Sean Owen ee07541e99 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		2014-07-30 08:55:15 -07:00
..
data	Fix all code examples in guide	2014-01-13 22:29:45 -08:00
src	SPARK-2748 [MLLIB] [GRAPHX] Loss of precision for small arguments to Math.exp, Math.log	2014-07-30 08:55:15 -07:00
pom.xml	[SPARK-2410][SQL] Merging Hive Thrift/JDBC server (with Maven profile fix)	2014-07-28 12:07:30 -07:00