diff options
| author | Sean Owen <srowen@gmail.com> | 2014-07-30 08:55:15 -0700 |
|---|---|---|
| committer | Xiangrui Meng <meng@databricks.com> | 2014-07-30 08:55:15 -0700 |
| commit | ee07541e99f0d262bf662b669b6542cf302ff39c (patch) | |
| tree | 005f8a40502e5868cdcfcd4b9afc868f7951700c /docs | |
| parent | 7c5fc28af42daaa6725af083d78c2372f3d0a338 (diff) | |
| download | spark-ee07541e99f0d262bf662b669b6542cf302ff39c.tar.gz spark-ee07541e99f0d262bf662b669b6542cf302ff39c.tar.bz2 spark-ee07541e99f0d262bf662b669b6542cf302ff39c.zip | |
SPARK-2748 [MLLIB] [GRAPHX] Loss of precision for small arguments to Math.exp, Math.log
In a few places in MLlib, an expression of the form `log(1.0 + p)` is evaluated. When p is so small that `1.0 + p == 1.0`, the result is 0.0. However the correct answer is very near `p`. This is why `Math.log1p` exists.
Similarly for one instance of `exp(m) - 1` in GraphX; there's a special `Math.expm1` method.
While the errors occur only for very small arguments, given their use in machine learning algorithms, this is entirely possible.
Also note the related PR for Python: https://github.com/apache/spark/pull/1652
Author: Sean Owen <srowen@gmail.com>
Closes #1659 from srowen/SPARK-2748 and squashes the following commits:
c5926d4 [Sean Owen] Use log1p, expm1 for better precision for tiny arguments
Diffstat (limited to 'docs')
0 files changed, 0 insertions, 0 deletions