Github user imatiach-msft commented on a diff in the pull request: https://github.com/apache/spark/pull/16699#discussion_r98300999 --- Diff: mllib/src/test/scala/org/apache/spark/ml/regression/GeneralizedLinearRegressionSuite.scala --- @@ -743,6 +743,84 @@ class GeneralizedLinearRegressionSuite } } + test("generalized linear regression with offset") { + /* + R code: + library(statmod) + df <- as.data.frame(matrix(c( + 1.0, 1.0, 2.0, 0.0, 5.0, + 2.0, 2.0, 0.5, 1.0, 2.0, + 1.0, 3.0, 1.0, 2.0, 1.0, + 2.0, 4.0, 0.0, 3.0, 3.0), 4, 5, byrow = TRUE)) + families <- list(gaussian, poisson, Gamma, tweedie(1.5)) + f1 <- V1 ~ -1 + V4 + V5 + f2 <- V1 ~ V4 + V5 + for (f in c(f1, f2)) { + for (fam in families) { + model <- glm(f, df, family = fam, weights = V2, offset = V3) + print(as.vector(coef(model))) + } + } + + [1] 0.535040431 0.005390836 + [1] 0.1968355 -0.2061711 + [1] 0.307996 -0.153579 + [1] 0.32166185 -0.09698986 + [1] -0.8800000 0.7342857 0.1714286 + [1] -1.9991044 0.7247511 0.1424392 + [1] -0.27378146 0.31599396 -0.06204946 + [1] -0.17118812 0.31200361 -0.02541656 + */ + val dataset = Seq( + OffsetInstance(1.0, 1.0, 2.0, Vectors.dense(0.0, 5.0)), + OffsetInstance(2.0, 2.0, 0.5, Vectors.dense(1.0, 2.0)), + OffsetInstance(1.0, 3.0, 1.0, Vectors.dense(2.0, 1.0)), + OffsetInstance(2.0, 4.0, 0.0, Vectors.dense(3.0, 3.0)) + ).toDF() + + val expected = Seq( + Vectors.dense(0.0, 0.535040431, 0.005390836), + Vectors.dense(0.0, 0.1968355, -0.2061711), + Vectors.dense(0.0, 0.307996, -0.153579), + Vectors.dense(0.0, 0.32166185, -0.09698986), + Vectors.dense(-0.88, 0.7342857, 0.1714286), + Vectors.dense(-1.9991044, 0.7247511, 0.1424392), + Vectors.dense(-0.27378146, 0.31599396, -0.06204946), + Vectors.dense(-0.17118812, 0.31200361, -0.02541656)) + + import GeneralizedLinearRegression._ + + var idx = 0 + for (fitIntercept <- Seq(false, true)) { + for (family <- Seq("gaussian", "poisson", "gamma", "tweedie")) { + var trainer = new GeneralizedLinearRegression().setFamily(family) --- End diff -- just a suggestion: maybe refactor the code below in a method and do fitIntercept.map(fi => family.map(fm => (fi, fm)).zip(expected).foreach(params => callMyMethod(params._1, params._2, params._3)) then you would get rid of the for loops and not have one long test case and you could remove the idx += 1 below
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org