search

aws-samples / aws-efa-eks

Deploying EFA in EKS utilizing GPUDirectRDMA where supported
MIT No Attribution
35 stars 19 forks source link

Cannot run bert-efa-example example #2

Open iankouls-aws opened 3 years ago

iankouls-aws commented 3 years ago

The example bert-efa-example.yaml refers to a private image 231748552833.dkr.ecr.us-east-1.amazonaws.com/bert-efa-hvd-tf1-py3:latest and gets ImagePullBackoff due to lack of permissions.

Here is the line number for reference: https://github.com/aws-samples/aws-efa-eks/blob/main/examples/bert-efa-example.yaml#L20

If possible, please provide a public version of this image so the example is reproducible. Thank you!

anyingl commented 1 year ago

Hello! I'm also having this issue with almost all of the Pytorch examples: https://github.com/aws-samples/aws-efa-eks/issues/2. It'd be great if these could be pushed to a public repo.