Refine Frame._reduce_for_stat_function. #1975
Conversation
|
|
||
| return self._reduce_for_stat_function(var, name="var", axis=axis, numeric_only=numeric_only) | ||
|
|
||
| def median( |
There was a problem hiding this comment.
This is moved here for ease of type annotation; bool is overwritten later..
There was a problem hiding this comment.
Where is bool overwritten?
There was a problem hiding this comment.
koalas/databricks/koalas/generic.py
Line 1885 in 5b1205b
Codecov Report
@@ Coverage Diff @@
## master #1975 +/- ##
==========================================
- Coverage 94.60% 94.56% -0.05%
==========================================
Files 50 50
Lines 10905 10935 +30
==========================================
+ Hits 10317 10341 +24
- Misses 588 594 +6
Continue to review full report at Codecov.
|
|
|
||
| def max(self, axis=None, numeric_only=None) -> Union[Scalar, "Series"]: | ||
| def max( | ||
| self, axis: Union[int, str] = None, numeric_only: bool = None |
There was a problem hiding this comment.
Do we use numeric_only: bool = None rather than numeric_only: bool = False for pandas compatibility?
There was a problem hiding this comment.
Yes, I think so. cc @itholic who changed it to None.
There was a problem hiding this comment.
Yes, It's for pandas compatibility.
| """ | ||
|
|
||
| def count(spark_column, spark_type): | ||
| # Special handle floating point types because Spark's count treats nan as a valid value, |
There was a problem hiding this comment.
The refactoring is much cleaner.
|
Looks great to me! Left some questions. |
|
Thanks! I'd merge this now. |
|
Great. Thanks! |
4 participants
Refines
DataFrame/Series._reduce_for_stat_functionto avoid special handling based on a specific function.Also:
countand supportnumeric_onlyparameter.