Add Documentation for new Python Based Visualizations (v0.7) #1076
Conversation
|
/assign @sarahmaddox |
There was a problem hiding this comment.
Thanks @ajchili! This will be a very useful doc. I've requested some updates to improve clarity. I've also suggested that some of the info is targeted at developers of the Kubeflow system rather than users, and therefore probably belongs in a GitHub README rather than on the website. Let me know what you think.
| visualizations can be found [below](/docs/pipelines/sdk/ | ||
| python-based-visualizations/#how-to-create-predefined-visualizations). | ||
|
|
||
| ## How to Create Predefined Visualizations |
There was a problem hiding this comment.
This section looks to me like info for KFP developers rather than users. If that's so, it should go in the KFP repo on GitHub rather than on the website. And then add a bullet item at the bottom of the page in a Next steps section that lets people know they can add predefined types by contributing to KFP. (We keep the developer-focused docs separate from the user docs, so that users don't think they need to build Kubeflow before they can use it.)
There was a problem hiding this comment.
Moved to KFP and is a new PR awaiting review.
| weight = 80 | ||
| +++ | ||
|
|
||
| This page describes python based visualizations, how to create them, and how to |
There was a problem hiding this comment.
As a reader, I wonder whether these Python-based visualisations will eventually replace the visualisations described in the other doc or whether they provide an alternative method, or whether I'm free to use both methods at once, perhaps for different reasons. It'd be good to add a bit of explanation at the top to address these questions.
Also, add info on which version of Kubeflow (e.g. v0.6 and later?) supports the new visualization method.
There was a problem hiding this comment.
Waiting for feature to be merged fully before adding a version it is available in.
|
Thanks @sarahmaddox for the detailed review! I'll make the suggested changes and push them as soon as they are finished. |
* Added details for how to enable custom visualizations * Added details about python based visualizations vs. current offering
|
/assign @neuromage |
|
/assign @IronPan |
Version info currently leads to dead links as releases have not yet been made
ajchili
changed the title
|
/hold |
|
/hold cancel |
|
/hold Thanks! Since the doc now states that this feature will be available in Kubeflow v0.7 and Pipelines 0.1.28, neither of which is available yet, I've applied a hold to prevent this doc from being merged too early. |
|
@sarahmaddox it is ready for the final tech review. |
| 6. Within the card, select the **CUSTOM** visualization type then provide a | ||
| source, and any necessary arguments (the source and argument variables are | ||
| optional for custom visualizations). | ||
| 7. Provide the custom visualization code. |
There was a problem hiding this comment.
We should include some guidelines about what the Python code looks like. Recommendation: Copy any relevant bits from the contributor README, including:
- The requirement for Python 3.
- How to pass variables from Pipelines to the UI.
- Anything else that the Python developer needs to know - please consult the engineering team for these details.
- An example of the code for a simple visualization.
There was a problem hiding this comment.
Still need to address:
- Anything else that the Python developer needs to know - please consult the engineering team for these details.
- An example of the code for a simple visualization.
|
@neuromage and @IronPan Please would you review, and provide assistance where I've commented about the Python code samples needed. |
|
Overall LGTM. Adding some examples would be great as well. @ajchili we could just add the existing predefined visualizations as samples here I think. |
sarahmaddox
changed the title
|
Fixes #1212 |
|
We now have an RC for Kubeflow v0.7, which means we can publish these docs. /hold cancel |
|
/lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: sarahmaddox The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
This PR aims to add detailed documentation for the new Python Based Visualizations feature within KFP. Currently, this PR is awaiting the feature's release before exiting a WIP state. Suggestions and reviews are welcome and encouraged even though this is marked as WIP.
This change is