Hi everyone,
Simple question… does Plotly (plotly.js specifically) have the capability to arbitrarily scale an axis to support P-P plots?
This is an example screenshot from Matplotlib:
I’m having the same question. Does anyone knows if is it possible?
Sorry to come here again, but I do really need to know if somebody has the answer to this question. I’ve googled alot and readed all the documentation, and didn’t find a solution. Does anyone have any idea if it’s possible?
1 Like
Hi. Has anyone found a solution to this question yet? Or a workaround. In matplotlib, you can simply select ‘probscale’ for an axis.
Sincerely
Anurag
I don’t think this is implemented in core plotly - axis types supported are linear, log, date/datetime and categorical. And I’ve not found any helper libraries, either in Python or js.
It should be possible to work round it, by transforming your probability values to standard normal quantiles, and then changing the axis appearance to look like probabilities (by customising tickvals and ticktext), and using your untransformed probability values in hovertext customdata.
(I could probably produce some Python code if that would be a useful starting point)
For anyone else who’s interested in background on this, the matplotlib capabilities are described here:
https://matplotlib.org/mpl-probscale/tutorial/closer_look_at_viz.html
Maybe this will help you