|==================================== | 52%
|
|===================================== | 52%
|
|===================================== | 53%Error in .h2o.doSafeREST(h2oRestApiVersion = h2oRestApiVersion, urlSuffix = urlSuffix, :
## Unexpected CURL error: Timeout was reached: [localhost:54321] Connection timed out after 14546 milliseconds
## [1] "Job request failed Unexpected CURL error: Timeout was reached: [localhost:54321] Connection timed out after 14546 milliseconds, will retry after 3s."
##
|
|====================================== | 54%
|
|====================================== | 55%
|
|======================================= | 55%
|
|======================================= | 56%
|
|======================================== | 57%
|
|======================================== | 58%
|
|========================================= | 58%
|
|========================================= | 59%
|
|========================================== | 59%
|
|========================================== | 60%
|
|=========================================== | 61%
|
|=========================================== | 62%
|
|============================================ | 62%
|
|============================================ | 63%
|
|============================================= | 64%
|
|============================================= | 65%
|
|============================================== | 65%
|
|============================================== | 66%
|
|=============================================== | 66%
|
|=============================================== | 67%
|
|=============================================== | 68%
|
|================================================ | 68%
|
|================================================ | 69%
|
|================================================= | 70%
|
|================================================= | 71%
|
|================================================== | 71%
|
|================================================== | 72%
|
|==================================================== | 75%
|
|===================================================== | 75%
|
|===================================================== | 76%
|
|====================================================== | 77%Error in .h2o.doSafeREST(h2oRestApiVersion = h2oRestApiVersion, urlSuffix = urlSuffix, :
## Unexpected CURL error: Timeout was reached: [localhost:54321] Connection timed out after 10108 milliseconds
## [1] "Job request failed Unexpected CURL error: Timeout was reached: [localhost:54321] Connection timed out after 10108 milliseconds, will retry after 3s."
##
|
|====================================================== | 78%
|
|======================================================= | 78%
|
|======================================================= | 79%
|
|======================================================== | 79%
|
|======================================================== | 80%
|
|======================================================== | 81%
|
|========================================================= | 81%
|
|========================================================= | 82%
|
|========================================================== | 82%
|
|========================================================== | 83%
|
|=========================================================== | 84%
|
|=========================================================== | 85%
|
|============================================================ | 85%
|
|============================================================ | 86%
|
|============================================================= | 86%
|
|============================================================= | 87%
|
|============================================================= | 88%
|
|============================================================== | 88%
|
|============================================================== | 89%
|
|=============================================================== | 89%
|
|=============================================================== | 90%
|
|=============================================================== | 91%
|
|================================================================ | 91%
|
|================================================================ | 92%
|
|================================================================= | 92%
|
|================================================================= | 93%
|
|================================================================== | 94%
|
|================================================================== | 95%
|
|=================================================================== | 95%
|
|=================================================================== | 96%
|
|==================================================================== | 96%
|
|==================================================================== | 97%
|
|==================================================================== | 98%
|
|===================================================================== | 98%
|
|===================================================================== | 99%
|
|======================================================================| 99%
|
|======================================================================| 100%Error in .h2o.doSafeREST(h2oRestApiVersion = h2oRestApiVersion, urlSuffix = urlSuffix, :
## Unexpected CURL error: Timeout was reached: [localhost:54321] Connection time-out
## [1] "Job request failed Unexpected CURL error: Timeout was reached: [localhost:54321] Connection time-out, will retry after 3s."## Model Details:
## ==============
##
## H2ORegressionModel: deeplearning
## Model ID: DeepLearning_model_R_1677385886459_6
## Status of Neuron Layers: predicting Rented.Bike.Count, regression, gaussian distribution, Quadratic loss, 82.601 weights/biases, 980,1 KB, 812.928 training samples, mini-batch size 1
## layer units type dropout l1 l2 mean_rate rate_rms momentum
## 1 1 9 Input 0.00 % NA NA NA NA NA
## 2 2 200 Rectifier 0.00 % 0.000000 0.000000 0.004968 0.005339 0.000000
## 3 3 200 Rectifier 0.00 % 0.000000 0.000000 0.018597 0.019647 0.000000
## 4 4 200 Rectifier 0.00 % 0.000000 0.000000 0.050648 0.076482 0.000000
## 5 5 1 Linear NA 0.000000 0.000000 0.001098 0.000517 0.000000
## mean_weight weight_rms mean_bias bias_rms
## 1 NA NA NA NA
## 2 -0.026467 0.270889 0.368908 0.233145
## 3 -0.036813 0.106407 0.949768 0.084539
## 4 -0.032635 0.106966 0.571763 0.420903
## 5 0.004344 0.155217 0.991146 0.000000
##
##
## H2ORegressionMetrics: deeplearning
## ** Reported on training data. **
## ** Metrics reported on full training frame **
##
## MSE: 99088.71
## RMSE: 314.7836
## MAE: 228.2953
## RMSLE: NaN
## Mean Residual Deviance : 99088.71
##
##
##
## H2ORegressionMetrics: deeplearning
## ** Reported on cross-validation data. **
## ** 12-fold cross-validation on training data (Metrics computed for combined holdout predictions) **
##
## MSE: 118717.8
## RMSE: 344.5545
## MAE: 224.4222
## RMSLE: NaN
## Mean Residual Deviance : 118717.8
##
##
## Cross-Validation Metrics Summary:
## mean sd cv_1_valid cv_2_valid
## mae 224.538210 12.931416 229.943740 227.762940
## mean_residual_deviance 118745.140000 12390.155000 118970.470000 127198.870000
## mse 118745.140000 12390.155000 118970.470000 127198.870000
## r2 0.714554 0.028988 0.708876 0.717753
## residual_deviance 118745.140000 12390.155000 118970.470000 127198.870000
## rmse 344.159180 18.078484 344.920960 356.649500
## rmsle 1.262317 0.000000 NA NA
## cv_3_valid cv_4_valid cv_5_valid cv_6_valid
## mae 240.753400 221.731830 241.072950 205.896090
## mean_residual_deviance 131161.280000 119731.270000 138527.360000 100553.220000
## mse 131161.280000 119731.270000 138527.360000 100553.220000
## r2 0.673274 0.686078 0.697237 0.768110
## residual_deviance 131161.280000 119731.270000 138527.360000 100553.220000
## rmse 362.161960 346.022060 372.192630 317.101260
## rmsle 1.262317 NA NA NA
## cv_7_valid cv_8_valid cv_9_valid cv_10_valid
## mae 241.040420 233.207140 218.719680 211.715530
## mean_residual_deviance 132417.160000 117900.625000 118666.990000 103972.950000
## mse 132417.160000 117900.625000 118666.990000 103972.950000
## r2 0.684157 0.700316 0.742950 0.748056
## residual_deviance 132417.160000 117900.625000 118666.990000 103972.950000
## rmse 363.891700 343.366600 344.480770 322.448360
## rmsle NA NA NA NA
## cv_11_valid cv_12_valid
## mae 213.779880 208.834950
## mean_residual_deviance 114975.770000 100865.695000
## mse 114975.770000 100865.695000
## r2 0.711945 0.735897
## residual_deviance 114975.770000 100865.695000
## rmse 339.080780 317.593600
## rmsle NA NAh2<-h2o.performance(dl2);r2<-h2@metrics$r2;r2## [1] 0.7624303d<-data.frame(hidden=c("Default","{300,200,100}","{200,200,200}"),
r2=round(c(r,r1,r2),2))Plot comparation of R2 for each layer
ggplot(d,aes(x=hidden,y=r2,fill=hidden))+geom_bar(stat="identity")+
theme_ipsum_rc(axis_title_just = "center",axis_text_size = 12)+
labs(fill="Hidden layer",y="")+scale_fill_brewer(palette="Blues")+
coord_polar("y")+theme(axis.text.y.left = element_blank(),
axis.title.y.left = element_blank())+coord_polar("y")## Coordinate system already present. Adding new coordinate system, which will
## replace the existing one.## Warning in grid.Call(C_stringMetric, as.graphicsAnnot(x$label)): font family not
## found in Windows font database
## Warning in grid.Call(C_stringMetric, as.graphicsAnnot(x$label)): font family not
## found in Windows font database
## Warning in grid.Call(C_stringMetric, as.graphicsAnnot(x$label)): font family not
## found in Windows font database
## Warning in grid.Call(C_stringMetric, as.graphicsAnnot(x$label)): font family not
## found in Windows font database## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database
## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database
## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database
## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database
## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database
#best model: Model 2 (3 hidden layer, 300 neuron layer 1, 200 neuron di layer 2, dan 100 neuron di layer 1)Learning Curve Plot
lc1<-h2o.learning_curve_plot(dl1,"rmse",cv_lines = F,cv_ribbon = F)
ggplot(lc1$data,aes(x=x,y=metric))+geom_point(col="coral",size=4)+
geom_line(col="steelblue",lwd=1.6)+
theme_ipsum_rc(axis_title_just = "center",axis_title_size = 12)+
labs(x="\nEpochs",y="RMSE\n")+geom_vline(xintercept = 116.034709514103,
col="darkgreen",lwd=1.4,lty=5)+
geom_text(aes(x=116.04,y=250,label="Selected epochs: 115.53"),
hjust=1.05,col="darkgreen",size=4.5,vjust=-16.8)## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database
## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database
## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database
## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database
## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database
## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database
## Warning in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, : font
## family not found in Windows font database## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
Variable Importance Plot
hpi<-h2o.permutation_importance_plot(dl1,tsh,metric="RMSE",seed=123);hpi
## Variable Relative Importance Scaled Importance Percentage
## 1 Hour 286.70008 1.00000000 0.26325138
## 2 Temperature..C. 228.83708 0.79817584 0.21012090
## 3 Solar.Radiation..MJ.m2. 214.40920 0.74785189 0.19687305
## 4 Dew.point.temperature..C. 147.85937 0.51572837 0.13576621
## 5 Humidity... 121.82426 0.42491882 0.11186047
## 6 Rainfall.mm. 32.02179 0.11169091 0.02940279
## 7 Wind.speed..m.s. 22.33465 0.07790250 0.02050794
## 8 Snowfall..cm. 20.29776 0.07079789 0.01863764
## 9 Visibility..10m. 14.78921 0.05158427 0.01357963
## 10 Holiday 0.00000 0.00000000 0.00000000
## 11 Functioning.Day 0.00000 0.00000000 0.00000000
## 12 Seasons 0.00000 0.00000000 0.00000000Prediction
h2o.predict(dl1,tsh)##
|
| | 0%
|
|======================================================================| 100%## predict
## 1 273.35520
## 2 563.26095
## 3 259.65934
## 4 93.31756
## 5 88.21706
## 6 365.32417
##
## [1752 rows x 1 column]Prediction Performance
hv<-h2o.performance(dl1,tsh);hv;hv@metrics$r2## H2ORegressionMetrics: deeplearning
##
## MSE: 108592.1
## RMSE: 329.5332
## MAE: 208.7487
## RMSLE: NaN
## Mean Residual Deviance : 108592.1## [1] 0.7360788#sedikit lebih kecil dibandingkan r2 trainingDifference of R2 (training vs testing)
h1@metrics$r2-hv@metrics$r2 #Kecil perbedaannya sehingga tidak terdapat overfitting## [1] 0.1019592