Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[SPARK-21622][ML][SparkR] Support offset in SparkR GLM #18831

Closed
wants to merge 3 commits into from
Closed

[SPARK-21622][ML][SparkR] Support offset in SparkR GLM #18831

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
6ec068e
add offset to SparkR
actuaryzhang Aug 3, 2017
dc8ccbc
fix unit test
actuaryzhang Aug 3, 2017
3c4ebf9
fix unit test
actuaryzhang Aug 3, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 18 additions & 4 deletions R/pkg/R/mllib_regression.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -76,6 +76,8 @@ setClass("IsotonicRegressionModel", representation(jobj = "jobj"))
#' "frequencyDesc", "frequencyAsc", "alphabetDesc", and "alphabetAsc".
#' The default value is "frequencyDesc". When the ordering is set to
#' "alphabetDesc", this drops the same category as R when encoding strings.
#' @param offsetCol the offset column name. If this is not set or empty, we treat all instance offsets
#' as 0.0. The feature specified as offset has a constant coefficient of 1.0.
#' @param ... additional arguments passed to the method.
#' @aliases spark.glm,SparkDataFrame,formula-method
#' @return \code{spark.glm} returns a fitted generalized linear model.
Expand Down Expand Up @@ -127,7 +129,8 @@ setMethod("spark.glm", signature(data = "SparkDataFrame", formula = "formula"),
function(data, formula, family = gaussian, tol = 1e-6, maxIter = 25, weightCol = NULL,
regParam = 0.0, var.power = 0.0, link.power = 1.0 - var.power,
stringIndexerOrderType = c("frequencyDesc", "frequencyAsc",
"alphabetDesc", "alphabetAsc")) {
"alphabetDesc", "alphabetAsc"),
offsetCol = NULL) {

stringIndexerOrderType <- match.arg(stringIndexerOrderType)
if (is.character(family)) {
Expand Down Expand Up @@ -159,12 +162,19 @@ setMethod("spark.glm", signature(data = "SparkDataFrame", formula = "formula"),
weightCol <- as.character(weightCol)
}

if (!is.null(offsetCol)) {
offsetCol <- as.character(offsetCol)
if (nchar(offsetCol) == 0) {
offsetCol <- NULL
}
}

# For known families, Gamma is upper-cased
jobj <- callJStatic("org.apache.spark.ml.r.GeneralizedLinearRegressionWrapper",
"fit", formula, data@sdf, tolower(family$family), family$link,
tol, as.integer(maxIter), weightCol, regParam,
as.double(var.power), as.double(link.power),
stringIndexerOrderType)
stringIndexerOrderType, offsetCol)
new("GeneralizedLinearRegressionModel", jobj = jobj)
})

Expand Down Expand Up @@ -192,6 +202,8 @@ setMethod("spark.glm", signature(data = "SparkDataFrame", formula = "formula"),
#' "frequencyDesc", "frequencyAsc", "alphabetDesc", and "alphabetAsc".
#' The default value is "frequencyDesc". When the ordering is set to
#' "alphabetDesc", this drops the same category as R when encoding strings.
#' @param offsetCol the offset column name. If this is not set or empty, we treat all instance offsets
#' as 0.0. The feature specified as offset has a constant coefficient of 1.0.
#' @return \code{glm} returns a fitted generalized linear model.
#' @rdname glm
#' @export
Expand All @@ -209,10 +221,12 @@ setMethod("glm", signature(formula = "formula", family = "ANY", data = "SparkDat
function(formula, family = gaussian, data, epsilon = 1e-6, maxit = 25, weightCol = NULL,
var.power = 0.0, link.power = 1.0 - var.power,
stringIndexerOrderType = c("frequencyDesc", "frequencyAsc",
"alphabetDesc", "alphabetAsc")) {
"alphabetDesc", "alphabetAsc"),
offsetCol = NULL) {
spark.glm(data, formula, family, tol = epsilon, maxIter = maxit, weightCol = weightCol,
var.power = var.power, link.power = link.power,
stringIndexerOrderType = stringIndexerOrderType)
stringIndexerOrderType = stringIndexerOrderType,
offsetCol = offsetCol)
})

# Returns the summary of a model produced by glm() or spark.glm(), similarly to R's summary().
Expand Down
8 changes: 8 additions & 0 deletions R/pkg/tests/fulltests/test_mllib_regression.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -173,6 +173,14 @@ test_that("spark.glm summary", {
expect_equal(stats$df.residual, rStats$df.residual)
expect_equal(stats$aic, rStats$aic)

# Test spark.glm works with offset
training <- suppressWarnings(createDataFrame(iris))
stats <- summary(spark.glm(training, Sepal_Width ~ Sepal_Length + Species,
family = poisson(), offsetCol = "Petal_Length"))
rStats <- suppressWarnings(summary(glm(Sepal.Width ~ Sepal.Length + Species,
data = iris, family = poisson(), offset = iris$Petal.Length)))
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

that's interesting - perhaps we should take col in addition to col name too

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Then do you want to make the change for weight as well?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

probably across every in ml.
let's discuss this in a new JIRA.

Copy link
Contributor

@yanboliang yanboliang Aug 5, 2017 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I vote to keep the name as it is, because it's the column name of offset rather than the offset itself. weightCol is the same. We would like to keep SparkR MLlib wrappers' argument name consistent with R only when it's applicable. I'm ok to create a new JIRA to discuss it. Thanks.

expect_true(all(abs(rStats$coefficients - stats$coefficients) < 1e-3))

# Test summary works on base GLM models
baseModel <- stats::glm(Sepal.Width ~ Sepal.Length + Species, data = iris)
baseSummary <- summary(baseModel)
Expand Down
4 changes: 3 additions & 1 deletion mllib/src/main/scala/org/apache/spark/ml/r/GeneralizedLinearRegressionWrapper.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -77,7 +77,8 @@ private[r] object GeneralizedLinearRegressionWrapper
regParam: Double,
variancePower: Double,
linkPower: Double,
stringIndexerOrderType: String): GeneralizedLinearRegressionWrapper = {
stringIndexerOrderType: String,
offsetCol: String): GeneralizedLinearRegressionWrapper = {
// scalastyle:on
val rFormula = new RFormula().setFormula(formula)
.setStringIndexerOrderType(stringIndexerOrderType)
Expand All @@ -99,6 +100,7 @@ private[r] object GeneralizedLinearRegressionWrapper
glr.setLink(link)
}
if (weightCol != null) glr.setWeightCol(weightCol)
if (offsetCol != null) glr.setOffsetCol(offsetCol)

val pipeline = new Pipeline()
.setStages(Array(rFormulaModel, glr))
Expand Down