PERF-modin-project#7068: Provide shape_hint="column" for some more op…

…erations with Series (modin-project#7069)

Signed-off-by: Iaroslav Igoshev <iaroslav.igoshev@intel.com>

modin/core/dataframe/algebra/map.py

@@ -43,9 +43,7 @@ def register(cls, function, *call_args, **call_kwds):
 
         def caller(query_compiler, *args, **kwargs):
             """Execute Map function against passed query compiler."""
-            shape_hint = call_kwds.pop("shape_hint", None) or kwargs.pop(
-                "shape_hint", None
-            )
+            shape_hint = call_kwds.pop("shape_hint", None) or query_compiler._shape_hint
             return query_compiler.__constructor__(
                 query_compiler._modin_frame.map(
                     lambda x: function(x, *args, **kwargs), *call_args, **call_kwds

modin/core/storage_formats/base/query_compiler.py

@@ -1567,11 +1567,10 @@ def isin(self, values, ignore_indices=False, **kwargs):  # noqa: PR02
             Boolean mask for self of whether an element at the corresponding
             position is contained in `values`.
         """
-        shape_hint = kwargs.pop("shape_hint", None)
         if isinstance(values, type(self)) and ignore_indices:
             # Pandas logic is that it ignores indexing if 'values' is a 1D object
             values = values.to_pandas().squeeze(axis=1)
-        if shape_hint == "column":
+        if self._shape_hint == "column":
             return SeriesDefault.register(pandas.Series.isin)(self, values, **kwargs)
         else:
             return DataFrameDefault.register(pandas.DataFrame.isin)(

modin/core/storage_formats/pandas/query_compiler.py

@@ -1702,7 +1702,8 @@ def stack(self, level, dropna):
 
     # Map partitions operations
     # These operations are operations that apply a function to every partition.
-    def isin(self, values, ignore_indices=False, shape_hint=None):
+    def isin(self, values, ignore_indices=False):
+        shape_hint = self._shape_hint
         if isinstance(values, type(self)):
             # HACK: if we don't cast to pandas, then the execution engine will try to
             # propagate the distributed Series to workers and most likely would have
@@ -1975,7 +1976,10 @@ def astype(self, col_dtypes, errors: str = "raise"):
         # other query compilers may not take care of error handling at the API
         # layer. This query compiler assumes there won't be any errors due to
         # invalid type keys.
-        return self.__constructor__(self._modin_frame.astype(col_dtypes, errors=errors))
+        return self.__constructor__(
+            self._modin_frame.astype(col_dtypes, errors=errors),
+            shape_hint=self._shape_hint,
+        )
 
     def infer_objects(self):
         return self.__constructor__(self._modin_frame.infer_objects())

modin/experimental/core/storage_formats/hdk/query_compiler.py

@@ -814,7 +814,7 @@ def astype(self, col_dtypes, errors: str = "raise"):
             )
         return self.__constructor__(
             self._modin_frame.astype(col_dtypes),
-            self._shape_hint,
+            shape_hint=self._shape_hint,
         )
 
     def setitem(self, axis, key, value):

modin/pandas/base.py

@@ -1919,7 +1919,7 @@ def convert_dtypes(
             )
         )
 
-    def isin(self, values, **kwargs):  # noqa: PR01, RT01, D200
+    def isin(self, values):  # noqa: PR01, RT01, D200
         """
         Whether elements in `BasePandasDataset` are contained in `values`.
         """
@@ -1929,7 +1929,7 @@ def isin(self, values, **kwargs):  # noqa: PR01, RT01, D200
         values = getattr(values, "_query_compiler", values)
         return self.__constructor__(
             query_compiler=self._query_compiler.isin(
-                values=values, ignore_indices=ignore_indices, **kwargs
+                values=values, ignore_indices=ignore_indices
             )
         )
 

modin/pandas/dataframe.py

@@ -1107,12 +1107,6 @@ def insert(
 
         self._update_inplace(new_query_compiler=new_query_compiler)
 
-    def isin(self, values):  # noqa: PR01, RT01, D200
-        """
-        Whether elements in `DataFrame` are contained in `values`.
-        """
-        return super(DataFrame, self).isin(values)
-
     def isna(self):
         """
         Detect missing values.

modin/pandas/series.py

@@ -1180,12 +1180,6 @@ def info(
             show_counts=show_counts,
         )
 
-    def isin(self, values):  # noqa: PR01, RT01, D200
-        """
-        Whether elements in `Series` are contained in `values`.
-        """
-        return super(Series, self).isin(values, shape_hint="column")
-
     def isna(self):
         """
         Detect missing values.