Xiangrui Meng

[Proposal] MLflow Pipelines (previously named as MLX)

18

## Proposal Summary (You can find [the latest version of the proposal in this Google doc](https://docs.google.com/document/d/1KDNxLvmS392LdNpqc-bU9EAyoTvv0U3FiB5vRMfQuzI/edit).) ## What are you trying to do? Articulate your objectives using absolutely no jargon....

Switch back to the official xgboost-predictor release after it supports xgboost 1.0.0

1

See https://github.com/combust/mleap/pull/686#discussion_r434005552

Java examples for scoring

7

Many users can only use Java in a production environment. It would be nice if we can provide an example project for scoring in Java. I saw some Java interfaces...

Prepare spark-tensorflow-distributor 0.1.0 release

4

* Use official Spark 3.0.0 release. * Bump version number to 0.1.0.

Support import notebooks in HTML format

2

We support export in HTML but not import. HTML format doesn't need `--language`, which is a required parameter.

Single-line ScalaDoc plus deprecated annotation would trigger a bug

1

We recently deprecated setLabelCol in ChiSqSelectorModel in Spark (https://github.com/apache/spark/pull/13823): ``` scala /** @group setParam */ @Since("1.6.0") @deprecated("labelCol is not used by ChiSqSelectorModel.", "2.0.0") def setLabelCol(value: String): this.type = set(labelCol, value)...

We updated Spark to use genjavadoc 0.10 and found the package private objects are not hidden with strictVisibility switch on. See https://issues.apache.org/jira/browse/SPARK-15006. cc: @jodersky

It is convenient for a developer to ship code in English and the cache data together to optimize speed, save cost, and more importantly reproduce the result. However, we need...

We should exclude them in the release and keep docker tests working.

#139 switches to sbt-release to update version numbers. We should make it manage the entire release so we can save some manual effort.