Jiaxin Shan

> @Jeffwan Is it a problem that dependabot assigns the approvers and reviewers? The functionality can easily be removed, but it does cause the script to be different that in...

@DavidSpek Are you using chainer now? This repo is inactive and we would like to get more feedbacks from users and then determine what's the next step? This probably needs...

@DavidSpek Thanks. We may do some clean up on these repos. I think this repo is inactive and we can skip Dependabot for this one. I talked with few folks...

> @Jeffwan We faced with the same problem in Katib. > We currently using `FashionMNIST ` instead of `MNIST`: https://github.com/kubeflow/katib/blob/master/examples/v1beta1/pytorch-mnist/mnist.py#L137. > I believe it hosts in the PyTorch servers. Sounds...

Code has been changed https://github.com/kubeflow/pytorch-operator/pull/327 We need a better way to publish images. This can be done after 1.3 release

@wuchen03 Can you rebase the master changes to pass tests? I can approve the change later

@johnugeorge Yeah, If we can not catch 1.1 release, I think we can postpone to next release. As TF and Pytorch already graduate to v1, we definitely need more test...

@MATRIX4284 Can you rebase master changes? We recently update test pipeline. Once rebase is done, test should pass

`809251082950.dkr.ecr.us-west-2.amazonaws.com/pytorch-operator` is used for internal testing. Once we move to public registry, we will make a change. It's been changed to use GCR in master now.

I wrote the pytorch ealstic controller in `pytorch/elastic` repo. Overall, it's very similar to existing operators. However, we can not use it seamlessly here because it requires a strong consistency...