You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In talking with @haleyjeppson, I think we see a way forward to deal with stats. Haley has already contributed some thoughts, in the Rmd file on the ggplot2 and Vega-Lite parts. I think it remains to sort out the ggspec part. I am proposing this issue as a gathering-place for that discussion.
I propose that Haley establish a branch to mirror this issue, so that we will merge a PR when we are "done". We can pr_pull_upstream(), as new things are merged into master from testing, ggschema, etc.
What follows is just a transcription from my notebook on a first idea for how a stat would appear in a ggspec layer; it is meant as a starting point for continuing the Rmd file.
This is meant to illustrate "new" parts of the spec, using stat-bin as an example:
In talking with @haleyjeppson, I think we see a way forward to deal with stats. Haley has already contributed some thoughts, in the
Rmd
file on the ggplot2 and Vega-Lite parts. I think it remains to sort out the ggspec part. I am proposing this issue as a gathering-place for that discussion.I propose that Haley establish a branch to mirror this issue, so that we will merge a PR when we are "done". We can
pr_pull_upstream()
, as new things are merged into master from testing, ggschema, etc.What follows is just a transcription from my notebook on a first idea for how a stat would appear in a ggspec layer; it is meant as a starting point for continuing the
Rmd
file.This is meant to illustrate "new" parts of the spec, using stat-bin as an example:
The content of
stat_params
would depend on the class of thestat
.Also, I have modified some of this since talking with Haley, so I think we will have some discussions before "finalizing" anything.
The text was updated successfully, but these errors were encountered: