Aurick Qiao

Hi! I think integrating with Lightning would be a great idea. For your two questions: 1. Not sure about Lightning, but from our experience integrating with other frameworks, this is...

Thanks for the inquiry! It simply calculates the size of the cluster each individual solution in the genetic algorithm uses. If cluster auto-scaling is enabled, the policy will check for...

> Unsupported storageclass from available storageclasses [] It looks like your K8s might not have any storageclasses installed. AdaptDL requires a shared filesystem which can be used to store checkpoints...

@SHu0421 This error could be caused by a variety of reasons. You can start by checking `kubectl -n get all` (replacing `` with the namespace in which you installed the...

@gudiandian it sounds like it's related to the problem you are having in https://github.com/petuum/adaptdl/issues/124

Could you provide the output of `helm list` and `kubectl get all`?

It seems a bit like certain pairs (or all pairs?) of Kubernetes pods are not able to reach each other. Could you try: - `kubectl exec` into a pod and...

You'll want to create a pod running a container that can run `ping`.

The scheduler pods running on worker nodes is expected behavior. I suggest making sure that inter-pod networking in your Kubernetes cluster is correctly working first, e.g. by following https://projectcalico.docs.tigera.io/getting-started/kubernetes/hardway/test-networking. AdaptDL...

Hi, unfortunately we're not able to host the datasets for public access due to cost reasons and (for certain datasets like ImageNet) license reasons. However, all the datasets we used...