ml_glm: Generalized Linear Models

In danzafar/tidyspark: A Tidy Interface to Spark

Description

Fits generalized linear model against a spark_tbl. Users can call summary to print a summary of the fitted model, predict to make predictions on new data, and write_ml/read_ml to save/load fitted models.

Usage

ml_glm(
  data,
  formula,
  family = "gaussian",
  tol = 1e-06,
  maxIter = 25,
  weightCol = NULL,
  regParam = 0,
  var.power = 0,
  link.power = 1 - var.power,
  stringIndexerOrderType = c("frequencyDesc", "frequencyAsc", "alphabetDesc",
    "alphabetAsc"),
  offsetCol = NULL
)

## S4 method for signature 'GeneralizedLinearRegressionModel'
summary(object)

## S3 method for class 'summary.tidyGeneralizedLinearRegressionModel'
print(x, ...)

## S4 method for signature 'GeneralizedLinearRegressionModel'
predict(object, newData)

## S4 method for signature 'GeneralizedLinearRegressionModel,character'
write_ml(object, path, overwrite = FALSE)

Arguments

data	a spark_tbl for training.
formula	a symbolic description of the model to be fitted. Currently only a few formula operators are supported, including '~', '.', ':', '+', and '-'.
family	a description of the error distribution and link function to be used in the model. This can be a character string naming a family function, a family function or the result of a call to a family function. Refer R family at https://stat.ethz.ch/R-manual/R-devel/library/stats/html/family.html. Currently these families are supported: binomial, gaussian, Gamma, poisson and tweedie. Note that there are two ways to specify the tweedie family. Set family = "tweedie" and specify the var.power and link.power; When package statmod is loaded, the tweedie family is specified using the family definition therein, i.e., tweedie(var.power, link.power).
tol	positive convergence tolerance of iterations.
maxIter	integer giving the maximal number of IRLS iterations.
weightCol	the weight column name. If this is not set or NULL, we treat all instance weights as 1.0.
regParam	regularization parameter for L2 regularization.
var.power	the power in the variance function of the Tweedie distribution which provides the relationship between the variance and mean of the distribution. Only applicable to the Tweedie family.
link.power	the index in the power link function. Only applicable to the Tweedie family.
stringIndexerOrderType	how to order categories of a string feature column. This is used to decide the base level of a string feature as the last category after ordering is dropped when encoding strings. Supported options are "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.
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.
object	a fitted generalized linear model.
x	summary object of fitted generalized linear model returned by summary function.
...	additional arguments passed to the method.
newData	a spark_tbl for testing.
path	the directory where the model is saved.
overwrite	overwrites or not if the output path already exists. Default is FALSE which means throw exception if the output path exists.

Value

ml_glm returns a fitted generalized linear model.

summary returns summary information of the fitted model, which is a list. The list of components includes at least the coefficients (coefficients matrix, which includes coefficients, standard error of coefficients, t value and p value), null.deviance (null/residual degrees of freedom), aic (AIC) and iter (number of iterations IRLS takes). If there are collinear columns in the data, the coefficients matrix only provides coefficients.

predict returns a spark_tbl containing predicted labels in a column named "prediction".

Note

ml_glm since 2.0.0

summary(GeneralizedLinearRegressionModel) since 2.0.0

print.summary.GeneralizedLinearRegressionModel since 2.0.0

predict(GeneralizedLinearRegressionModel) since 1.5.0

write_ml(GeneralizedLinearRegressionModel, character) since 2.0.0

Examples

## Not run: 
spark_session()
t <- as.data.frame(Titanic, stringsAsFactors = FALSE)
df <- spark_tbl(t)
model <- df %>%
  ml_glm(Freq ~ Sex + Age, family = "gaussian")
summary(model)

# fitted values on training data
fitted <- predict(model, df)
fitted %>%
  select(Freq, prediction) %>%
  collect

# save fitted model to input path
path <- "path/to/model"
write_ml(model, path)

# can also read back the saved model and print
savedModel <- read_ml(path)
summary(savedModel)

# note that the default string encoding is different from R's glm
model2 <- df %>%
  ml_glm(Freq ~ Sex + Age, family = "gaussian")
summary(model2)
# use stringIndexerOrderType = "alphabetDesc" to force string encoding
# to be consistent with R
model3 <- ml_glm(df, Freq ~ Sex + Age, family = "gaussian",
                   stringIndexerOrderType = "alphabetDesc")
summary(model3)

# fit tweedie model
model <- df %>% ml_glm(Freq ~ Sex + Age, family = "tweedie",
                       var.power = 1.2, link.power = 0)
summary(model)

# use the tweedie family from statmod
library(statmod)
model <- ml_glm(df, Freq ~ Sex + Age, family = tweedie(1.2, 0))
summary(model)

## End(Not run)

danzafar/tidyspark documentation built on Sept. 30, 2020, 12:19 p.m.