DEPR: DataFrame.get_dtype_counts (#27145)

Author: mroeschke

doc/source/getting_started/basics.rst

@@ -1968,11 +1968,11 @@ dtype of the column will be chosen to accommodate all of the data types
    pd.Series([1, 2, 3, 6., 'foo'])
 
 The number of columns of each type in a ``DataFrame`` can be found by calling
-:meth:`~DataFrame.get_dtype_counts`.
+``DataFrame.dtypes.value_counts()``.
 
 .. ipython:: python
 
-   dft.get_dtype_counts()
+   dft.dtypes.value_counts()
 
 Numeric dtypes will propagate and can coexist in DataFrames.
 If a dtype is passed (either directly via the ``dtype`` keyword, a passed ``ndarray``,

doc/source/user_guide/io.rst

@@ -3767,7 +3767,7 @@ defaults to `nan`.
     store.append('df_mixed', df_mixed, min_itemsize={'values': 50})
     df_mixed1 = store.select('df_mixed')
     df_mixed1
-    df_mixed1.get_dtype_counts()
+    df_mixed1.dtypes.value_counts()
 
     # we have provided a minimum string column size
     store.root.df_mixed.table

doc/source/user_guide/missing_data.rst

@@ -105,7 +105,7 @@ pandas objects provide compatibility between ``NaT`` and ``NaN``.
    df2
    df2.loc[['a', 'c', 'h'], ['one', 'timestamp']] = np.nan
    df2
-   df2.get_dtype_counts()
+   df2.dtypes.value_counts()
 
 .. _missing.inserting:
 

doc/source/whatsnew/v0.10.1.rst

@@ -89,7 +89,7 @@ You can now store ``datetime64`` in data columns
     store.append('df_mixed', df_mixed)
     df_mixed1 = store.select('df_mixed')
     df_mixed1
-    df_mixed1.get_dtype_counts()
+    df_mixed1.dtypes.value_counts()
 
 You can pass ``columns`` keyword to select to filter a list of the return
 columns, this is equivalent to passing a

doc/source/whatsnew/v0.11.0.rst

@@ -296,7 +296,7 @@ Furthermore ``datetime64[ns]`` columns are created by default, when passed datet
    df
 
    # datetime64[ns] out of the box
-   df.get_dtype_counts()
+   df.dtypes.value_counts()
 
    # use the traditional nan, which is mapped to NaT internally
    df.loc[df.index[2:4], ['A', 'timestamp']] = np.nan

doc/source/whatsnew/v0.25.0.rst

@@ -786,6 +786,7 @@ Other deprecations
 - :meth:`Index.item` and :meth:`Series.item` is deprecated. (:issue:`18262`)
 - The default value ``ordered=None`` in :class:`~pandas.api.types.CategoricalDtype` has been deprecated in favor of ``ordered=False``. When converting between categorical types ``ordered=True`` must be explicitly passed in order to be preserved. (:issue:`26336`)
 - :meth:`Index.contains` is deprecated. Use ``key in index`` (``__contains__``) instead (:issue:`17753`).
+- :meth:`DataFrame.get_dtype_counts` is deprecated. (:issue:`18262`)
 
 .. _whatsnew_0250.prior_deprecations:
 

pandas/core/computation/expressions.py

@@ -79,11 +79,11 @@ def _can_use_numexpr(op, op_str, a, b, dtype_check):
             # check for dtype compatibility
             dtypes = set()
             for o in [a, b]:
-                if hasattr(o, 'get_dtype_counts'):
-                    s = o.get_dtype_counts()
+                if hasattr(o, 'dtypes'):
+                    s = o.dtypes.value_counts()
                     if len(s) > 1:
                         return False
-                    dtypes |= set(s.index)
+                    dtypes |= set(s.index.astype(str))
                 elif isinstance(o, np.ndarray):
                     dtypes |= {o.dtype.name}
 

pandas/core/frame.py

@@ -2326,7 +2326,7 @@ def _sizeof_fmt(num, size_qualifier):
             else:
                 _verbose_repr()
 
-        counts = self.get_dtype_counts()
+        counts = self._data.get_dtype_counts()
         dtypes = ['{k}({kk:d})'.format(k=k[0], kk=k[1]) for k
                   in sorted(counts.items())]
         lines.append('dtypes: {types}'.format(types=', '.join(dtypes)))

pandas/core/generic.py

@@ -5263,6 +5263,10 @@ def get_dtype_counts(self):
         """
         Return counts of unique dtypes in this object.
 
+        .. deprecated:: 0.25.0
+
+        Use `.dtypes.value_counts()` instead.
+
         Returns
         -------
         dtype : Series
@@ -5288,6 +5292,10 @@ def get_dtype_counts(self):
         object     1
         dtype: int64
         """
+        warnings.warn("`get_dtype_counts` has been deprecated and will be "
+                      "removed in a future version. For DataFrames use "
+                      "`.dtypes.value_counts()", FutureWarning,
+                      stacklevel=2)
         from pandas import Series
         return Series(self._data.get_dtype_counts())
 

pandas/tests/frame/test_api.py

@@ -7,8 +7,8 @@
 
 import pandas as pd
 from pandas import (
-    Categorical, DataFrame, Series, SparseDataFrame, compat, date_range,
-    timedelta_range)
+    Categorical, DataFrame, Series, SparseDataFrame, SparseDtype, compat,
+    date_range, timedelta_range)
 import pandas.util.testing as tm
 from pandas.util.testing import (
     assert_almost_equal, assert_frame_equal, assert_series_equal)
@@ -433,11 +433,11 @@ def test_with_datetimelikes(self):
                          'B': timedelta_range('1 day', periods=10)})
         t = df.T
 
-        result = t.get_dtype_counts()
+        result = t.dtypes.value_counts()
         if self.klass is DataFrame:
-            expected = Series({'object': 10})
+            expected = Series({np.dtype('object'): 10})
         else:
-            expected = Series({'Sparse[object, nan]': 10})
+            expected = Series({SparseDtype(dtype=object): 10})
         tm.assert_series_equal(result, expected)
 
 

pandas/tests/frame/test_arithmetic.py

@@ -273,8 +273,8 @@ def test_df_flex_cmp_constant_return_types(self, opname):
         df = pd.DataFrame({'x': [1, 2, 3], 'y': [1., 2., 3.]})
         const = 2
 
-        result = getattr(df, opname)(const).get_dtype_counts()
-        tm.assert_series_equal(result, pd.Series([2], ['bool']))
+        result = getattr(df, opname)(const).dtypes.value_counts()
+        tm.assert_series_equal(result, pd.Series([2], index=[np.dtype(bool)]))
 
     @pytest.mark.parametrize('opname', ['eq', 'ne', 'gt', 'lt', 'ge', 'le'])
     def test_df_flex_cmp_constant_return_types_empty(self, opname):
@@ -283,8 +283,8 @@ def test_df_flex_cmp_constant_return_types_empty(self, opname):
         const = 2
 
         empty = df.iloc[:0]
-        result = getattr(empty, opname)(const).get_dtype_counts()
-        tm.assert_series_equal(result, pd.Series([2], ['bool']))
+        result = getattr(empty, opname)(const).dtypes.value_counts()
+        tm.assert_series_equal(result, pd.Series([2], index=[np.dtype(bool)]))
 
 
 # -------------------------------------------------------------------

pandas/tests/frame/test_block_internals.py

@@ -217,19 +217,21 @@ def test_construction_with_mixed(self, float_string_frame):
         df = DataFrame(data)
 
         # check dtypes
-        result = df.get_dtype_counts().sort_values()
+        result = df.dtypes
         expected = Series({'datetime64[ns]': 3})
 
         # mixed-type frames
         float_string_frame['datetime'] = datetime.now()
         float_string_frame['timedelta'] = timedelta(days=1, seconds=1)
         assert float_string_frame['datetime'].dtype == 'M8[ns]'
         assert float_string_frame['timedelta'].dtype == 'm8[ns]'
-        result = float_string_frame.get_dtype_counts().sort_values()
-        expected = Series({'float64': 4,
-                           'object': 1,
-                           'datetime64[ns]': 1,
-                           'timedelta64[ns]': 1}).sort_values()
+        result = float_string_frame.dtypes
+        expected = Series([np.dtype('float64')] * 4 +
+                          [np.dtype('object'),
+                           np.dtype('datetime64[ns]'),
+                           np.dtype('timedelta64[ns]')],
+                          index=list('ABCD') + ['foo', 'datetime',
+                                                'timedelta'])
         assert_series_equal(result, expected)
 
     def test_construction_with_conversions(self):
@@ -409,11 +411,12 @@ def test_get_numeric_data(self):
         df = DataFrame({'a': 1., 'b': 2, 'c': 'foo',
                         'f': Timestamp('20010102')},
                        index=np.arange(10))
-        result = df.get_dtype_counts()
-        expected = Series({'int64': 1, 'float64': 1,
-                           datetime64name: 1, objectname: 1})
-        result = result.sort_index()
-        expected = expected.sort_index()
+        result = df.dtypes
+        expected = Series([np.dtype('float64'),
+                           np.dtype('int64'),
+                           np.dtype(objectname),
+                           np.dtype(datetime64name)],
+                          index=['a', 'b', 'c', 'f'])
         assert_series_equal(result, expected)
 
         df = DataFrame({'a': 1., 'b': 2, 'c': 'foo',

pandas/tests/frame/test_combine_concat.py

@@ -17,8 +17,10 @@ def test_concat_multiple_frames_dtypes(self):
         A = DataFrame(data=np.ones((10, 2)), columns=[
                       'foo', 'bar'], dtype=np.float64)
         B = DataFrame(data=np.ones((10, 2)), dtype=np.float32)
-        results = pd.concat((A, B), axis=1).get_dtype_counts()
-        expected = Series(dict(float64=2, float32=2))
+        results = pd.concat((A, B), axis=1).dtypes
+        expected = Series([np.dtype('float64')] * 2 +
+                          [np.dtype('float32')] * 2,
+                          index=['foo', 'bar', 0, 1])
         assert_series_equal(results, expected)
 
     @pytest.mark.parametrize('data', [