Github user sethah commented on a diff in the pull request:
https://github.com/apache/spark/pull/10384#discussion_r48062304
--- Diff:
mllib/src/test/scala/org/apache/spark/mllib/evaluation/RegressionMetricsSuite.scala
---
@@ -22,91 +22,111 @@ import
org.apache.spark.mllib.util.MLlibTestSparkContext
import org.apache.spark.mllib.util.TestingUtils._
class RegressionMetricsSuite extends SparkFunSuite with
MLlibTestSparkContext {
+ val obs = List[Double](77, 85, 62, 55, 63, 88, 57, 81, 51)
+ val eps = 1E-5
test("regression metrics for unbiased (includes intercept term)
predictor") {
/* Verify results in R:
- preds = c(2.25, -0.25, 1.75, 7.75)
- obs = c(3.0, -0.5, 2.0, 7.0)
+ y = c(77, 85, 62, 55, 63, 88, 57, 81, 51)
+ x = c(16, 22, 14, 10, 13, 19, 12, 18, 11)
+ df <- as.data.frame(cbind(x, y))
+ model <- lm(y ~ x, data=df)
+ preds <- signif(predict(model), digits = 4)
- SStot = sum((obs - mean(obs))^2)
- SSreg = sum((preds - mean(obs))^2)
- SSerr = sum((obs - preds)^2)
+ cat("predictions: ", preds, "\n")
+ cat("explainedVariance =", mean((preds - mean(y))^2), "\n")
+ cat("meanAbsoluteError =", mean(abs(preds - y)), "\n")
+ cat("meanSquaredError =", mean((preds - y)^2), "\n")
+ cat("rmse =", sqrt(mean((preds - y)^2)), "\n")
+ cat("r2 =", summary(model)$r.squared, "\n")
- explainedVariance = SSreg / length(obs)
- explainedVariance
- > [1] 8.796875
- meanAbsoluteError = mean(abs(preds - obs))
- meanAbsoluteError
- > [1] 0.5
- meanSquaredError = mean((preds - obs)^2)
- meanSquaredError
- > [1] 0.3125
- rmse = sqrt(meanSquaredError)
- rmse
- > [1] 0.559017
- r2 = 1 - SSerr / SStot
- r2
- > [1] 0.9571734
+ Output of R code:
+ predictions: 72.08 91.88 65.48 52.28 62.18 81.98 58.88 78.68 55.58
+ explainedVariance = 157.3
+ meanAbsoluteError = 3.735556
+ meanSquaredError = 17.53951
+ rmse = 4.18802
+ r2 = 0.8996822
*/
- val predictionAndObservations = sc.parallelize(
- Seq((2.25, 3.0), (-0.25, -0.5), (1.75, 2.0), (7.75, 7.0)), 2)
+ val preds = List(72.08, 91.88, 65.48, 52.28, 62.18, 81.98, 58.88,
78.68, 55.58)
+ val pairs: Seq[(Double, Double)] = preds.zip(obs)
+ val predictionAndObservations = sc.parallelize(pairs, 2)
val metrics = new RegressionMetrics(predictionAndObservations)
- assert(metrics.explainedVariance ~== 8.79687 absTol 1E-5,
+ assert(metrics.explainedVariance ~== 157.3 absTol eps,
"explained variance regression score mismatch")
- assert(metrics.meanAbsoluteError ~== 0.5 absTol 1E-5, "mean absolute
error mismatch")
- assert(metrics.meanSquaredError ~== 0.3125 absTol 1E-5, "mean squared
error mismatch")
- assert(metrics.rootMeanSquaredError ~== 0.55901 absTol 1E-5,
+ assert(metrics.meanAbsoluteError ~== 3.735556 absTol eps, "mean
absolute error mismatch")
+ assert(metrics.meanSquaredError ~== 17.53951 absTol eps, "mean squared
error mismatch")
+ assert(metrics.rootMeanSquaredError ~== 4.18802 absTol eps,
"root mean squared error mismatch")
- assert(metrics.r2 ~== 0.95717 absTol 1E-5, "r2 score mismatch")
+ assert(metrics.r2 ~== 0.8996822 absTol eps, "r2 score mismatch")
}
test("regression metrics for biased (no intercept term) predictor") {
/* Verify results in R:
- preds = c(2.5, 0.0, 2.0, 8.0)
- obs = c(3.0, -0.5, 2.0, 7.0)
+ y = c(77, 85, 62, 55, 63, 88, 57, 81, 51)
+ x = c(16, 22, 14, 10, 13, 19, 12, 18, 11)
+ df <- as.data.frame(cbind(x, y))
+ model <- lm(y ~ 0 + x, data=df)
+ preds <- signif(predict(model), digits = 4)
- SStot = sum((obs - mean(obs))^2)
- SSreg = sum((preds - mean(obs))^2)
- SSerr = sum((obs - preds)^2)
+ cat("predictions: ", preds, "\n")
+ cat("explainedVariance =", mean((preds - mean(y))^2), "\n")
+ cat("meanAbsoluteError =", mean(abs(preds - y)), "\n")
+ cat("meanSquaredError =", mean((preds - y)^2), "\n")
+ cat("rmse =", sqrt(mean((preds - y)^2)), "\n")
+ cat("r2 =", summary(model)$r.squared, "\n")
- explainedVariance = SSreg / length(obs)
- explainedVariance
- > [1] 8.859375
- meanAbsoluteError = mean(abs(preds - obs))
- meanAbsoluteError
- > [1] 0.5
- meanSquaredError = mean((preds - obs)^2)
- meanSquaredError
- > [1] 0.375
- rmse = sqrt(meanSquaredError)
- rmse
- > [1] 0.6123724
- r2 = 1 - SSerr / SStot
- r2
- > [1] 0.9486081
+ Output of R code:
--- End diff --
We should stick with the convention of declaring R code as if it were
computed in an R shell.
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
