You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by jaceklaskowski <gi...@git.apache.org> on 2016/08/25 14:49:37 UTC
[GitHub] spark pull request #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logisti...
GitHub user jaceklaskowski opened a pull request:
https://github.com/apache/spark/pull/14808
[SPARK-17156][ML][EXAMPLE] Add multiclass logistic regression Scala Example
## What changes were proposed in this pull request?
New example for `MultinomialLogisticRegression` ML algorithm.
## How was this patch tested?
local build + run-example
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/jaceklaskowski/spark SPARK-17156-multiclass-logreg-example
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/14808.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 #14808
----
commit ba5a4e2cc14e253ea3465d887c3cf5a2a9d82a80
Author: Jacek Laskowski <ja...@japila.pl>
Date: 2016-08-25T14:47:18Z
[SPARK-17156][ML][EXAMPLE] Add multiclass logistic regression Scala Example
----
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/14808
Test PASSed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/64423/
Test PASSed.
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by sethah <gi...@git.apache.org>.
Github user sethah commented on the issue:
https://github.com/apache/spark/pull/14808
https://github.com/apache/spark/pull/14834 is merged now. We did not implement a new API, but we can still update the logistic regression examples to show the new multiclass functionality.
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the issue:
https://github.com/apache/spark/pull/14808
Merged build finished. Test PASSed.
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by jaceklaskowski <gi...@git.apache.org>.
Github user jaceklaskowski commented on the issue:
https://github.com/apache/spark/pull/14808
@sethah Anything to improve the example. I'm open for suggestions and improve the example (to learn that part better). Thanks!
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by sethah <gi...@git.apache.org>.
Github user sethah commented on the issue:
https://github.com/apache/spark/pull/14808
@jaceklaskowski You can see https://github.com/apache/spark/pull/14834
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by jaceklaskowski <gi...@git.apache.org>.
Github user jaceklaskowski commented on the issue:
https://github.com/apache/spark/pull/14808
I don't mind adapting the example to the upcoming changes and I can work on it. But..
SPARK-17163 has no Fix Version/s field assigned so I'm reading it that it's not clear whether the change gets into the next release or not. Given 2.0.1 might be out in a couple of weeks I believe accepting the example now and changing it later (once SPARK-17163 gets merged) is a viable strategy. What do you think?
/cc @srowen @wangmiao1981
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by jaceklaskowski <gi...@git.apache.org>.
Github user jaceklaskowski commented on the issue:
https://github.com/apache/spark/pull/14808
Thanks @MLnick. What would you suggest with the PR then (after the comments from @sethah)? Please guide.
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by sethah <gi...@git.apache.org>.
Github user sethah commented on the issue:
https://github.com/apache/spark/pull/14808
We are trying to move quite fast on 17163. I will submit a PR for it soon (not sure how long review will take). The issue here is that the current example will show users how to use an estimator called `MultinomialLogisticRegression` which won't even exist after 17163. I'm not sure it's worth it, but I'm not sure what the goals are for minor releases either. I would appreciate others' opinions.
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by jaceklaskowski <gi...@git.apache.org>.
Github user jaceklaskowski commented on the issue:
https://github.com/apache/spark/pull/14808
Any news on that @sethah? I've seen some discussions about the changes to unify the interfaces, but am wondering how close the other PRs are so I could help with that one myself.
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by sethah <gi...@git.apache.org>.
Github user sethah commented on the issue:
https://github.com/apache/spark/pull/14808
This is going to have to be changed after [SPARK-17163](https://issues.apache.org/jira/browse/SPARK-17163). Sorry about the confusion! We'll still want to make an example with multiclass, though, so maybe we can reuse some of this :)
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark pull request #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logisti...
Posted by jaceklaskowski <gi...@git.apache.org>.
Github user jaceklaskowski closed the pull request at:
https://github.com/apache/spark/pull/14808
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/14808
**[Test build #64423 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/64423/consoleFull)** for PR 14808 at commit [`ba5a4e2`](https://github.com/apache/spark/commit/ba5a4e2cc14e253ea3465d887c3cf5a2a9d82a80).
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by sethah <gi...@git.apache.org>.
Github user sethah commented on the issue:
https://github.com/apache/spark/pull/14808
I think we should close this. The new example and the user guide should be updated against [SPARK-17239](https://issues.apache.org/jira/browse/SPARK-17239). @jaceklaskowski If you'd still like to do it, please let me know otherwise I am happy to do it. We should try to get this in soon. Thanks!
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by jaceklaskowski <gi...@git.apache.org>.
Github user jaceklaskowski commented on the issue:
https://github.com/apache/spark/pull/14808
@sethah I seem to have missed the other comment with the notes what has to be done to make the PR an example for the change. Sorry.
Since I'm very new to it and the only way to learn it better is to help with the PR here or the one that's coming, I'd like to be engaged one way or the other.
I'm gonna close this PR (to make it consistent JIRA-wise) and open another with the changes you've mentioned to have the example aligned with the changes and the requirements.
Thanks @sethah for your help! See you in the other PR...
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by sethah <gi...@git.apache.org>.
Github user sethah commented on the issue:
https://github.com/apache/spark/pull/14808
I think we should add an example for multiclass classification, to demonstrate that it works in 2.1+. We also need to update the documentation that says we only support binary.
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by MLnick <gi...@git.apache.org>.
Github user MLnick commented on the issue:
https://github.com/apache/spark/pull/14808
by the way multi-class LR will be in 2.1 not 2.0.1, so we do have some time to wait for #14834
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
[GitHub] spark issue #14808: [SPARK-17156][ML][EXAMPLE] Add multiclass logistic regre...
Posted by SparkQA <gi...@git.apache.org>.
Github user SparkQA commented on the issue:
https://github.com/apache/spark/pull/14808
**[Test build #64423 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/64423/consoleFull)** for PR 14808 at commit [`ba5a4e2`](https://github.com/apache/spark/commit/ba5a4e2cc14e253ea3465d887c3cf5a2a9d82a80).
* This patch passes all tests.
* This patch merges cleanly.
* This patch adds the following public classes _(experimental)_:
* ` case class Params(`
---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org