FEAT-modin-project#2375: code cleaning

Signed-off-by: Dmitry Chigarev <dmitry.chigarev@intel.com>
dchigarev · Dec 8, 2020 · bc921c8 · bc921c8
1 parent 7a6ac46
commit bc921c8
Show file tree

Hide file tree

Showing 4 changed files with 14 additions and 11 deletions.
diff --git a/modin/backends/pandas/query_compiler.py b/modin/backends/pandas/query_compiler.py
@@ -2561,7 +2561,11 @@ def groupby_agg_builder(df, by=None, drop=False):
                 if isinstance(internal_by_df, pandas.DataFrame):
                     df = pandas.concat(
                         [df]
-                        + [internal_by_df[[o for o in internal_by_df if o not in df]]],
+                        + [
+                            internal_by_df.iloc[
+                                :, ~internal_by_df.columns.isin(df.columns)
+                            ]
+                        ],
                         axis=1,
                     )
                     internal_by_cols = list(internal_by_df.columns)

diff --git a/modin/engines/base/frame/axis_partition.py b/modin/engines/base/frame/axis_partition.py
@@ -148,11 +148,11 @@ def apply(
             if not isinstance(other_axis_partition, list):
                 other_axis_partition = [other_axis_partition]
 
-            other_shape = np.zeros(len(other_axis_partition) + 1, dtype=np.int)
-            for i in range(1, len(other_axis_partition) + 1):
-                other_shape[i] = other_shape[i - 1] + len(
-                    other_axis_partition[i - 1].list_of_blocks
-                )
+            # (other_shape[i-1], other_shape[i]) will indicate slice
+            # to restore i axis partition
+            other_shape = np.cumsum(
+                [0] + [len(o.list_of_blocks) for o in other_axis_partition]
+            )
 
             return self._wrap_partitions(
                 self.deploy_func_between_two_axis_partitions(
@@ -273,7 +273,7 @@ def deploy_func_between_two_axis_partitions(
         # reshaping flattened `rt_parts` array into a frame with shape `other_shape`
         combined_axis = [
             pandas.concat(
-                [rt_parts[j] for j in range(other_shape[i - 1], other_shape[i])],
+                rt_parts[other_shape[i - 1] : other_shape[i]],
                 axis=axis,
                 copy=False,
             )

diff --git a/modin/engines/base/frame/data.py b/modin/engines/base/frame/data.py
@@ -1735,7 +1735,9 @@ def broadcast_apply_full_axis(
             None,
             None,
             dtypes,
-            validate_axes="all" if new_partitions.size != 0 else False,
+            validate_axes="all"
+            if any(o is None for o in [new_index, new_columns])
+            else False,
         )
 
     def _copartition(self, axis, other, how, sort, force_repartition=False):

diff --git a/modin/pandas/dataframe.py b/modin/pandas/dataframe.py
@@ -414,9 +414,6 @@ def groupby(
                 if len(external_by) == 0:
                     by = self[internal_by]._query_compiler
 
-                # if len(by) == 1 and isinstance(by[0], type(self._query_compiler)):
-                #     by = by[0]
-
                 drop = True
             else:
                 mismatch = len(by) != len(self.axes[axis])