-
Notifications
You must be signed in to change notification settings - Fork 1
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
feat: Fix Scatterplot #529
Comments
I would add
|
Some comments:
Can this be overridden in the settings, with the above behavior only being the default?
In its current functionality, you can specify columns that are shown in the tooltip. Are we removing this feature again? |
No, we are not removing this. To the contrary, Moritz is currently working on major refactorings of the scatterplot, and this ticket is here to make sure all features find their way back |
Some fundamental design drawbacks make this refactoring necessary. Moritz can explain the technical details. @thinkh, @dvmartinweigl, and @oltionchampari were involved in the discussion and decision. |
Currently the scatterplot has severe problems as described in #470
Step 1 is to collect a list of requirements for the scatter plot that we want to support in accordance with @dg-datavisyn
Should split up the data by a categorical attribute and create one plot for each category. The axis of the plots should be synced (vertical & horizontal). Each plot has its own selection which are merged in the vis config. So one could select points from one plot and add points from another plot by holding shift/ctrl, but it would also be possible to double click any plot to clear the selection.
Should create a scatterplot matrix as soon as we have 3 or more numerical attributes. Each plot should show the data for the attribute in its column/row. Selections are synced across the plots. Double clicking any plot clears the selection.
We want to color by numerical/categorical attributes. In the faceting and matrix case the color scale is global, ensuring that the categories have the same color and the numerical relationship between plots is preserved.
Analog to the color channel (in contrast to the current plot it should also be supported in faceting and matrix mode)
Analog to the color channel
Correct me @dg-datavisyn but this should just add the attribute to the hover tooltip?
Clear what it does
Shows a regression line and an overlay for the coefficients. Should also work in the faceting/matrix mode
The text was updated successfully, but these errors were encountered: