cancel
Showing results for 
Search instead for 
Did you mean: 
Data Engineering
Join discussions on data engineering best practices, architectures, and optimization strategies within the Databricks Community. Exchange insights and solutions with fellow data engineers.
cancel
Showing results for 
Search instead for 
Did you mean: 

Forum Posts

Alex_Persin
by New Contributor III
  • 6997 Views
  • 6 replies
  • 8 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 ...

  • 6997 Views
  • 6 replies
  • 8 kudos
Latest Reply
stevewb
New Contributor II
  • 8 kudos

Bump again... does anyone have a solution for this?

  • 8 kudos
5 More Replies
hvsk
by New Contributor
  • 14707 Views
  • 4 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 ...

  • 14707 Views
  • 4 replies
  • 0 kudos
Latest Reply
User16502773013
Databricks Employee
  • 0 kudos

Hello @sshssh  Databricks support Notebook scoped environments for databricks runtime (regular and ML) Kindly please check blog post here for ways to install Notebook scoped libraries in databricks Regards

  • 0 kudos
3 More Replies
Smu_Tan
by New Contributor
  • 3805 Views
  • 6 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...

  • 3805 Views
  • 6 replies
  • 1 kudos
Latest Reply
-werners-
Esteemed Contributor III
  • 1 kudos

If only the driver is active, this probably means you are not using Spark.  When running pure python,... code, the driver will execute that.If Spark is active, workers receive their tasks from the driver.  Generally the driver is not that active, the...

  • 1 kudos
5 More Replies
Labels