Even Oldridge

@sararb any updates here?

Datasets used will be example driven. Closing.

@marcromeyn @oliverholworthy what's the status of this? This is blocking the creation of example notebooks for end to end using systems.

@bschifferer @radekosmulski This would be a good example. How to prepare data for RecSys.

And instructions commented out for how to deal with this in 2.6 and lower.

https://github.com/NVIDIA-Merlin/Merlin/issues/471 has details on the customer request side.

We're working on a model evaluation framework that allows for slicing, but we don't have plans for this on our roadmap yet. @zhiruiwang is this something you'd be interested in...

@jperez999 @benfred @rjzamora It looks like batching isn't correctly splitting the datasets? Is this particular to multi-gpu or does the problem also occur on single GPU but it just doesn't...

@bschifferer can you follow up and make sure we're using HV correctly. We'll probably need to find a way to make clear to our customers how to properly set this...

Since this isn't a part of our stack I'm closing. @vonodiripsa were you able to resolve this through llvm and horovod?