jsleight

Apologies for the delay. I just finished releasing mleap v0.22.0 which is on spark 3.3.0. See the release notes for the full changelong. You should see the relevant artifacts on...

> Step 1: Create a Spark session > spark = SparkSession.builder\ > .config('spark.jars.packages', 'ml.combust.mleap:mleap-spark_2.12:0.19.0') \ > .getOrCreate() Your example is using mleap 0.19.0. Does this go away if you use...

Looks like the op [isn't serializing the impurities](https://github.com/combust/mleap/blob/master/mleap-spark/src/main/scala/org/apache/spark/ml/bundle/ops/regression/DecisionTreeRegressionOp.scala#L23) right now. Looking at what the `withImpurities` is doing, it seems that is extra meta-data that can aid in debugging, but that...

Yeah for sure. But I'd argue that serializing to mleap is for inference tasks. To do evaluation and introspection you could just ``` pipeline.save(path) pipeline.load(path) ``` using spark's built in...

Ah, in mleap you can do that exact process by altering the ops registry. We use it for xgboost in order to allow xgboost models to be serialized in different...

The latest tensorflow-java versions are only building for java 11 https://github.com/tensorflow/java/blob/master/README.md#tensorflowjava-version-support So downgrading to java 8 would prohibit us from upgrading tensorflow.

I'll look into the dockerhub at some point (we changed who does the releases and I was not aware we even had a dockerhub artifact). For your specific error I...

Can you tell us a bit about your environment. What mleap and spark version do you have?

Well, I can see that we aren't storing k in spark.ml.bundle.ops.feature.PcaOp, so storing then loading back into Spark will crash like you observed. Looking through changelogs on the files I...

Last I knew there are changes needed in mleap to support a newer scikit-learn version. I'm happy to help review/ship a PR if someone creates one. Would be great to...