Github user viirya commented on a diff in the pull request: https://github.com/apache/spark/pull/8587#discussion_r43395797 --- Diff: sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/AggregationQuerySuite.scala --- @@ -556,6 +556,33 @@ abstract class AggregationQuerySuite extends QueryTest with SQLTestUtils with Te Row(0, null, 1, 1, null, 0) :: Nil) } + test("pearson correlation") { + val df = Seq.tabulate(10)(i => (1.0 * i, 2.0 * i, i * -1.0)).toDF("a", "b", "c") + val corr1 = df.repartition(2).groupBy().agg(corr("a", "b")).collect()(0).getDouble(0) + assert(math.abs(corr1 - 1.0) < 1e-12) + val corr2 = df.groupBy().agg(corr("a", "c")).collect()(0).getDouble(0) + assert(math.abs(corr2 + 1.0) < 1e-12) + // non-trivial example. To reproduce in python, use: + // >>> from scipy.stats import pearsonr + // >>> import numpy as np + // >>> a = np.array(range(20)) + // >>> b = np.array([x * x - 2 * x + 3.5 for x in range(20)]) + // >>> pearsonr(a, b) + // (0.95723391394758572, 3.8902121417802199e-11) + // In R, use: + // > a <- 0:19 + // > b <- mapply(function(x) x * x - 2 * x + 3.5, a) + // > cor(a, b) + // [1] 0.957233913947585835 + val df2 = Seq.tabulate(20)(x => (1.0 * x, x * x - 2 * x + 3.5)).toDF("a", "b") + val corr3 = df2.groupBy().agg(corr("a", "b")).collect()(0).getDouble(0) + assert(math.abs(corr3 - 0.95723391394758572) < 1e-12) + + val df3 = Seq.tabulate(0)(i => (1.0 * i, 2.0 * i)).toDF("a", "b") + val corr4 = df3.groupBy().agg(corr("a", "b")).collect()(0).getDouble(0) + assert(corr4.isNaN) + } --- End diff -- I will add ImplicitCastInputTypes to case class Corr. So the other NumericType can be automatically casting to double.
--- 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