GitHub user dbtsai opened a pull request:
https://github.com/apache/spark/pull/40
Initialized the regVal for first iteration in SGD optimizer
Ported from https://github.com/apache/incubator-spark/pull/633
In runMiniBatchSGD, the regVal (for 1st iter) should be initialized
as sum of sqrt of weights if it's L2 update; for L1 update, the same logic
is followed.
It maybe not be important here for SGD since the updater doesn't take the
loss
as parameter to find the new weights. But it will give us the correct
history of loss.
However, for LBFGS optimizer we implemented, the correct loss with regVal
is crucial to
find the new weights.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/AlpineNow/spark dbtsai-smallRegValFix
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/40.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 #40
----
commit 9d2670330ebcde4240d97f0f51d7cfbc71509780
Author: DB Tsai <[email protected]>
Date: 2014-02-22T10:59:00Z
In runMiniBatchSGD, the regVal (for 1st iter) should be initialized
as sum of sqrt of weights if it's L2 update; for L1 update, the same logic
is followed.
It maybe not be important here for SGD since the updater doesn't take the
loss
as parameter to find the new weights. But it will give us the correct
history of loss.
However, for LBFGS optimizer we implemented, the correct loss with regVal
is crucial to
find the new weights.
commit 1c15cfebf8a130744ddea915dd457a8e4291bdfe
Author: DB Tsai <[email protected]>
Date: 2014-02-25T22:02:12Z
Added unittest for the change in runMiniBatchSGD
commit 594a288acd91dd9b426d9327f2b8f4db828321d3
Author: DB Tsai <[email protected]>
Date: 2014-02-27T09:37:39Z
Removed unnecessary parentheses, and fixed a typo.
----
---
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.
---
