[
https://issues.apache.org/jira/browse/MAHOUT-1856?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15833758#comment-15833758
]
ASF GitHub Bot commented on MAHOUT-1856:
----------------------------------------
Github user rawkintrevo commented on the issue:
https://github.com/apache/mahout/pull/246
Thanks for the review @andrewpalumbo !
[sklearn parameters are set when the `Estimator` is
instantiated.](http://scikit-learn.org/stable/tutorial/statistical_inference/settings.html).
MLlib on the other hand, [passes parameter maps in `fit` as you
suggest](https://spark.apache.org/docs/2.0.1/api/java/org/apache/spark/ml/Estimator.html#fit(org.apache.spark.sql.Dataset,%20org.apache.spark.ml.param.ParamMap))
BOTH however, allow hyper parameters to be updated. And in the case you
refer too, the model would not be re-instantiated, but something like this:
```scala
model.param1 = 1
model.fit(X, y)
model.param2 = 2
```
To your point, I also want to make this as easy as possible for new users-
so I think it would be best to leave the option to pass a parameter map at
initiation, and also expose it as a optional parameter of the `fit` method.
> Create a framework for new Mahout Clustering, Classification, and
> Optimization Algorithms
> ------------------------------------------------------------------------------------------
>
> Key: MAHOUT-1856
> URL: https://issues.apache.org/jira/browse/MAHOUT-1856
> Project: Mahout
> Issue Type: New Feature
> Affects Versions: 0.12.1
> Reporter: Andrew Palumbo
> Assignee: Trevor Grant
> Priority: Critical
> Fix For: 0.13.0
>
>
> To ensure that Mahout does not become "A loose bag of algorithms", Create
> basic traits with funtions common to each class of algorithm.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
