Learn R Programming

⚠️There's a newer version (2.2) of this package.Take me there.

TDMR (version 2.0)

Tuned Data Mining in R

Description

Tuned Data Mining in R ('TDMR') performs the complete tuning of a data mining task (predictive analytics, that is classification and regression). Preprocessing parameters and modeling parameters can be tuned simultaneously. It incorporates a variety of tuners (among them 'SPOT' and 'CMA' with package 'rCMA') and allows integration of additional tuners. Noise handling in the data mining optimization process is supported, see Koch et al. (2015) .

Copy Link

Version

Install

install.packages('TDMR')

Monthly Downloads

23

Version

2.0

License

GPL (>= 2)

Maintainer

Wolfgang Konen

Last Published

May 4th, 2018

Functions in TDMR (2.0)

tdmClassifySummary

Print summary output for result from tdmClassifiyLoop and add result$y.

Construct a new environment envT of class TDMenvir.

tdmGraAndLogFinalize

Finalize graphics and log file

Load an envT-type environment from file fileRData.

Return list opts from the k-th element of envT$sCList

tdmBindResponse

Bind a column to a data frame.

tdmPreLevel2Target

Relate levels of a column with a target (column).

Merge the parameters from a partial list and the default list

Print objects using print if opts$VERBOSE>=1.

tdmEnvTAddGetters

Add getter functions getBst and getRes to environment envT

Load the mapping files.

Perform LHD tuning.

tdmROCR.TDMclassifier

Interactive plot of ROC, lift or other charts for a TDMclassifier object.

Tuning and unbiased evaluation in a big loop.

Tuning and unbiased evaluation (single tuning).

Perform a SPOT tuning.

Calculate confusion matrix, gain and RGain measure.

tdmModVote2Target

Analyze how the vote fraction corresponds to reliability of prediction.

Sensitivity Report Helper Function

tdmPreGroupLevels

Group the levels of factor variable in dset[,colname].

tdmGraAndLogInitialize

Initialize graphics and log file.

tdmOptsDefaultsFill

Fill the current opts.

tdmPreSFA.train

SFA (Slow Feature Analysis) for numeric columns in a data frame.

tdmPreAddMonomials

Add monomials of degree 2 to a data frame.

Single plot of ROC, lift or other chart for a TDMclassifier object.

tdmRegressSummary

Print summary output for result from tdmRegressLoop and add result$y.

tdmEnvTAddBstRes

Add BST and RES data frames to an existing envT environment.

tdmDispatchTuner

Helper function for tdmBigLoop.

Set list opts for the k-th element of envT$sCList

tdmClassifyLoop

Core classification double loop returning a TDMclassifier object.

tdmPreSFA.apply

Apply SFA (Slow Feature Analysis) to new data.

Core classification function of TDMR.

tdmGraphicCloseDev

Close all open graphic devices.

tdmDefaultsFill

Default values for list tdm.

Interactive plot of ROC charts

Initialize graphic device.

tdmPrePCA.train

PCA (Principal Component Analysis) for numeric columns in a data frame.

Function called by SPOT 2.0 tuner

Print objects using print if opts$VERBOSE>=2.

tdmGraphicNewWin

Initialize a new window.

tdmPreFindConstVar

Find constant columns.

Perform unbiased runs with best-solution parameters.

Update envT$tdm

Make a report plot based on envT

just to make example check happy

tdmPreNAroughfix

Replace <NA> values with suitable non <NA> values

tdmGraphicCloseWin

Close active file ("png").

tdmReadTaskData

Read task data.

tdmParaBootstrap

Parametric bootstrap: add 'noisy copies' to a data frame (training data).

tdmPrePCA.apply

Apply PCA (Principal Component Analysis) to new data.

tdmEnvTReportSens

Function to generate a report with sensitivity plot.

Apply the mapping from des to opts.

tdmOptsDefaultsSet

Default values for list opts.

Read data accoroding to opts settings.

Core regression function of TDMR.

tdmGraphicToTop

Bring the acitve window to the top

tdmModSortedRFimport

Sort the input variables decreasingly by their RF-importance.

tdmModCreateCVindex

Create and return a training-validation-set index vector.

Generates pseudo-random random number seeds.

tdmReadAndSplit

Read and split the task data.

Core regression double loop returning a TDMregressor object.

Default settings for the spotConfig part of TDMR.

Perform BFGS tuning.

Perform CMA-ES tuning (Java version).

Return the list 'opts'.

Perform CMA-ES tuning (R version).

dsetTrnVa.TDMdata

Return train-validation data of TDMdata object

dsetTest.TDMdata

Return test data of TDMdata object

Return test data

makeTdmRandomSeed

Factory method to make a function generating pseudo-random random number seeds.

Tuned Data Mining in R

Output objects to cat if opts$VERBOSE>=1.

Default settings for the data mining part of TDMR (list opts).

Perform Powell's tuning.

predict.TDMenvir

Make a prediction using the last model.

print.TDMclassifier

Print an overview for a TDMclassifier object.

Output objects to cat if opts$VERBOSE>=2.

Return train-validataion data

Print an overview for a TDMdata object.

print.TDMregressor

Print an overview for a TDMregressor object.