Github user orhankislal commented on a diff in the pull request:
https://github.com/apache/madlib/pull/179#discussion_r136677915
--- Diff: src/ports/postgres/modules/convex/utils_regularization.py_in ---
@@ -156,14 +211,21 @@ def __utils_normalize_data_grouping(y_decenter=True,
**kwargs):
"""
group_col = kwargs.get('grouping_col')
group_col_list = split_quoted_delimited_str(group_col)
- group_where_x = ' AND
'.join(['{tbl_data}.{grp}=__x__.{grp}'.format(grp=grp,
- **kwargs) for grp in group_col_list])
- group_where_y = ' AND
'.join(['{tbl_data}.{grp}=__y__.{grp}'.format(grp=grp,
- **kwargs) for grp in group_col_list])
+ x_mean_join_clause = ''
+ y_mean_join_clause = ''
+ if kwargs.get('x_mean_table'):
+ group_where_x = ' AND
'.join(['{tbl_data}.{grp}=__x__.{grp}'.format(grp=grp,
+ **kwargs) for grp in group_col_list])
+ x_mean_join_clause = "INNER JOIN {0} AS __x__ ON {1}".format(
+ kwargs.get('x_mean_table'), group_where_x)
+ if kwargs.get('y_mean_table'):
+ group_where_y = ' AND
'.join(['{tbl_data}.{grp}=__y__.{grp}'.format(grp=grp,
--- End diff --
We can use the new `_check_groups` function from utilities.
---
