Adam Wróbel

This can be worked around by using older version `sagemaker-training==4.3.0` This means the Python 3.6 backwards incompatibility was introduced quite recently

@RobinLoxlyHood it does not work - can you confirm it works for you?

I would be also interested in this. I am wondering if we can do a workaround with regular `deploy` function in `mlflow.sagemaker` ? There is a `mode=mlflow.sagemaker.DEPLOY_ADD` or something like...

@cjolif I eventually managed to do a workaround. For each of model included in multimodel endpoint, you need to download from `mlflow` registry the model binary, but only those files...

@tonofll hey sorry for asking in an old topic, I am having issues even adding the JAR to the cluster. How did you do it?

Oh yeah I just noticed you switched to Maven Central from Spark Packages. In there, the latest is 2.3.0. I managed today to workaround this by just pasting the coordinates,...