Ce Gao

envd is a **machine learning development environment** for data science and AI/ML engineering teams. 🐍 **No Docker, only Python** - Focus on writing Python code, we will take care of...

[request] Avoid hard code

1

https://github.com/liamcli/randomNAS_release/blob/master/searchers/random_weight_share.py#L2 hard coded the path of the dep, which caused an error when I run it.

Our project contains one job for unit test and the other one for e2e. We want to merge the coverage result, then send them to coveralls. The current [code](https://github.com/tensorchord/envd/blob/main/.github/workflows/CI.yml) looks...

[feature request] Add Go Client

1

Hi I am one of the contributors of Kubeflow Katib, we are using modeldb as storage backend to store models. Currently, we maintain a go client https://github.com/kubeflow/katib/tree/master/pkg/manager/modelstore/modeldb, which is auto-generated...

[question] MPIDistribution is missing in v1 CRD

10

MPIDistribution is missing in v1 but it exists in v1alpha2. Did we remove it? /cc @terrytangyuan @alculquicondor @zw0610

Run tests at merge time

1

> Have you considered validations at merge time, in addition to the time of the PR being created? https://github.com/kubeflow/mpi-operator/pull/387

Now we support Horovod elastic, but we need to support scale subresource for HPA

Ref https://github.com/kubeflow/mpi-operator/pull/332#discussion_r587031118

[feature request] Support elastic

11

https://github.com/horovod/horovod/blob/master/docs/elastic.rst It will be better if we support elastic training.

https://github.com/kubeflow/tf-operator/issues/1241#issuecomment-775607747 CleanPodPolicy now keeps in RunPolicy, I think we should keep API consistency with it.