MACHINE LEARNING WITH R ++
DATA POPULASI
Input Data
MACHINE LEARNING
Pra Proses Data
Eksplorasi dan Pemilihan Variabel Model
Training & Tuning Model Tunggal
Training & Tuning Multimodel
Ensembele & Stacking Models
EXPLAINING UNI MODEL ML
Penjelasan Umum
LIME
Penjelasan SHAPLEY
Penjelasan DALEX
DEEP LEARNING
Multi Layer Perception (MLP)
Unsupervised Learning
SURVIVAL
Bangun Objek
Model CoxPH
Regresi Survival
Info ML Survival
ML untuk Survival Analisis
Input Data
Di bagian ini pengguna dapat membaca data terutama dalam format excel dan teks (csv)
Pilih Data:
Import Excel
Import CSV
cars
GermanCredit
orange
iris
UnasSim
segmentationData
lung(survival)
tempEng (TS1)
FoReco_data(TS2)
simts (TS3)
Pilih File EXCEL ...
Browse...
Nomor Sheet yang Dibaca
Pilih File CSV/ Teks
Browse...
Header
Separator
Comma
Semicolon
Tab
Quote
None
Double Quote
Single Quote
I Made Tirta (2022)
Referensi:
S. Prabhakaran. 2018.
Caret Package – A Practical Guide to Machine Learning in R
https://www.machinelearningplus.com/machine-learning/caret-package/
M Kuhn. 2018.
The caret Package.
https://topepo.github.io/caret/index.html
M. Kuhn & K. Johnson. 2013.
Applied Predictive Modeling.
Springer
Informasi Data
MANUAL
Dafar Data
Struktur Data
Ringkasan Data
INPUT DATA TRAINING
Prosentase Training (75% -85 %)
Pre Proses (PP) Data Hilang atau Transformasi
Metode Proses
none
BoxCox
YeoJohnson
expoTrans
center
scale
range
knnImpute
bagImpute
medianImpute
pca
ica
spatialSign
corr
zv
nzv
conditionalX
Total Data
Data Training
Data Testing
Praproses Data
Plot Pencar PP
Eksplorasi Data dan Pemilihan Variabel
Info Algoritma yang sesuai dengan tujuan analisis
Info Algoritma
----ADA---------
ada
AdaBag
AdaBoost.M1
adaboost
amdai
ANFIS
avNNet
awnb
awtan
----BAGGING----------
bag
bagEarth
bagEarthGCV
bagFDA
bagFDAGCV
bam
bartMachine
bayesglm
binda
blackboost
blasso
blassoAveraged
bridge
brnn
BstLm
bstSm
bstTree
C5.0
C5.0Cost
C5.0Rules
C5.0Tree
cforest
chaid
CSimca
ctree
ctree2
cubist
dda
deepboost
DENFIS
dnn
dwdLinear
dwdPoly
dwdRadial
earth
elm
enet
evtree
extraTrees
fda
FH.GBML
FIR.DM
foba
FRBCS.CHI
FRBCS.W
FS.HGD
-----GAM/GLM---------------
gam
gamboost
gamLoess
gamSpline
gaussprLinear
gaussprPoly
gaussprRadial
gbmh2o
gbm
gcvEarth
GFS.FR.MOGUL
GFS.LT.RS
GFS.THRIFT
glm.nb
glm
glmboost
glmneth2o
glmnet
glmStepAIC
gpls
hda
hdda
hdrda
HYFIS
icr
J48
JRip
kernelpls
kknn
knn
krlsPoly
krlsRadial
lars
lars2
lasso
lda
lda2
leapBackward
leapForward
leapSeq
Linda
lm
lmStepAIC
LMT
loclda
logicBag
LogitBoost
logreg
lssvmLinear
lssvmPoly
lssvmRadial
lvq
M5
M5Rules
manb
mda
Mlda
mlp
mlpKerasDecay
mlpKerasDecayCost
mlpKerasDropout
mlpKerasDropoutCost
mlpML
mlpSGD
mlpWeightDecay
mlpWeightDecayML
monmlp
msaenet
multinom
mxnet
mxnetAdam
naive_bayes
nb
nbDiscrete
nbSearch
neuralnet
nnet
nnls
nodeHarvest
null
OneR
ordinalNet
ORFlog
ORFpls
ORFridge
ORFsvm
ownn
pam
parRF
PART
partDSA
pcaNNet
pcr
pda
pda2
penalized
PenalizedLDA
plr
pls
plsRglm
polr
ppr
PRIM
protoclass
pythonKnnReg
qda
QdaCov
qrf
qrnn
randomGLM
ranger
rbf
rbfDDA
Rborist
rda
regLogistic
relaxo
rf
rFerns
RFlda
rfRules
ridge
rlda
rlm
rmda
rocc
rotationForest
rotationForestCp
rpart
rpart1SE
rpart2
rpartCost
rpartScore
rqlasso
rqnc
RRF
RRFglobal
rrlda
RSimca
rvmLinear
rvmPoly
rvmRadial
SBC
sda
sdwd
simpls
SLAVE
slda
smda
snn
sparseLDA
spikeslab
spls
stepLDA
stepQDA
superpc
----SVM----------------
svmBoundrangeString
svmExpoString
svmLinear
svmLinear2
svmLinear3
svmLinearWeights
svmLinearWeights2
svmPoly
svmRadial
svmRadialCost
svmRadialSigma
svmRadialWeights
svmSpectrumString
tan
tanSearch
treebag
vbmpRadial
vglmAdjCat
vglmContRatio
vglmCumulative
widekernelpls
WM
wsrf
xgbDART
xgbLinear
xgbTree
xyf
Eksplorasi Pemilihan Variabel Model
Jenis Fitur Plot
box
strip
density
pairs
ellipse
scatter
Info Algoritma
Plot Fitur
Plot Pencar
Model Training
Berdasarkan pilihan variabel pada Eksplorasi Data
Algoritma Training
----ADA---------
ada
AdaBag
AdaBoost.M1
adaboost
amdai
ANFIS
avNNet
awnb
awtan
----BAGGING----------
bag
bagEarth
bagEarthGCV
bagFDA
bagFDAGCV
bam
bartMachine
bayesglm
binda
blackboost
blasso
blassoAveraged
bridge
brnn
BstLm
bstSm
bstTree
C5.0
C5.0Cost
C5.0Rules
C5.0Tree
cforest
chaid
CSimca
ctree
ctree2
cubist
dda
deepboost
DENFIS
dnn
dwdLinear
dwdPoly
dwdRadial
earth
elm
enet
evtree
extraTrees
fda
FH.GBML
FIR.DM
foba
FRBCS.CHI
FRBCS.W
FS.HGD
-----GAM/GLM---------------
gam
gamboost
gamLoess
gamSpline
gaussprLinear
gaussprPoly
gaussprRadial
gbmh2o
gbm
gcvEarth
GFS.FR.MOGUL
GFS.LT.RS
GFS.THRIFT
glm.nb
glm
glmboost
glmneth2o
glmnet
glmStepAIC
gpls
hda
hdda
hdrda
HYFIS
icr
J48
JRip
kernelpls
kknn
knn
krlsPoly
krlsRadial
lars
lars2
lasso
lda
lda2
leapBackward
leapForward
leapSeq
Linda
lm
lmStepAIC
LMT
loclda
logicBag
LogitBoost
logreg
lssvmLinear
lssvmPoly
lssvmRadial
lvq
M5
M5Rules
manb
mda
Mlda
mlp
mlpKerasDecay
mlpKerasDecayCost
mlpKerasDropout
mlpKerasDropoutCost
mlpML
mlpSGD
mlpWeightDecay
mlpWeightDecayML
monmlp
msaenet
multinom
mxnet
mxnetAdam
naive_bayes
nb
nbDiscrete
nbSearch
neuralnet
nnet
nnls
nodeHarvest
null
OneR
ordinalNet
ORFlog
ORFpls
ORFridge
ORFsvm
ownn
pam
parRF
PART
partDSA
pcaNNet
pcr
pda
pda2
penalized
PenalizedLDA
plr
pls
plsRglm
polr
ppr
PRIM
protoclass
pythonKnnReg
qda
QdaCov
qrf
qrnn
randomGLM
ranger
rbf
rbfDDA
Rborist
rda
regLogistic
relaxo
rf
rFerns
RFlda
rfRules
ridge
rlda
rlm
rmda
rocc
rotationForest
rotationForestCp
rpart
rpart1SE
rpart2
rpartCost
rpartScore
rqlasso
rqnc
RRF
RRFglobal
rrlda
RSimca
rvmLinear
rvmPoly
rvmRadial
SBC
sda
sdwd
simpls
SLAVE
slda
smda
snn
sparseLDA
spikeslab
spls
stepLDA
stepQDA
superpc
----SVM----------------
svmBoundrangeString
svmExpoString
svmLinear
svmLinear2
svmLinear3
svmLinearWeights
svmLinearWeights2
svmPoly
svmRadial
svmRadialCost
svmRadialSigma
svmRadialWeights
svmSpectrumString
tan
tanSearch
treebag
vbmpRadial
vglmAdjCat
vglmContRatio
vglmCumulative
widekernelpls
WM
wsrf
xgbDART
xgbLinear
xgbTree
xyf
Metrik Luaran
Accuracy
Kappa
RMSE
ROC
Rsquared
Metode Search
grid
random
Training Control (Tuning)
none
boot
boot632
optimism_boot
boot_all
cv
repeatedcv
LOOCV
LGOCV
oob
adaptive_cv
adaptive_boot
adaptive_LGOCV
Tipe Ringkasan
postResample
twoClassSummary
multiClassSummary
prSummary
mnLogLoss
Banyaknya Fold (5-15):
DIAGRAM PENCAR: Pilih X (numerik) dan Y (Target) untuk Plot Final
Proses Training
Ringkasan Training
Kepentingan Variansi
CM Testing
Peluang (Testing)
Training Plot
Var Impt Plot
Data Pred Testing
Plot Final Klas
Plot Final Reg
Model Majemuk Pilih Maksimum 5 Algoritma yang relevan
Algoritma Training
----ADA---------
ada
AdaBag
AdaBoost.M1
adaboost
amdai
ANFIS
avNNet
awnb
awtan
----BAGGING----------
bag
bagEarth
bagEarthGCV
bagFDA
bagFDAGCV
bam
bartMachine
bayesglm
binda
blackboost
blasso
blassoAveraged
bridge
brnn
BstLm
bstSm
bstTree
C5.0
C5.0Cost
C5.0Rules
C5.0Tree
cforest
chaid
CSimca
ctree
ctree2
cubist
dda
deepboost
DENFIS
dnn
dwdLinear
dwdPoly
dwdRadial
earth
elm
enet
evtree
extraTrees
fda
FH.GBML
FIR.DM
foba
FRBCS.CHI
FRBCS.W
FS.HGD
-----GAM/GLM---------------
gam
gamboost
gamLoess
gamSpline
gaussprLinear
gaussprPoly
gaussprRadial
gbmh2o
gbm
gcvEarth
GFS.FR.MOGUL
GFS.LT.RS
GFS.THRIFT
glm.nb
glm
glmboost
glmneth2o
glmnet
glmStepAIC
gpls
hda
hdda
hdrda
HYFIS
icr
J48
JRip
kernelpls
kknn
knn
krlsPoly
krlsRadial
lars
lars2
lasso
lda
lda2
leapBackward
leapForward
leapSeq
Linda
lm
lmStepAIC
LMT
loclda
logicBag
LogitBoost
logreg
lssvmLinear
lssvmPoly
lssvmRadial
lvq
M5
M5Rules
manb
mda
Mlda
mlp
mlpKerasDecay
mlpKerasDecayCost
mlpKerasDropout
mlpKerasDropoutCost
mlpML
mlpSGD
mlpWeightDecay
mlpWeightDecayML
monmlp
msaenet
multinom
mxnet
mxnetAdam
naive_bayes
nb
nbDiscrete
nbSearch
neuralnet
nnet
nnls
nodeHarvest
null
OneR
ordinalNet
ORFlog
ORFpls
ORFridge
ORFsvm
ownn
pam
parRF
PART
partDSA
pcaNNet
pcr
pda
pda2
penalized
PenalizedLDA
plr
pls
plsRglm
polr
ppr
PRIM
protoclass
pythonKnnReg
qda
QdaCov
qrf
qrnn
randomGLM
ranger
rbf
rbfDDA
Rborist
rda
regLogistic
relaxo
rf
rFerns
RFlda
rfRules
ridge
rlda
rlm
rmda
rocc
rotationForest
rotationForestCp
rpart
rpart1SE
rpart2
rpartCost
rpartScore
rqlasso
rqnc
RRF
RRFglobal
rrlda
RSimca
rvmLinear
rvmPoly
rvmRadial
SBC
sda
sdwd
simpls
SLAVE
slda
smda
snn
sparseLDA
spikeslab
spls
stepLDA
stepQDA
superpc
----SVM----------------
svmBoundrangeString
svmExpoString
svmLinear
svmLinear2
svmLinear3
svmLinearWeights
svmLinearWeights2
svmPoly
svmRadial
svmRadialCost
svmRadialSigma
svmRadialWeights
svmSpectrumString
tan
tanSearch
treebag
vbmpRadial
vglmAdjCat
vglmContRatio
vglmCumulative
widekernelpls
WM
wsrf
xgbDART
xgbLinear
xgbTree
xyf
Metrik Luaran Multi Model
Accuracy
Kappa
RMSE
ROC
Rsquared
Metode Search
grid
random
Training Control
none
boot
boot632
optimism_boot
boot_all
cv
repeatedcv
LOOCV
LGOCV
oob
adaptive_cv
adaptive_boot
adaptive_LGOCV
Execute!
Luaran Mult Training
Ringkasan Luaran Mult
Korelasi Mult
Training Plot
Kepentingan Variansi
Menggabungkan Model Majemuk
Jenis Machine Learning:
Jenis ML:
Regresi
KLasifikasi
ENSEMBLE: Metrik
Accuracy
Kappa
RMSE
ROC
Rsquared
Metode Stacking:
STACKING: Metode
gbm
glm
rf
glmnet
Hasil Multi Model
Hasil Ensemble
Plot Ensemble
ProsesStack
Final Stack
Stack Plot
Diagnostik Plot Regresi
Sumber:
Dingari, N.C. (2018)
https://rpubs.com/dnchari/explainable_machine_learning
Penjelasan Umum:
Jenis Jarak
manhattan
gower
euclidean
maximum
canberra
binary
minkowski
Seleksi Fitur
auto
highest_weights
lasso_path
L-Kernel ( >0 ):
n-Fitur:
LIME, no klp (acuan):
LIME, n-indv untuk visualisasi:
Info (LIME)
Plot Feature (LIME)
Plot Explain (LIME)
Penjelasan model dengan LIME
PENJELASAN LIME:
Jenis Jarak
manhattan
gower
euclidean
maximum
canberra
binary
minkowski
Seleksi Fitur
auto
highest_weights
lasso_path
L-Kernel ( >0 ):
n-Fitur:
LIME, no klp (acuan):
LIME, n-indv untuk visualisasi:
Info (LIME)
Plot Feature (LIME)
Plot Explain (LIME)
Penjelasan model dengan Shapley
PENJELASAN LIME:
Jenis Jarak
manhattan
gower
euclidean
maximum
canberra
binary
minkowski
Seleksi Fitur
auto
highest_weights
lasso_path
L-Kernel ( >0 ):
n-Fitur:
LIME, no klp (acuan):
LIME, n-indv untuk visualisasi:
Info (LIME)
Plot Feature (LIME)
Plot Explain (LIME)
Penjelasan model dengan DALEX
PENJELASAN LIME:
Jenis Jarak
manhattan
gower
euclidean
maximum
canberra
binary
minkowski
Seleksi Fitur
auto
highest_weights
lasso_path
L-Kernel ( >0 ):
n-Fitur:
LIME, no klp (acuan):
LIME, n-indv untuk visualisasi:
Info (LIME)
Plot Feature (LIME)
Plot Explain (LIME)
MLP dengan Keras
Luaran1
Luaran2
Plot
Analisis Klaster dengan NN
PRAKATA
IMPOR DATA
MANUAL
Eksplorasi Objek Survival
Cara Input Formula
menu
skrip
Tipe Sensor:
nos
right
left
interval
counting
interval2
mstate
CI Pada Plot
Ya
Tidak
Objek Surv
Estimasi Surv
Diff
Plot Kaplan Meier
Bar Plot
MODEL COXPH
Lakukan eksplorasi terlebih dahulu, lalu lanjutkan dengan CoxPH
Tentukan variabel X untuk model CoxPH
Model CoxPH
Anova CoxPH
Asumsi CoxPH
Diagnostik CoxPH Plot
REGRESI SURVIVAL
Harus kerjakan Objek dan CoxPH dulu
Input Formula
menu
skrip
Jenis Distribusi Respon Y:
extreme
weibull
exponential
gaussian
logistic
lognormal
loglogistic
rayleigh
t
Fit Surv Reg
GOF Surv Reg
INFO SURVIVAL ML
berdasar paket mlr [Bischl e.al, https://mlr.mlr-org.com/]
Jenis Luaran
prob
numerics
factors
ordered
missings
weights
oneclass
twoclass
multiclass
class.weights
featimp
oobpreds
functionals
single.functional
se
lcens
rcens
icens
Tipe Survival Learner
surv.cforest
surv.CoxBoost
surv.coxph
surv.cv.CoxBoost
surv.cvglmnet
surv.gamboost
surv.gbm
surv.glmboost
surv.glmnet
surv.randomForestSRC
surv.ranger
surv.rpart
Info Klas
Info Measure
Menggabungkan Model Majemuk
Tipe Resample
CV
LOO
RepCV
Bootstrap
Holdout
GrowingWindowCV
FixedWindowCV
Tipe Luaran Prediksi
prob
response
Tipe Sensor:
nos
right
left
interval
counting
interval2
mstate
Prediksi Train dan Uji
Detail Khusus
Prediksi
Daftar Prediksi
Deskripsi
Features Importances
Plot1
Plot2