[ENH] Box Plot: Add box for missing group values#4292
[ENH] Box Plot: Add box for missing group values#4292ajdapretnar merged 2 commits intobiolab:masterfrom
Conversation
3f4d121 to
3542a0e
Compare
Codecov Report
@@ Coverage Diff @@
## master #4292 +/- ##
==========================================
+ Coverage 86.68% 86.77% +0.09%
==========================================
Files 396 396
Lines 71510 71549 +39
==========================================
+ Hits 61990 62090 +100
+ Misses 9520 9459 -61 |
737775e to
b7dd941
Compare
b7dd941 to
9da0bba
Compare
| self.dist = [] | ||
| self.conts = contingency.get_contingency( | ||
| dataset, attr, self.group_var) | ||
| group_var_labels = self.group_var.values + ["Missing value"] |
There was a problem hiding this comment.
I was thinking whether we could simplify the user experience by instead printing, say "missing thal" or "missing sepal length". Same for splitting by missing values. This would tell the user more clearly what those missing values are.
There was a problem hiding this comment.
I think it is a great idea. I added the names of the variables to the missing label.
9da0bba to
250c582
Compare
|
I have one trivial suggestion, which you may feel free to reject or ignore if no other changes are made. I think writing I would also put the attribute name into (single) quotes. |
250c582 to
2467507
Compare
|
@janezd I agree with you it looks nicer and also using a single quote more clearly shows that it is not the name of the value. It is modified. |

Issue
@lanzagar and I discussed that there should be also a box for missing values in subgropus for countinuous variables as it is done for discrete.
Description of changes
When the selected variable is continuous there is now box with missing values if they are present in the subgroup. It shows the distribution of values that are missing in the subgroups variable.
Includes