tomasatdatabricks

Results 6 comments of tomasatdatabricks

Update scoring api.

Thanks for the review @dbczumar ! I am planning to update the docs, but I am still fixing tests :/ The PR grew unexpectedly in scope as I am working...

[FR] Add Pandas category dtype to mlflow.types.schema

Possibly yes, but it needs a bit of work. The tricky part about categoricals is that the dictionary can be different between training and inference data. So you need to...

[FR] Expose the current model from mlserver `/inference` endpoint

@BenWilson2 @dingobar Note that the PR 6575 is against 2.0 branch - it will be released with mlflow 2.0. and we're not planning on porting it to mlflow 1.x. MLServer...

[FR] Add Pandas category dtype to mlflow.types.schema

@dbczumar seems reasonable to support this based on the feedback.

[WIP - do not merge!] Move sparkdl utilities for conversion between numpy arrays and image schema to ImageSchema

@sueann Yes I agree, I would merge spark version first and merge this one only after spark 2.4 is released. I made the PR here mostly because that's what we...

[BUG] mlflow.spark Model predictions cannot reliably be matched to the input data

Hi @vrr187, I think that's because the path exists only on the driver and not on the workers. Spark ML models are loaded in a distributed task so the exact...