2015-07-20 23:49:38 -04:00
|
|
|
#
|
|
|
|
# Licensed to the Apache Software Foundation (ASF) under one or more
|
|
|
|
# contributor license agreements. See the NOTICE file distributed with
|
|
|
|
# this work for additional information regarding copyright ownership.
|
|
|
|
# The ASF licenses this file to You under the Apache License, Version 2.0
|
|
|
|
# (the "License"); you may not use this file except in compliance with
|
|
|
|
# the License. You may obtain a copy of the License at
|
|
|
|
#
|
|
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
#
|
|
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
# See the License for the specific language governing permissions and
|
|
|
|
# limitations under the License.
|
|
|
|
#
|
|
|
|
|
|
|
|
library(testthat)
|
|
|
|
|
|
|
|
context("MLlib functions")
|
|
|
|
|
|
|
|
# Tests for MLlib functions in SparkR
|
|
|
|
|
|
|
|
sc <- sparkR.init()
|
|
|
|
|
|
|
|
sqlContext <- sparkRSQL.init(sc)
|
|
|
|
|
|
|
|
test_that("glm and predict", {
|
|
|
|
training <- createDataFrame(sqlContext, iris)
|
|
|
|
test <- select(training, "Sepal_Length")
|
|
|
|
model <- glm(Sepal_Width ~ Sepal_Length, training, family = "gaussian")
|
|
|
|
prediction <- predict(model, test)
|
|
|
|
expect_equal(typeof(take(select(prediction, "prediction"), 1)$prediction), "double")
|
|
|
|
})
|
|
|
|
|
2015-11-05 19:34:10 -05:00
|
|
|
test_that("glm should work with long formula", {
|
|
|
|
training <- createDataFrame(sqlContext, iris)
|
|
|
|
training$LongLongLongLongLongName <- training$Sepal_Width
|
|
|
|
training$VeryLongLongLongLonLongName <- training$Sepal_Length
|
|
|
|
training$AnotherLongLongLongLongName <- training$Species
|
|
|
|
model <- glm(LongLongLongLongLongName ~ VeryLongLongLongLonLongName + AnotherLongLongLongLongName,
|
|
|
|
data = training)
|
|
|
|
vals <- collect(select(predict(model, training), "prediction"))
|
|
|
|
rVals <- predict(glm(Sepal.Width ~ Sepal.Length + Species, data = iris), iris)
|
|
|
|
expect_true(all(abs(rVals - vals) < 1e-6), rVals - vals)
|
|
|
|
})
|
|
|
|
|
2015-07-20 23:49:38 -04:00
|
|
|
test_that("predictions match with native glm", {
|
|
|
|
training <- createDataFrame(sqlContext, iris)
|
2015-07-27 20:17:49 -04:00
|
|
|
model <- glm(Sepal_Width ~ Sepal_Length + Species, data = training)
|
2015-07-20 23:49:38 -04:00
|
|
|
vals <- collect(select(predict(model, training), "prediction"))
|
2015-07-27 20:17:49 -04:00
|
|
|
rVals <- predict(glm(Sepal.Width ~ Sepal.Length + Species, data = iris), iris)
|
|
|
|
expect_true(all(abs(rVals - vals) < 1e-6), rVals - vals)
|
2015-07-20 23:49:38 -04:00
|
|
|
})
|
2015-07-28 17:16:57 -04:00
|
|
|
|
|
|
|
test_that("dot minus and intercept vs native glm", {
|
|
|
|
training <- createDataFrame(sqlContext, iris)
|
|
|
|
model <- glm(Sepal_Width ~ . - Species + 0, data = training)
|
|
|
|
vals <- collect(select(predict(model, training), "prediction"))
|
|
|
|
rVals <- predict(glm(Sepal.Width ~ . - Species + 0, data = iris), iris)
|
|
|
|
expect_true(all(abs(rVals - vals) < 1e-6), rVals - vals)
|
|
|
|
})
|
2015-07-30 19:15:43 -04:00
|
|
|
|
2015-09-25 03:43:22 -04:00
|
|
|
test_that("feature interaction vs native glm", {
|
|
|
|
training <- createDataFrame(sqlContext, iris)
|
|
|
|
model <- glm(Sepal_Width ~ Species:Sepal_Length, data = training)
|
|
|
|
vals <- collect(select(predict(model, training), "prediction"))
|
|
|
|
rVals <- predict(glm(Sepal.Width ~ Species:Sepal.Length, data = iris), iris)
|
|
|
|
expect_true(all(abs(rVals - vals) < 1e-6), rVals - vals)
|
|
|
|
})
|
|
|
|
|
2015-07-30 19:15:43 -04:00
|
|
|
test_that("summary coefficients match with native glm", {
|
|
|
|
training <- createDataFrame(sqlContext, iris)
|
2015-11-09 11:56:22 -05:00
|
|
|
stats <- summary(glm(Sepal_Width ~ Sepal_Length + Species, data = training, solver = "normal"))
|
|
|
|
coefs <- unlist(stats$Coefficients)
|
|
|
|
devianceResiduals <- unlist(stats$DevianceResiduals)
|
|
|
|
|
2015-07-30 19:15:43 -04:00
|
|
|
rCoefs <- as.vector(coef(glm(Sepal.Width ~ Sepal.Length + Species, data = iris)))
|
2015-11-09 11:56:22 -05:00
|
|
|
rStdError <- c(0.23536, 0.04630, 0.07207, 0.09331)
|
|
|
|
rTValue <- c(7.123, 7.557, -13.644, -10.798)
|
|
|
|
rPValue <- c(0.0, 0.0, 0.0, 0.0)
|
|
|
|
rDevianceResiduals <- c(-0.95096, 0.72918)
|
|
|
|
|
|
|
|
expect_true(all(abs(rCoefs - coefs[1:4]) < 1e-6))
|
|
|
|
expect_true(all(abs(rStdError - coefs[5:8]) < 1e-5))
|
|
|
|
expect_true(all(abs(rTValue - coefs[9:12]) < 1e-3))
|
|
|
|
expect_true(all(abs(rPValue - coefs[13:16]) < 1e-6))
|
|
|
|
expect_true(all(abs(rDevianceResiduals - devianceResiduals) < 1e-5))
|
2015-07-30 19:15:43 -04:00
|
|
|
expect_true(all(
|
2015-11-09 11:56:22 -05:00
|
|
|
rownames(stats$Coefficients) ==
|
2015-09-25 03:43:22 -04:00
|
|
|
c("(Intercept)", "Sepal_Length", "Species_versicolor", "Species_virginica")))
|
2015-07-30 19:15:43 -04:00
|
|
|
})
|
2015-11-04 11:28:33 -05:00
|
|
|
|
|
|
|
test_that("summary coefficients match with native glm of family 'binomial'", {
|
|
|
|
df <- createDataFrame(sqlContext, iris)
|
|
|
|
training <- filter(df, df$Species != "setosa")
|
|
|
|
stats <- summary(glm(Species ~ Sepal_Length + Sepal_Width, data = training,
|
|
|
|
family = "binomial"))
|
2015-11-09 11:56:22 -05:00
|
|
|
coefs <- as.vector(stats$Coefficients)
|
2015-11-04 11:28:33 -05:00
|
|
|
|
|
|
|
rTraining <- iris[iris$Species %in% c("versicolor","virginica"),]
|
|
|
|
rCoefs <- as.vector(coef(glm(Species ~ Sepal.Length + Sepal.Width, data = rTraining,
|
|
|
|
family = binomial(link = "logit"))))
|
2015-11-09 11:56:22 -05:00
|
|
|
rStdError <- c(3.0974, 0.5169, 0.8628)
|
|
|
|
rTValue <- c(-4.212, 3.680, 0.469)
|
|
|
|
rPValue <- c(0.000, 0.000, 0.639)
|
2015-11-04 11:28:33 -05:00
|
|
|
|
2015-11-09 11:56:22 -05:00
|
|
|
expect_true(all(abs(rCoefs - coefs[1:3]) < 1e-4))
|
|
|
|
expect_true(all(abs(rStdError - coefs[4:6]) < 1e-4))
|
|
|
|
expect_true(all(abs(rTValue - coefs[7:9]) < 1e-3))
|
|
|
|
expect_true(all(abs(rPValue - coefs[10:12]) < 1e-3))
|
2015-11-04 11:28:33 -05:00
|
|
|
expect_true(all(
|
2015-11-09 11:56:22 -05:00
|
|
|
rownames(stats$Coefficients) ==
|
2015-11-04 11:28:33 -05:00
|
|
|
c("(Intercept)", "Sepal_Length", "Sepal_Width")))
|
|
|
|
})
|
2015-11-10 00:06:01 -05:00
|
|
|
|
|
|
|
test_that("summary works on base GLM models", {
|
|
|
|
baseModel <- stats::glm(Sepal.Width ~ Sepal.Length + Species, data = iris)
|
|
|
|
baseSummary <- summary(baseModel)
|
|
|
|
expect_true(abs(baseSummary$deviance - 12.19313) < 1e-4)
|
|
|
|
})
|