[
https://issues.apache.org/jira/browse/CLIMATE-466?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14020465#comment-14020465
]
ASF GitHub Bot commented on CLIMATE-466:
----------------------------------------
Github user agoodm commented on a diff in the pull request:
https://github.com/apache/climate/pull/67#discussion_r13513738
--- Diff: ocw/plotter.py ---
@@ -61,8 +61,19 @@ def _nice_intervals(data, nlevs):
# Find the min and max levels by cutting off the tails of the
distribution
# This mitigates the influence of outliers
data = data.ravel()
- mnlvl = mstats.scoreatpercentile(data, 5)
- mxlvl = mstats.scoreatpercentile(data, 95)
+ mn = mstats.scoreatpercentile(data, 5)
+ mx = mstats.scoreatpercentile(data, 95)
+ #if there min less than 0 then
+ #put 0 in center of color bar
+ if mn < 0:
--- End diff --
This line should be changed to
```python
if mn < 0 and mx > 0:
```
because otherwise you will end up centering the contours at 0 when the
maximum value is negative. This results in superfluous contour levels.
> Plotter need better configuration for generating color bar
> ----------------------------------------------------------
>
> Key: CLIMATE-466
> URL: https://issues.apache.org/jira/browse/CLIMATE-466
> Project: Apache Open Climate Workbench
> Issue Type: Improvement
> Components: visualization
> Affects Versions: 0.3-incubating
> Reporter: Maziyar Boustani
> Assignee: Maziyar Boustani
> Fix For: 0.4
>
>
> In plotter.py function "_nice_intervals" need to understand the range of
> values in different cases and be able to make a correct color bar for plot.
> If the values all are above 0 then make color bar starts from min to max.
> But if the values are between negative and positive then always keep 0 in
> center of color bar.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
