diff options
| author | sethah <seth.hendrickson16@gmail.com> | 2017-03-25 17:41:59 +0000 |
|---|---|---|
| committer | DB Tsai <dbtsai@dbtsai.com> | 2017-03-25 17:41:59 +0000 |
| commit | be85245a98d58f636ff54956cdfde15ea5cd6122 (patch) | |
| tree | 06073069774144c9c5a6d36a62850f0cc0b35413 /sql | |
| parent | e8ddb91c7ea5a0b4576cf47aaf969bcc82860b7c (diff) | |
| download | spark-be85245a98d58f636ff54956cdfde15ea5cd6122.tar.gz spark-be85245a98d58f636ff54956cdfde15ea5cd6122.tar.bz2 spark-be85245a98d58f636ff54956cdfde15ea5cd6122.zip | |
[SPARK-17137][ML][WIP] Compress logistic regression coefficients
## What changes were proposed in this pull request?
Use the new `compressed` method on matrices to store the logistic regression coefficients as sparse or dense - whichever is requires less memory.
Marked as WIP so we can add some performance test results. Basically, we should see if prediction is slower because of using a sparse matrix over a dense one. This can happen since sparse matrices do not use native BLAS operations when computing the margins.
## How was this patch tested?
Unit tests added.
Author: sethah <seth.hendrickson16@gmail.com>
Closes #17426 from sethah/SPARK-17137.
Diffstat (limited to 'sql')
0 files changed, 0 insertions, 0 deletions