Github user tillrohrmann commented on a diff in the pull request:
https://github.com/apache/flink/pull/715#discussion_r31011074
--- Diff: docs/libs/ml/optimization.md ---
@@ -160,10 +158,20 @@ mini-batch SGD will be performed.
<td><strong>RegularizationParameter</strong></td>
<td>
<p>
- The amount of regularization to apply. (Default
value:<strong>0</strong>)
+ The amount of regularization to apply. (Default
value:<strong>0.0</strong>)
+ </p>
+ </td>
+ </tr>
+ <tr>
+ <td><strong>PredictionFunction</strong></td>
+ <td>
+ <p>
+ Class that provides the prediction function, used to calculate
$\hat{y}$ based on the
+ weights and the example features, and the prediction gradient.
+ (Default value: <strong>LinearPrediction</strong>)
--- End diff --
The same applies to the loss functions and the regularization functions. I
know they are already described in the text, but maybe it is helpful to have
again a more doc-style listing of all available options with a description.
---
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.
---
