cancel
Showing results for 
Search instead for 
Did you mean: 
Data Engineering
cancel
Showing results for 
Search instead for 
Did you mean: 

Forum Posts

hvsk
by New Contributor
  • 9178 Views
  • 2 replies
  • 0 kudos

Using a Virtual environment

Hi All,We are working on training NHits/TFT (a Pytorch-forecasting implementation) for timeseries forecasting. However, we are having some issues with package dependency conflicts.Is there a way to consistently use a virtual environment across cells ...

  • 9178 Views
  • 2 replies
  • 0 kudos
Latest Reply
Anonymous
Not applicable
  • 0 kudos

Hi @Harsh Kalra​ Hope everything is going great.Just wanted to check in if you were able to resolve your issue. If yes, would you be happy to mark an answer as best so that other members can find the solution more quickly? If not, please tell us so w...

  • 0 kudos
1 More Replies
Smu_Tan
by New Contributor
  • 1442 Views
  • 3 replies
  • 1 kudos

Resolved! Does Databricks supports the Pytorch Distributed Training for multiple devices?

Hi, Im trying to use the databricks platform to do the pytorch distributed training, but I didnt find any info about this. What I expected is using multiple clusters to run a common job using pytorch distributed data parallel (DDP) with the code belo...

  • 1442 Views
  • 3 replies
  • 1 kudos
Latest Reply
axb0
New Contributor III
  • 1 kudos

With Databricks MLR, HorovodRunner is provided which supports distributed training and inference with PyTorch. Here's an example notebook for your reference: PyTorchDistributedDeepLearningTraining - Databricks.

  • 1 kudos
2 More Replies
Alex_Persin
by New Contributor II
  • 3221 Views
  • 2 replies
  • 2 kudos

How can the shared memory size (/dev/shm) be increased on databricks worker nodes with custom docker images?

PyTorch uses shared memory to efficiently share tensors between its dataloader workers and its main process. However in a docker container the default size of the shared memory (a tmpfs file system mounted at /dev/shm) is 64MB, which is too small to ...

  • 3221 Views
  • 2 replies
  • 2 kudos
Latest Reply
mstuder
New Contributor II
  • 2 kudos

Also interested in increasing shared memory for use with ray

  • 2 kudos
1 More Replies
Labels