pmml (version 1.5.7)

pmml: Generate PMML for R objects

Description

pmml is a generic function implementing S3 methods used to produce the PMML (Predictive Model Markup Language) representation of an R model. The resulting PMML file can then be imported into other systems that accept PMML.

The same function can also be used to output variable transformations in PMML format. In particular, it can be used as a transformations generator. Various transformation operations can be implemented in R and those transformations can then be output in PMML format by calling the function with a NULL value for the model input and a pmmlTransformations object as the transforms input. Please see the R pmmlTransformations package for more information on how to create the pmmlTransformations object.

In addition, the pmml function can also be called using a pre-existing PMML model as the first input and a pmmlTransformations object as the transforms input. The result is a new PMML model with the transformation inserted as a "LocalTransformations" element in the original model. If the original model already had a "LocalTransformations" element, the new information will be appended to that element. If the model variables are derived directly from a chain of transformations defined in the transforms input, the field names in the model are replaced with the original field names with the correct data types to make a consistent model. The covered cases include model fields derived from an original field, model fields derived from a chain of transformations starting from an original field and mutiple fields derived from the same original field.

This package converts models to PMML version 4.3.

Please note that package XML_3.95-0.1 or later is required to perform the full and correct functionality of the pmml package.

If data used for an R model contains features of type character, these must be converted to factors before the model is trained and converted with pmml.

A list of all the supported packages is available in the vignette:

vignette("packages_and_functions", package="pmml").

Usage

pmml(model=NULL, model.name="Rattle_Model", 
     app.name="Rattle/PMML", description=NULL, 
     copyright=NULL, transforms=NULL, …)

Arguments

model

an object to be converted to PMML.

model.name

a name to be given to the model in the PMML code.

app.name

the name of the application that generated the PMML code.

description

a descriptive text for the Header element of the PMML code.

copyright

the copyright notice for the model.

transforms

data transformations represented in PMML via pmmlTransformations.

…

further arguments passed to or from other methods.

Value

An object of class XMLNode as that defined by the XML package. This represents the top level, or root node, of the XML document and is of type PMML. It can be written to file with saveXML.

Details

The Predictive Model Markup Language (PMML) is an XML-based language which provides a way for applications to define machine learning, statistical and data mining models and to share models between PMML compliant applications. More information about the PMML industry standard and the Data Mining Group can be found at http://www.dmg.org.

The generated PMML can be imported into any PMML consuming application, such as Zementis Predictive Analytics products, which integrate with web services, relational database systems and deploy natively on Hadoop in conjunction with Hive, Spark or Storm, as well as allow predictive analytics to be executed for IBM z Systems mainframe applications and real-time, streaming analytics platforms.

References

• Rattle home page: https://rattle.togaware.com/

• PMML home page: http://www.dmg.org

• A. Guazzelli, W. Lin, T. Jena (2012), PMML in Action: Unleashing the Power of Open Standards for Data Mining and Predictive Analytics. CreativeSpace (Second Edition) - Available on Amazon.com.

• A. Guazzelli, M. Zeller, W. Lin, G. Williams (2009), PMML: An Open Standard for Sharing Models. The R journal, Volume 1/1, 60-65

• A. Guazzelli, T. Jena, W. Lin, M. Zeller (2013). Extending the Naive Bayes Model Element in PMML: Adding Support for Continuous Input Variables. In Proceedings of the 19th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

• T. Jena, A. Guazzelli, W. Lin, M. Zeller (2013). The R pmmlTransformations Package. In Proceedings of the 19th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

See Also

pmml.ada, pmml.rules, pmml.coxph, pmml.cv.glmnet, pmml.glm, pmml.hclust, pmml.kmeans, pmml.ksvm, pmml.lm, pmml.multinom, pmml.naiveBayes, pmml.neighbr, pmml.nnet, pmml.randomForest, pmml.rfsrc, pmml.rpart, pmml.svm, pmml.xgb.Booster

Examples

# NOT RUN {
# Build a simple lm model
iris.lm <- lm(Sepal.Length ~ ., data=iris)

# Convert to pmml
pmml(iris.lm)

# Create a pmmlTransformations object
library(pmmlTransformations)
xo <- WrapData(iris)

# Transform the 'Sepal.Length' variable
xo <- MinMaxXform(xo,xformInfo="column1->d_sl")

# Output the tranformation in PMML format
pmml(NULL, transforms=xo)

# }