Scribd
Upload
121 views1 page

Caret Package Infographic PDF

The Caret package in R provides functions to streamline the process of creating predictive models. It contains functions for data splitting, pre-processing, feature selection, model tuning, and visualization. Some key functions include createDataPartition for splitting data, preprocess for pre-processing, gafs for feature selection using genetic algorithms, trainControl for model tuning parameters, and calibration for visualizing model calibration. Caret attempts to simplify the model building process in R.

Uploaded by

Ali Habib
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
121 views1 page

Caret Package Infographic PDF

The Caret package in R provides functions to streamline the process of creating predictive models. It contains functions for data splitting, pre-processing, feature selection, model tuning, and visualization. Some key functions include createDataPartition for splitting data, preprocess for pre-processing, gafs for feature selection using genetic algorithms, trainControl for model tuning parameters, and calibration for visualizing model calibration. Caret attempts to simplify the model building process in R.

Uploaded by

Ali Habib
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Cheatsheet:Caret Package

CARET ( Classification And Regression Training) is a library in R which provides a set of functions
that attempt to streamline the process for creating predictive models.

1.Data Splitting
Function Description
1
createDaIt splits a vector 'y' with 80 percent
createDataPartition(y,p=0.8) data in one part and 20 percent in other
parttaPartition(y,p=0.8)
2
It creates subsamples from 'b' which are at a
maxDissim(a,b,n=2) maximum Dissimilarity from 'a'(a,b,n=2)

2.Data Pre-Processing
Function Description
1
It is used to perform preprocessing tasks like
preprocess(x, method=c("center","scale") centering, scaling and imputing missing values
in a dataset
2
BoxCoxTrans(y,...) To remove skeweness in a vector by using
boxcoxtransformations on it.
3
It is used to randomly sample the data so that
downSample(x,y,yname="class") every class has the same frequency as the
4
minority class.
It creates a full set of dummy variables for
dummyVars(formula,...) categorical variables

3.Feature Selection
Function Description
1

gafs.default(x,y,...) It is used to perform supervised feature selection


using genetic algorithms
2
nearZeroVar(x,..) It is used to identify predictors that have zero or
near zero variance.
3

pickSizeBest(x,metric,maximise) It is used to perform backward selection


4

rfe(x,..) It is used to perform a simple backward selection


5
It is used to calculate variable importance for
varImp(object,...) classification and regression models

4.Model Tuning
Function Description
1

trainControl It is used for controlling training parameters like


resampling, number of folds, iteration etc.
2
oneSE(x,metric,maximise) This function is used to set tuning paramters of a
model.

5.Visualization
Function Description
1
It is used to draw calibration plot that describe
calibration(x,data) show consistent model probabilities are with
2 the observed event rate.
densityplot.rfe(x,data,...) Lattice functions for plotting resampling results of
recursive feature selection
3

featureplot(x,y,plot...) A shortcut to produce lattice plots


4
plotClassProbs It is used to plot predicted probabilities in
classification model .
5
plotObsVsPred It is used to plot observed vs predicted results in
Classification and Regression Models

For more Infographics log on :


http://www.analyticsvidhya.com

You might also like