Jeremy Lewi

Results 140 issues of Jeremy Lewi

Opening this issue to start a discussion about whether it would be worth investing to make it easy to run tensorflow agents K8s. For some inspiration you can look at...

enhancement

### Expected behavior Setting ephmeralStorage when using the kaniko builder should set ephmeral-storage for the kaniko container. ### Actual behavior I observe ephmeral-storage being changed for the init container but...

kind/feature-request
area/build
priority/p2
build/kaniko

The flux grafana dashboard shows the status of a deploy but not the commit. I think it would be very useful to show the commit as well if possible.

area/monitoring

/kind feature Our test suite for GCP currently doesn't include any coverage for KFServing. Here's our existing test grid: https://k8s-testgrid.appspot.com/sig-big-data#kubeflow-gcp-blueprints-periodic-1-1&group-by-hierarchy-pattern=%5B%5Cw-%5D%2B Do we have a python program or notebook that we...

area/testing
platform/gcp
kind/feature
area/kfctl

I think we will want conformance tests to establish a clear, technical understanding of what can be expected from a conformant distribution of Kubeflow. One question that comes up immediately...

kind/feature
priority/p1
lifecycle/frozen

Adding more people to the issue triage rotation is blocked on the fact that issue triagers need fairly expansive permissions in order to triage issues. Right now issue triagers need...

kind/feature
area/engprod
lifecycle/frozen

* tensorflow/federated#3109 indicates that Python 3.10 is required to avoid various import errors. * Update the documentation to indicate this. * https://github.com/tensorflow/federated/issues/3109

/kind process Per the governance discussions. All projects need to find WG sponsorship or they will eventually be archived. Opening this issue to begin discussions about what to do about...

kind/feature
kind/process
area/community

/kind feature A common task for fairing Kubeflow is building docker images in cluster. Right now we use ClusterBuilder and some of the other libraries to do this. I wonder...

kind/feature
priority/p1

/kind feature I think it would be useful in our SDK to have a python function to generate stackdriver logs given a filter experession. Basically this function https://github.com/kubeflow/testing/blob/95955e506e9d1dd1aab22ed0412644533a90fd7d/py/kubeflow/testing/argo_build_util.py#L239

kind/feature