diff options
| author | Sean Owen <sowen@cloudera.com> | 2016-12-07 17:34:45 +0800 |
|---|---|---|
| committer | Sean Owen <sowen@cloudera.com> | 2016-12-07 17:34:45 +0800 |
| commit | 79f5f281bb69cb2de9f64006180abd753e8ae427 (patch) | |
| tree | a722ebc403cc4655e96e48b9e3de7502e04271a0 /R | |
| parent | b8280271396eb74638da6546d76bbb2d06c7011b (diff) | |
| download | spark-79f5f281bb69cb2de9f64006180abd753e8ae427.tar.gz spark-79f5f281bb69cb2de9f64006180abd753e8ae427.tar.bz2 spark-79f5f281bb69cb2de9f64006180abd753e8ae427.zip | |
[SPARK-18678][ML] Skewed reservoir sampling in SamplingUtils
## What changes were proposed in this pull request?
Fix reservoir sampling bias for small k. An off-by-one error meant that the probability of replacement was slightly too high -- k/(l-1) after l element instead of k/l, which matters for small k.
## How was this patch tested?
Existing test plus new test case.
Author: Sean Owen <sowen@cloudera.com>
Closes #16129 from srowen/SPARK-18678.
Diffstat (limited to 'R')
| -rw-r--r-- | R/pkg/inst/tests/testthat/test_mllib.R | 9 |
1 files changed, 5 insertions, 4 deletions
diff --git a/R/pkg/inst/tests/testthat/test_mllib.R b/R/pkg/inst/tests/testthat/test_mllib.R index 0802a2ae48..4758e40e41 100644 --- a/R/pkg/inst/tests/testthat/test_mllib.R +++ b/R/pkg/inst/tests/testthat/test_mllib.R @@ -1007,10 +1007,11 @@ test_that("spark.randomForest", { model <- spark.randomForest(data, Employed ~ ., "regression", maxDepth = 5, maxBins = 16, numTrees = 20, seed = 123) predictions <- collect(predict(model, data)) - expect_equal(predictions$prediction, c(60.379, 61.096, 60.636, 62.258, - 63.736, 64.296, 64.868, 64.300, - 66.709, 67.697, 67.966, 67.252, - 68.866, 69.593, 69.195, 69.658), + expect_equal(predictions$prediction, c(60.32820, 61.22315, 60.69025, 62.11070, + 63.53160, 64.05470, 65.12710, 64.30450, + 66.70910, 67.86125, 68.08700, 67.21865, + 68.89275, 69.53180, 69.39640, 69.68250), + tolerance = 1e-4) stats <- summary(model) expect_equal(stats$numTrees, 20) |