Closed pkaf closed 2 years ago
@pkaf thank you again for contributing to Yellowbrick! Since this is a pretty simple change; I think this will work well - the tests are failing because the baseline images are now incorrect; I'm going to regenerate the images and spot-check a few of them. I'll let you know when I do so that you can also view the new baseline images. Note that this means I need write access to your fork - I'll let you know if you need to do anything extra.
@bbengfort may be the _sharedlimits is worth keeping with True as a default value. With following code modification inside the draw function?
self.ax.set_xlim(min(min(y), min(y_pred)), max(max(y), max(y_pred)))
self.ax.set_ylim(self.ax.get_xlim())
to
if self.shared_limits is True:
self.ax.set_xlim(min(min(y), min(y_pred)), max(max(y), max(y_pred)))
self.ax.set_ylim(self.ax.get_xlim())
This will make the current test in-place still relevant. Suggestion?
Merging #1208 (33ca4af) into develop (838063e) will increase coverage by
0.00%
. The diff coverage is100.00%
.
@@ Coverage Diff @@
## develop #1208 +/- ##
========================================
Coverage 90.42% 90.42%
========================================
Files 90 90
Lines 5096 5097 +1
========================================
+ Hits 4608 4609 +1
Misses 488 488
Impacted Files | Coverage Δ | |
---|---|---|
yellowbrick/regressor/prediction_error.py | 93.54% <100.00%> (+0.10%) |
:arrow_up: |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update 838063e...33ca4af. Read the comment docs.
@pkaf that works for me - looks like the tests are passing; I'll handle the merge conflicts and we'll get this merged! Thanks again for your patience.
This fixes Issue #1193. X and Y axis limits are synced and set to overall max/min values of (X, Y).