GitHub user mandar2812 opened a pull request:
https://github.com/apache/spark/pull/5503
[MLLIB][WIP] SPARK-4638: Kernels feature for MLLIB
1) Class hierarchy for SVM Kernels, with unit tests.
2) Entropy based subset selection for low rank approximation of Large
Kernel Matrices, with unit tests.
3) Kernels for density estimation, with 'plug in' based optimum bandwidth
selection, with unit tests.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/mandar2812/spark kernels
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/5503.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #5503
----
commit 9d7b00b08423dae455e57e1f00fa6471f8fd901c
Author: mandar2812 <[email protected]>
Date: 2014-12-22T01:47:54Z
Initial commit for Kernels feature: Contains class heirarchies,
implementation of the Nystrom method for feature map extractions, RBF and
Polynomial Kernels. Also a bare bones test suite for SVM Kernels is included
commit 2e29b123665b19f3f06d5f4f90d7884646da71a6
Author: mandar2812 <[email protected]>
Date: 2015-01-21T17:34:29Z
1) Changes to class hierarchy 2) Commenting Nystrom approximation code
commit 5239082426911032df7228163a94eca8281e5676
Author: mandar2812 <[email protected]>
Date: 2015-01-27T16:03:26Z
Added optimal bandwidth selection procedure for Gaussian Density Kernel
commit bf1e9829b798ca76d408f6d1a66220720948c038
Author: mandar2812 <[email protected]>
Date: 2015-01-29T13:42:53Z
Entropy based subset selection done, unit tests passing
commit 43f85c20f1b8ca8ed7324f828df3a88295d26a3d
Author: mandar2812 <[email protected]>
Date: 2015-01-31T23:26:03Z
1) Optimized code for feature map extraction, kernel matrix multiplication
2) Code indentation changes
commit 025d214802621e6a8f8b05732723bbd0025f136d
Author: mandar2812 <[email protected]>
Date: 2015-02-01T13:29:27Z
Minor edits to kernel test suite.
commit 98d7730270d804e2140741ee0154cd03c2dde350
Author: mandar2812 <[email protected]>
Date: 2014-12-22T01:47:54Z
Initial commit for Kernels feature: Contains class heirarchies,
implementation of the Nystrom method for feature map extractions, RBF and
Polynomial Kernels. Also a bare bones test suite for SVM Kernels is included
commit bc0209b1b1d0e72e1335bb349ec04dba2a306864
Author: mandar2812 <[email protected]>
Date: 2015-01-21T17:34:29Z
1) Changes to class hierarchy 2) Commenting Nystrom approximation code
commit f8baf1e7da0ab6d71898f150d6b01cf3f38b2baf
Author: mandar2812 <[email protected]>
Date: 2015-01-27T16:03:26Z
Added optimal bandwidth selection procedure for Gaussian Density Kernel
commit 1fedafdbe7794dcd9a34f6a7317b1ff5e936c1f1
Author: mandar2812 <[email protected]>
Date: 2015-01-29T13:42:53Z
Entropy based subset selection done, unit tests passing
commit 6b8d8db8466b8795f8710d8861fc8f1ed541543f
Author: mandar2812 <[email protected]>
Date: 2015-01-31T23:26:03Z
1) Optimized code for feature map extraction, kernel matrix multiplication
2) Code indentation changes
commit 9367dc14f672976781a9ea6bdf54d503104e6d2a
Author: mandar2812 <[email protected]>
Date: 2015-02-01T13:29:27Z
Minor edits to kernel test suite.
commit 7f4dfae134b915e9c882700c1df695dbaaaee489
Author: mandar2812 <[email protected]>
Date: 2015-04-13T14:54:28Z
Minor comment clean up
commit 6b95548db3bbf180c330ee69121b371bcbc0f83e
Author: mandar2812 <[email protected]>
Date: 2015-04-14T00:10:13Z
Scala style check changes
commit fdafe16790f4b11bdc2f2965c7b3bf4243031cd4
Author: mandar2812 <[email protected]>
Date: 2015-04-14T00:13:12Z
Merging from remote fork
----
---
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]
