API: value_counts consistently return int64 dtype

pandas/core/arrays/masked.py

@@ -905,7 +905,6 @@ def value_counts(self, dropna: bool = True) -> Series:
             Index,
             Series,
         )
-        from pandas.arrays import IntegerArray
 
         # compute counts on the data with no nans
         data = self._data[~self._mask]
@@ -925,9 +924,6 @@ def value_counts(self, dropna: bool = True) -> Series:
 
         index = index.astype(self.dtype)
 
-        mask = np.zeros(len(counts), dtype="bool")
-        counts = IntegerArray(counts, mask)
-
         return Series(counts, index=index)
 
     @doc(ExtensionArray.equals)

pandas/core/arrays/string_.py

@@ -485,7 +485,7 @@ def max(self, axis=None, skipna: bool = True, **kwargs) -> Scalar:
     def value_counts(self, dropna: bool = True):
         from pandas import value_counts
 
-        result = value_counts(self._ndarray, dropna=dropna).astype("Int64")
+        result = value_counts(self._ndarray, dropna=dropna)
         result.index = result.index.astype(self.dtype)
         return result
 

pandas/core/arrays/string_arrow.py

@@ -566,7 +566,7 @@ def value_counts(self, dropna: bool = True) -> Series:
 
         index = Index(type(self)(values))
 
-        return Series(counts, index=index).astype("Int64")
+        return Series(counts, index=index)
 
     def astype(self, dtype, copy: bool = True):
         dtype = pandas_dtype(dtype)

pandas/tests/arrays/boolean/test_function.py

@@ -92,20 +92,20 @@ def test_ufunc_reduce_raises(values):
 def test_value_counts_na():
     arr = pd.array([True, False, pd.NA], dtype="boolean")
     result = arr.value_counts(dropna=False)
-    expected = pd.Series([1, 1, 1], index=arr, dtype="Int64")
+    expected = pd.Series([1, 1, 1], index=arr)
     assert expected.index.dtype == arr.dtype
     tm.assert_series_equal(result, expected)
 
     result = arr.value_counts(dropna=True)
-    expected = pd.Series([1, 1], index=arr[:-1], dtype="Int64")
+    expected = pd.Series([1, 1], index=arr[:-1])
     assert expected.index.dtype == arr.dtype
     tm.assert_series_equal(result, expected)
 
 
 def test_value_counts_with_normalize():
     ser = pd.Series([True, False, pd.NA], dtype="boolean")
     result = ser.value_counts(normalize=True)
-    expected = pd.Series([1, 1], index=ser[:-1], dtype="Float64") / 2
+    expected = pd.Series([1, 1], index=ser[:-1]) / 2
     assert expected.index.dtype == "boolean"
     tm.assert_series_equal(result, expected)
 

pandas/tests/arrays/floating/test_function.py

@@ -100,11 +100,11 @@ def test_value_counts_na():
     result = arr.value_counts(dropna=False)
     idx = pd.Index([0.1, 0.2, pd.NA], dtype=arr.dtype)
     assert idx.dtype == arr.dtype
-    expected = pd.Series([2, 1, 1], index=idx, dtype="Int64")
+    expected = pd.Series([2, 1, 1], index=idx)
     tm.assert_series_equal(result, expected)
 
     result = arr.value_counts(dropna=True)
-    expected = pd.Series([2, 1], index=idx[:-1], dtype="Int64")
+    expected = pd.Series([2, 1], index=idx[:-1])
     tm.assert_series_equal(result, expected)
 
 
@@ -113,14 +113,14 @@ def test_value_counts_empty():
     result = ser.value_counts()
     idx = pd.Index([], dtype="Float64")
     assert idx.dtype == "Float64"
-    expected = pd.Series([], index=idx, dtype="Int64")
+    expected = pd.Series([], index=idx, dtype="int64")
     tm.assert_series_equal(result, expected)
 
 
 def test_value_counts_with_normalize():
     ser = pd.Series([0.1, 0.2, 0.1, pd.NA], dtype="Float64")
     result = ser.value_counts(normalize=True)
-    expected = pd.Series([2, 1], index=ser[:2], dtype="Float64") / 3
+    expected = pd.Series([2, 1], index=ser[:2]) / 3
     assert expected.index.dtype == ser.dtype
     tm.assert_series_equal(result, expected)
 

pandas/tests/arrays/integer/test_function.py

@@ -111,11 +111,11 @@ def test_value_counts_na():
     result = arr.value_counts(dropna=False)
     ex_index = pd.Index([1, 2, pd.NA], dtype="Int64")
     assert ex_index.dtype == "Int64"
-    expected = pd.Series([2, 1, 1], index=ex_index, dtype="Int64")
+    expected = pd.Series([2, 1, 1], index=ex_index)
     tm.assert_series_equal(result, expected)
 
     result = arr.value_counts(dropna=True)
-    expected = pd.Series([2, 1], index=arr[:2], dtype="Int64")
+    expected = pd.Series([2, 1], index=arr[:2])
     assert expected.index.dtype == arr.dtype
     tm.assert_series_equal(result, expected)
 
@@ -125,16 +125,16 @@ def test_value_counts_empty():
     ser = pd.Series([], dtype="Int64")
     result = ser.value_counts()
     idx = pd.Index([], dtype=ser.dtype)
+    expected = pd.Series([], index=idx, dtype="int64")
     assert idx.dtype == ser.dtype
-    expected = pd.Series([], index=idx, dtype="Int64")
     tm.assert_series_equal(result, expected)
 
 
 def test_value_counts_with_normalize():
     # GH 33172
     ser = pd.Series([1, 2, 1, pd.NA], dtype="Int64")
     result = ser.value_counts(normalize=True)
-    expected = pd.Series([2, 1], index=ser[:2], dtype="Float64") / 3
+    expected = pd.Series([2, 1], index=ser[:2]) / 3
     assert expected.index.dtype == ser.dtype
     tm.assert_series_equal(result, expected)
 

pandas/tests/arrays/string_/test_string.py

@@ -481,18 +481,18 @@ def test_arrow_load_from_zero_chunks(dtype, string_storage2):
 def test_value_counts_na(dtype):
     arr = pd.array(["a", "b", "a", pd.NA], dtype=dtype)
     result = arr.value_counts(dropna=False)
-    expected = pd.Series([2, 1, 1], index=arr[[0, 1, 3]], dtype="Int64")
+    expected = pd.Series([2, 1, 1], index=arr[[0, 1, 3]])
     tm.assert_series_equal(result, expected)
 
     result = arr.value_counts(dropna=True)
-    expected = pd.Series([2, 1], index=arr[:2], dtype="Int64")
+    expected = pd.Series([2, 1], index=arr[:2])
     tm.assert_series_equal(result, expected)
 
 
 def test_value_counts_with_normalize(dtype):
     ser = pd.Series(["a", "b", "a", pd.NA], dtype=dtype)
     result = ser.value_counts(normalize=True)
-    expected = pd.Series([2, 1], index=ser[:2], dtype="Float64") / 3
+    expected = pd.Series([2, 1], index=ser[:2]) / 3
     tm.assert_series_equal(result, expected)
 
 

pandas/tests/extension/base/methods.py

@@ -50,10 +50,6 @@ def test_value_counts_with_normalize(self, data):
             expected = pd.Series(0.0, index=result.index)
             expected[result > 0] = 1 / len(values)
 
-        if isinstance(data.dtype, pd.core.dtypes.dtypes.BaseMaskedDtype):
-            # TODO(GH#44692): avoid special-casing
-            expected = expected.astype("Float64")
-
         self.assert_series_equal(result, expected)
 
     def test_count(self, data_missing):

pandas/tests/extension/test_string.py

@@ -156,14 +156,10 @@ def test_reduce_series_numeric(self, data, all_numeric_reductions, skipna):
 
 
 class TestMethods(base.BaseMethodsTests):
-    @pytest.mark.skip(reason="returns nullable")
+    @pytest.mark.xfail(reason="returns nullable")
     def test_value_counts(self, all_data, dropna):
         return super().test_value_counts(all_data, dropna)
 
-    @pytest.mark.skip(reason="returns nullable")
-    def test_value_counts_with_normalize(self, data):
-        pass
-
 
 class TestCasting(base.BaseCastingTests):
     pass