Brian Hulette

Run Python Examples_Direct

Run Python Examples_Dataflow

Run Python 3.8 PostCommit

PythonDocs PreCommit has passed (https://ci-beam.apache.org/job/beam_PreCommit_PythonDocs_Commit/9575/), merging

I think this would be difficult to do in a general (cross-ModelHandler) way as each ModelHandler is responsible for invoking it's model, and they currently have different ways of doing...

@y1chi do you have time to review this? It's pretty small :)

I pulled out some data from the exception. Here is the assertion: ``` Unable to match metrics for matcher name: 'ElementCount' (label_key: 'output_user_name' label_value: 'GroupByKey/Reify-out0'). (label_key: 'original_name' label_value: 'GroupByKey/Reify-out0-ElementCount'). attempted:...

But this would disable pruning within pipelines that use the DataFrame API. I'd consider that just a temporary workaround. Shouldn't we investigate what's causing these transforms to be incorrectly pruned?

> It just skips the pruning for a specific transform, not the whole pruning feature. Right, but all usages of the DataFrame API build a single composite DataFrameTransform. These usages...

Re-opening this to track the follow-up work discussed in https://github.com/apache/beam/pull/23069#issuecomment-1245899612 With #23069 we just disabled pruning for DataFrames, we should find a way to correctly prune them.