【Spark ML】How to get access of the MLlib's LogisticRegressionWithSGD after 3.0.0?

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Lyx
Reply | Threaded
Open this post in threaded view
|

【Spark ML】How to get access of the MLlib's LogisticRegressionWithSGD after 3.0.0?

Lyx
Hi,
    I have updated my Spark to the version of 3.0.0,
and it seems that the LogisticRegressionWithSGD's  constructor is private within the mllib  package, 
So i can't initialize an instance of this class.

How can i still use this class rather than switch to the LogisticRegressionWithLBFGS ? 

Reply | Threaded
Open this post in threaded view
|

Re: 【Spark ML】How to get access of the MLlib's LogisticRegressionWithSGD after 3.0.0?

srowen
-dev

See the migration guide:
https://spark.apache.org/docs/3.0.0/ml-migration-guide.html

Use ml.LogisticRegression, which should still let you use SGD

On Tue, Sep 22, 2020 at 12:54 AM Lyx <[hidden email]> wrote:
>
> Hi,
>     I have updated my Spark to the version of 3.0.0,
> and it seems that the LogisticRegressionWithSGD's  constructor is private within the mllib  package,
> So i can't initialize an instance of this class.
>
> How can i still use this class rather than switch to the LogisticRegressionWithLBFGS ?
>

---------------------------------------------------------------------
To unsubscribe e-mail: [hidden email]