[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-12 Thread srowen
Github user srowen commented on the issue: https://github.com/apache/spark/pull/14949 Oh, I get it now. That makes sense. If this were being applied to decision trees only, that would make sense and we could fix this up and document the meaning. I agree it only makes sense to return "

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-12 Thread MLnick
Github user MLnick commented on the issue: https://github.com/apache/spark/pull/14949 The original JIRA [SPARK-8069](https://issues.apache.org/jira/browse/SPARK-8069) refers to https://cran.r-project.org/web/packages/randomForest/randomForest.pdf. That R package calls it "cu

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-12 Thread srowen
Github user srowen commented on the issue: https://github.com/apache/spark/pull/14949 The problem is that it's called 'threshold' and not 'weight', and 'threshold' means something different. Is anyone suggesting that it was always meant as a 'weight', and/or has a reference for this t

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-12 Thread zhengruifeng
Github user zhengruifeng commented on the issue: https://github.com/apache/spark/pull/14949 I think both this change and current design are reasonable. And I personally prefer to current one which treat threshould as a kind of weight. --- If your project is set up for it, you can rep

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-12 Thread srowen
Github user srowen commented on the issue: https://github.com/apache/spark/pull/14949 Trying @holdenk or @mengxr maybe. I think this behavior should be changed because it doesn't match the common meaning of 'threshold', but I feel like I'm missing context about why it was done this wa

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-07 Thread srowen
Github user srowen commented on the issue: https://github.com/apache/spark/pull/14949 @jkbradley @zhengruifeng @MLnick I wonder if I could ask you for comments on this change? it's a behavior change, so not something I'd do lightly, but I do think it improves the semantics here. ---

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-03 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/14949 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 e

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-03 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/14949 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/64903/ Test PASSed. ---

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-03 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/14949 **[Test build #64903 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/64903/consoleFull)** for PR 14949 at commit [`2fa331e`](https://github.com/apache/spark/commit/

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-03 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/14949 **[Test build #64903 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/64903/consoleFull)** for PR 14949 at commit [`2fa331e`](https://github.com/apache/spark/commit/2

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-03 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/14949 Merged build finished. Test FAILed. --- 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 e

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-03 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/14949 Test FAILed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/64900/ Test FAILed. ---

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-03 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/14949 **[Test build #64900 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/64900/consoleFull)** for PR 14949 at commit [`08dbe43`](https://github.com/apache/spark/commit/

[GitHub] spark issue #14949: [SPARK-17057] [ML] ProbabilisticClassifierModels' predic...

2016-09-03 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/14949 **[Test build #64900 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/64900/consoleFull)** for PR 14949 at commit [`08dbe43`](https://github.com/apache/spark/commit/0