Support the versioning of artifacts and/or pipelines. #467
Comments
|
Riley or Pavel, do you are more details on this one? I am not exactly sure of what the scope would be. Is it about versions of the pipeline system? Or is it about versioned pipelines? Should it be about versioned pipelines in general instead of just the front-end? |
|
This is about versioned pipelines, not the system itself. Yes, this should not be frontend specific. |
|
Thanks @rileyjbauer. I am going to repurpose this issue to track the versioning of artifacts in general (not just in the UI). This seems closely related to the ongoing work on a Metadata store that keeps track of which artifacts were generated by which pipelines. |
vicaire
changed the title
vicaire
changed the title
|
I'm just checking out Pipelines at the moment but it's looking great, thanks for your work on it! With the current UI, it doesn't seem there is any way to edit or upload a new version of a pipeline. Considering this, how do you test your pipelines? As an example, I had a typo in the URI for a container in my pipeline, and fixing that one typo required deleting and re-uploading the whole thing. Is that the current workflow, and is it the plan to add some flexibility here? "Upload a new version" or "edit the config in the browser" would be some nice features to have :) |
|
@JohnPaton, yes, at the moment, modifying a pipeline requires uploading/executing a new pipeline. |
|
What about implementing a git clone from a url additionally to import from file ? |
vicaire
added
area/wide-impact
and removed
area/api
area/backend
area/frontend
area/sdk/client
labels
|
@Bobgy: Closing this issue. In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
…low repository, I ran into problems with the import failing because the module kubeflow.kubeflow could not be found (kubeflow#466) The issue is tracked kubeflow#467 Essentially dynamically adding the kubeflow/kubeflow/py (i.e. from inside python) didn't work Could be an issue with namespace packaging I spent hours trying to make dynamically importing the modules work but I couldn't figure it out Setting python path before invoking python appears to work. So as a workaround we have run_e2e_workflow.py shell out to e2e_tool which writes the workflow to a file. run_e2e_workflow.py then loads it from the file. Create a K8s job to run run_e2e_workflow.py to simulate what happens under prow Related to kubeflow/kubeflow#3035 migrate e2e tests off ksonnet
* Request Logging in Control Plane * change executor to logger in release scripts * change executor into logger * initial inference logger commit * initial Q version of inference logger * Add initial cloudevents code * add cloudevents and opencensus vendor additions * update cloudevents and create example * add log_type and sample * Add logging to transformers and explainers * Add validation tests * Allow default log url to be knative default broker * Bring inference logger annotations into single method * add model_url and update examples * refactor to logger * Update logger Dockerfile * Fix compile errors after rebase * Logger files, and test * Remove sample and add modelId * Updates from review * Updates from review * Updates from testing * Update from review comments * update workflow to replace executor with logger * regenerate tf2openapi * Updates from review * review updates * Updates from review
kubeflow#467] (kubeflow#468) * fix: update the api to latest kubeflow pipelines for katib sample[ Fixes kubeflow#467] * Interesting set of changes seemed to fix that error. * added return statements. * Looks like a bug in compiler forces us to do this. * move contrib/katib to sample/katib. * review feedback * Remove mpi operator based samples as it does not come installed by default. * resolve k8s client version conflict by rearranging the deps in notebook. * Contains all the outputs of a successfully executed notebook.
This is a high level work item to track work needed to support pipeline versioning, including what links versions together, identifying a version when creating runs, comparing versions.. etc.
The text was updated successfully, but these errors were encountered: