bschifferer

How about HugeCTR?

@viswa-nvidia @EvenOldridge We need to add following success criteria : * Analyze scaling factor by using multiple GPUs: If we go from 1x GPU -> 2x GPUs -> 4x GPUs...

I provided an example to show that Merlin Models work with horovods: https://github.com/NVIDIA-Merlin/models/pull/778 However, we need to address the points above + the bug ( https://github.com/NVIDIA-Merlin/dataloader/issues/75 ). In addition, we...

I provided following example based on the current code: https://github.com/NVIDIA-Merlin/models/pull/855 - as I am OOO next week. @edknv did a great job to provide the horovod functionality in Merlin Models....

Let's keep this as the main ticket. We want to collect metrics based on the criteo example: https://github.com/NVIDIA-Merlin/Merlin/issues/235

@sararb can you review the PR if the documentation is still accurate?

We have many duplication of this ticket. I think we decided on: - collect only a few metrics instead of metrics from all notebooks (too overwhelming) - focus on session-based...

@EvenOldridge @karlhigley @viswa-nvidia is this ticket still relevant?

I think we decided to collect the metrics here: https://nvidia.slack.com/archives/CVBDJUPEZ/p1679617586380369 and we continue the ticket with https://github.com/NVIDIA-Merlin/models/issues/1047 .

@viswa-nvidia I closed the ticket as there was no progress for a long time. Please reopen, if we should work on it.