Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Support dataset-only smart tag analysis #478

Merged
merged 10 commits into from
Mar 9, 2023

Conversation

JosephMarinier
Copy link
Contributor

@JosephMarinier JosephMarinier commented Mar 8, 2023

Resolve #433

Description:

The overall diff is pretty messy, so I recommend looking at the different commits individually.

I made a couple of decisions what we can debate:

  • Color the dataset-only bars (with no outcomes) dark blue (palette.secondary.dark) like other data visualizations.
  • Hide the pipeline-specific columns, including accuracy and some smart tag families.
  • Disable the Prediction and Outcome options from the menu (since hiding the options would result in a weird one-option menu).

Here is an example:
Screen Shot 2023-03-08 at 11 45 51 AM

For reference, here is how it looks with a pipeline:
Screen Shot 2023-03-08 at 11 46 04 AM

Checklist:

You should check all boxes before the PR is ready. If a box does not apply, check it to acknowledge it.

  • ISSUE NUMBER. You linked the issue number (Ex: Resolve #XXX).
  • PRE-COMMIT. You ran pre-commit on all commits, or else, you
    ran pre-commit run --all-files at the end.
  • USER CHANGES. The changes are added to CHANGELOG.md and the documentation, if they impact
    our users.
  • DEV CHANGES.
    • Update the documentation if this PR changes how to develop/launch on the app.
    • Update the README files and our wiki for any big design decisions, if relevant.
    • Add unit tests, docstrings, typing and comments for complex sections.

that way, skipping columns (for example defining columns 1 and 3) won't leave extra gaps in between.
- `classCount` to `row`, as it is of type `Row` and it's not necessarily a label/prediction class since we added per-outcome aggregation; and
- `classIndex` to `rowIndex` accordingly.
@gabegma
Copy link
Contributor

gabegma commented Mar 8, 2023

Awesome!!! A few comments on your questions:

  • I'm ok with the blue, but I think the color should be the same one as in the control panel when there is no pipeline. So either both are blue, or we pick the same dark grey (black?) for this viz too. Side note - is the same dark blue as in the dataset warnings for example?
  • I'm not sure about moving the tile before the pipeline quality. I absolutely get the reasoning, but I do feel when you have a pipeline selected, you would care/want to see the overall quality first. Curious what others think.
  • The rest is great.

Comment on lines -141 to +164
{isPipelineSelected(pipeline) && (
<WithDatasetSplitNameState
defaultDatasetSplitName={firstAvailableDatasetSplit}
render={(datasetSplitName, setDatasetSplitName) => (
<PreviewCard
title="Smart Tag Analysis"
to={`/${jobId}/dataset_splits/${datasetSplitName}/smart_tags${searchString}`}
description={smartTagsDescription}
<WithDatasetSplitNameState
defaultDatasetSplitName={firstAvailableDatasetSplit}
render={(datasetSplitName, setDatasetSplitName) => (
<PreviewCard
title="Smart Tag Analysis"
to={`/${jobId}/dataset_splits/${datasetSplitName}/smart_tags${searchString}`}
description={smartTagsDescription}
>
<Box
display="flex"
flexDirection="column" // Important for the inner Box to overflow correctly
maxHeight={DEFAULT_PREVIEW_CONTENT_HEIGHT}
>
<Box
display="flex"
flexDirection="column" // Important for the inner Box to overflow correctly
maxHeight={DEFAULT_PREVIEW_CONTENT_HEIGHT}
>
<SmartTagsTable
jobId={jobId}
pipeline={pipeline}
availableDatasetSplits={datasetInfo.availableDatasetSplits}
datasetSplitName={datasetSplitName}
setDatasetSplitName={setDatasetSplitName}
/>
</Box>
</PreviewCard>
)}
/>
)}
<SmartTagsTable
jobId={jobId}
pipeline={pipeline}
availableDatasetSplits={datasetInfo.availableDatasetSplits}
datasetSplitName={datasetSplitName}
setDatasetSplitName={setDatasetSplitName}
/>
</Box>
</PreviewCard>
)}
/>
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nothing changed here, except removing the {isPipelineSelected(pipeline) && (...)} around the card. I hate how Git deals with changes of indent.

@JosephMarinier JosephMarinier merged commit 9986fad into main Mar 9, 2023
@JosephMarinier JosephMarinier deleted the joseph/dataset-only-smart-tag-analysis branch March 9, 2023 16:44
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Make smart tag aggregation plot work for dataset only
3 participants