Support Pandas 0.23.0 #153
Conversation
| cutoff_time['time'] = pd.to_numeric(cutoff_time['time']) | ||
| except (ValueError, TypeError): | ||
| raise TypeError("cutoff_time times must be a numeric") | ||
| else: |
There was a problem hiding this comment.
can we do an explicit check for DateTimeIndex here?
| pass_columns = [column_name for column_name in cutoff_time.columns[2:]] | ||
|
|
||
| if _check_time_type(cutoff_time['time'].iloc[0]) is None: | ||
| raise ValueError("cutoff_time time values must be datetime or numeric") | ||
| if cutoff_time['instance_id'].dtype == np.dtype('O') and target_entity.df[target_entity.index].dtype.name.find('int') > -1: |
There was a problem hiding this comment.
I think we can do without these 2 checks and conversations for now. let's assume the user has this fixed up before calling calculate feature matrix.
There was a problem hiding this comment.
These are to suppress warnings from the tests. Should we change the tests then? Or add in a filter to ignore pandas warnings when running the tests?
Codecov Report
@@ Coverage Diff @@
## master #153 +/- ##
==========================================
- Coverage 92.7% 92.61% -0.09%
==========================================
Files 72 72
Lines 7714 7718 +4
==========================================
- Hits 7151 7148 -3
- Misses 563 570 +7
Continue to review full report at Codecov.
|
|
Looks good to me. Merge it! |
Various changes to support Pandas 0.23. Issues related to the new ability to merge on a combination of index and column names, the need to supply a "sort" argument to
DataFrame.append, as well as some issues related to MultiIndex sorting.