Fix ess/rhat plot in plot_forest #1606
Conversation
|
Are the bokeh y-limits a bit off? We could share them? |
|
Yeah, it looks like they are a bit off. I will look into this. Thank You |
Codecov Report
@@ Coverage Diff @@
## main #1606 +/- ##
==========================================
- Coverage 90.92% 90.92% -0.01%
==========================================
Files 108 108
Lines 11645 11663 +18
==========================================
+ Hits 10588 10604 +16
- Misses 1057 1059 +2
Continue to review full report at Codecov.
|
|
@ahartikainen done!, Is there anything else which I need to change? Thanks |
|
@OriolAbril is there anything else that needs to be done here? Thanks |
| for i in range(idx): | ||
| axes[0, i].y_range._property_values[ | ||
| "start" | ||
| ] = -all_plotters[ # pylint: disable=protected-access | ||
| 0 | ||
| ].group_offset | ||
| axes[0, i].y_range._property_values["end"] = y_max # pylint: disable=protected-access |
There was a problem hiding this comment.
can this go above? I think the ymax definition does not depend on anything in between and can also be moved before the loop above. There already is a loop over axes, now that we loop here for the yaxis lims I think we can use the same loop that starts in line 175, mostly for clarity
Rishabh261998
force-pushed
the
plot_forest_fix
branch
from
e0b1785
to
0d95733
Compare
|
Thanks @Rishabh261998, sorry it took a while to get it merged |
* Fix ess/rhat plot in plot_forest * Aligned ess/rhat dots in forestplot bokeh * Minor changes Co-authored-by: Oriol Abril-Pla <[email protected]>
Description
To fix #1596
As it was seen in #1591 that for the
NoModellabeller theess/rhatdots were plotted for only one model for bothmatplotlibandbokeh.After modifying the code:
Matplotlib with
NoModellabeller:Bokeh:
Checklist