Skip to content

PERF: Regressions since v0.21 #18532

Closed
Closed
@mroeschke

Description

@mroeschke

xref

$ asv continuous -f 1.1 81372093f1fdc0c07e4b45ba0f47b upstream/master

+        54.0±9μs       1.40±0.01s 25895.28  indexing.IntervalIndexing.time_loc_list
+       65.6±20μs       1.39±0.02s 21250.19  indexing.IntervalIndexing.time_getitem_list
+     14.2±0.04μs      1.51±0.03ms   106.31  categoricals.CategoricalSlicing.time_getitem_bool_array('monotonic_decr')
+      35.6±0.5ms       1.99±0.01s    55.86  offset.ApplyIndex.time_apply_index(<BusinessDay>)
+      36.4±0.3ms       1.98±0.02s    54.20  offset.ApplyIndex.time_apply_index(<SemiMonthEnd: day_of_month=15>)
+      36.9±0.7ms          1.99±0s    53.86  offset.ApplyIndex.time_apply_index(<SemiMonthBegin: day_of_month=15>)
+         443±1ns       22.9±0.2μs    51.76  timestamp.TimestampProperties.time_weekday_name(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
+         444±5ns       23.0±0.2μs    51.71  timestamp.TimestampProperties.time_weekday_name(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, None)
+      22.3±0.5ms       1.04±0.01s    46.47  period.DataFramePeriodColumn.time_setitem_period_column
+     4.65±0.02ms          203±2ms    43.80  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<BusinessDay>)
+     4.87±0.06ms          202±1ms    41.57  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<SemiMonthBegin: day_of_month=15>)
+     5.01±0.09ms        202±0.6ms    40.33  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<SemiMonthEnd: day_of_month=15>)
+     5.15±0.03ms        204±0.9ms    39.65  offset.OffsetSeriesArithmetic.time_add_offset(<BusinessDay>)
+      13.8±0.1ms          522±2ms    37.99  timeseries.Iteration.time_iter_preexit(<function period_range at 0x1124ecea0>)
+     5.45±0.02ms          206±2ms    37.74  offset.OffsetSeriesArithmetic.time_add_offset(<SemiMonthBegin: day_of_month=15>)
+     5.51±0.02ms        206±0.7ms    37.43  offset.OffsetSeriesArithmetic.time_add_offset(<SemiMonthEnd: day_of_month=15>)
+       374±0.9ns      12.4±0.03μs    33.08  indexing.MethodLookup.time_lookup_ix
+     3.42±0.03ms        104±0.7ms    30.46  period.PeriodIndexConstructor.time_from_pydatetime('D')
+     1.71±0.01ms         50.1±1ms    29.26  indexing.CategoricalIndexIndexing.time_get_indexer_list('monotonic_decr')
+         438±3ns      8.69±0.04μs    19.85  timestamp.TimestampProperties.time_weekday_name(None, 'B')
+     5.01±0.08ms         99.1±4ms    19.78  timeseries.DatetimeIndex.time_timeseries_is_month_start('tz_aware')
+         444±2ns      8.65±0.08μs    19.50  timestamp.TimestampProperties.time_weekday_name(None, None)
+     8.82±0.09ms          170±2ms    19.24  multiindex_object.Values.time_datetime_level_values_copy
+      7.02±0.2μs        113±0.9μs    16.07  period.Indexing.time_get_loc
+      60.9±0.8ms          655±3ms    10.75  plotting.TimeseriesPlotting.time_plot_regular
+      6.36±0.1μs       67.8±0.5μs    10.67  period.Indexing.time_shallow_copy
+     7.30±0.02ms         75.1±3ms    10.29  frame_methods.Repr.time_frame_repr_wide
+     7.20±0.07μs       60.3±0.6μs     8.38  index_object.Indexing.time_slice('Int')
+      22.0±0.1ms          183±3ms     8.32  binary_ops.Ops2.time_frame_float_floor_by_zero
+     7.16±0.05μs       59.5±0.5μs     8.30  index_object.Indexing.time_slice_step('Int')
+       113±0.4μs        841±300μs     7.44  groupby.GroupByMethods.time_dtype_as_field('int', 'cummin', 'direct')
+      72.0±0.2μs        525±0.9μs     7.28  groupby.GroupByMethods.time_dtype_as_field('float', 'cummin', 'direct')
+      73.0±0.6μs          524±2μs     7.18  groupby.GroupByMethods.time_dtype_as_field('float', 'cummin', 'transformation')
+      73.1±0.7μs          519±3μs     7.10  groupby.GroupByMethods.time_dtype_as_field('float', 'cummax', 'transformation')
+      18.0±0.2μs        127±0.5μs     7.06  period.PeriodUnaryMethods.time_now('M')
+       116±0.9μs        814±300μs     7.03  groupby.GroupByMethods.time_dtype_as_field('int', 'cummax', 'transformation')
+      73.3±0.6μs          514±3μs     7.00  groupby.GroupByMethods.time_dtype_as_field('float', 'cummax', 'direct')
+      77.2±0.4μs          506±4μs     6.56  groupby.GroupByMethods.time_dtype_as_field('datetime', 'cummin', 'direct')
+      77.4±0.5μs          503±4μs     6.50  groupby.GroupByMethods.time_dtype_as_field('datetime', 'cummin', 'transformation')
+      81.3±0.3μs          528±5μs     6.49  groupby.GroupByMethods.time_dtype_as_field('float', 'cumprod', 'direct')
+      80.9±0.3μs          525±5μs     6.49  groupby.GroupByMethods.time_dtype_as_field('float', 'cumprod', 'transformation')
+        86.0±5μs          527±5μs     6.13  groupby.GroupByMethods.time_dtype_as_field('float', 'cumsum', 'direct')
+        86.8±5μs          531±3μs     6.12  groupby.GroupByMethods.time_dtype_as_field('float', 'cumsum', 'transformation')
+      18.3±0.1ms          112±7ms     6.11  frame_methods.Dropna.time_dropna('any', 1)
+      18.0±0.2ms          106±3ms     5.91  frame_methods.Dropna.time_dropna('any', 0)
+      30.9±0.7μs          178±1μs     5.78  period.PeriodUnaryMethods.time_asfreq('min')
+      31.1±0.4μs        178±0.9μs     5.73  period.PeriodUnaryMethods.time_asfreq('M')
+         116±1μs        628±200μs     5.40  groupby.GroupByMethods.time_dtype_as_field('int', 'cummax', 'direct')
+       107±0.3μs          570±5μs     5.31  groupby.GroupByMethods.time_dtype_as_group('datetime', 'cummin', 'transformation')
+      64.8±0.1μs          341±4μs     5.27  period.PeriodProperties.time_property('M', 'end_time')
+         108±1μs          566±4μs     5.22  groupby.GroupByMethods.time_dtype_as_group('float', 'cummin', 'direct')
+         110±1μs          572±4μs     5.22  groupby.GroupByMethods.time_dtype_as_group('datetime', 'cummin', 'direct')
+       109±0.5μs          567±4μs     5.20  groupby.GroupByMethods.time_dtype_as_group('float', 'cummin', 'transformation')
+       109±0.4μs          567±4μs     5.19  groupby.GroupByMethods.time_dtype_as_group('float', 'cummax', 'direct')
+         109±1μs          565±2μs     5.17  groupby.GroupByMethods.time_dtype_as_group('float', 'cummax', 'transformation')
+      65.7±0.7μs          340±1μs     5.17  period.PeriodProperties.time_property('min', 'end_time')
+         116±1μs         591±90μs     5.10  groupby.GroupByMethods.time_dtype_as_field('int', 'cummin', 'transformation')
+         116±2μs         578±10μs     5.00  groupby.GroupByMethods.time_dtype_as_group('int', 'cummin', 'transformation')
+         114±1μs          569±9μs     4.99  groupby.GroupByMethods.time_dtype_as_group('int', 'cummax', 'direct')
+       114±0.6μs          566±3μs     4.95  groupby.GroupByMethods.time_dtype_as_group('int', 'cummin', 'direct')
+         115±2μs          565±4μs     4.90  groupby.GroupByMethods.time_dtype_as_group('int', 'cummax', 'transformation')
+      32.2±0.1ms          155±4ms     4.83  eval.Eval.time_and('python', 1)
+      3.54±0.1μs       16.7±0.1μs     4.71  indexing.DataFrameStringIndexing.time_ix
+         124±4μs          583±6μs     4.68  groupby.GroupByMethods.time_dtype_as_group('float', 'cumsum', 'transformation')
+         125±4μs         583±30μs     4.68  groupby.GroupByMethods.time_dtype_as_field('int', 'cumsum', 'transformation')
+         124±5μs          573±2μs     4.63  groupby.GroupByMethods.time_dtype_as_group('int', 'cumsum', 'direct')
+         123±5μs          569±5μs     4.63  groupby.GroupByMethods.time_dtype_as_field('int', 'cumsum', 'direct')
+         124±6μs          574±5μs     4.62  groupby.GroupByMethods.time_dtype_as_group('int', 'cumsum', 'transformation')
+         128±5μs          577±3μs     4.52  groupby.GroupByMethods.time_dtype_as_group('float', 'cumsum', 'direct')
+      38.2±0.4ms          160±2ms     4.18  eval.Eval.time_and('python', 'all')
+      59.4±0.5μs          232±1μs     3.91  period.PeriodUnaryMethods.time_to_timestamp('min')
+      60.2±0.3μs          233±2μs     3.87  period.PeriodUnaryMethods.time_to_timestamp('M')
+      60.6±0.8μs        234±0.9μs     3.86  period.PeriodProperties.time_property('min', 'start_time')
+      60.5±0.5μs          232±2μs     3.84  period.PeriodProperties.time_property('M', 'start_time')
+      40.6±0.2ms          153±9ms     3.76  frame_methods.Dropna.time_dropna('all', 1)
+      38.4±0.3ms          144±9ms     3.75  frame_methods.Dropna.time_dropna('all', 0)
+     3.18±0.01μs       11.7±0.2μs     3.66  multiindex_object.GetLoc.time_string_get_loc
+     3.12±0.01ms       11.3±0.1ms     3.62  multiindex_object.GetLoc.time_small_get_loc_warm
+         102±3ms          360±4ms     3.52  groupby.Groups.time_series_groups('int64_large')
+     3.19±0.02ms      10.8±0.08ms     3.40  multiindex_object.GetLoc.time_med_get_loc_warm
+     27.3±0.08ms         90.3±2ms     3.30  binary_ops.Ops.time_frame_multi_and(False, 1)
+      51.2±0.4μs        169±0.7μs     3.30  period.Indexing.time_unique
+     3.36±0.09μs       11.1±0.1μs     3.30  multiindex_object.GetLoc.time_med_get_loc
+     5.58±0.02ms         18.3±2ms     3.28  frame_methods.Equals.time_frame_nonunique_equal
+      27.4±0.2ms         89.2±2ms     3.25  binary_ops.Ops.time_frame_multi_and(False, 'default')
+      53.7±0.5μs          172±2μs     3.21  period.PeriodUnaryMethods.time_now('min')
+     5.58±0.04ms         17.8±2ms     3.19  frame_methods.Equals.time_frame_nonunique_unequal
+      84.9±0.6μs          267±1μs     3.14  period.Algorithms.time_drop_duplicates('index')
+       227±0.9μs          696±8μs     3.06  groupby.GroupByMethods.time_dtype_as_field('int', 'cumprod', 'direct')
+         230±1μs          692±3μs     3.01  groupby.GroupByMethods.time_dtype_as_field('int', 'cumprod', 'transformation')
+         142±2μs          426±5μs     3.00  period.Indexing.time_intersection
+       139±0.9μs          415±4μs     2.98  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x1133397b8>, False)
+      31.2±0.1ms         92.9±1ms     2.98  binary_ops.Ops.time_frame_multi_and(True, 1)
+       139±0.7μs          412±4μs     2.97  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x1133396a8>, False)
+       139±0.7μs          410±3μs     2.96  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x11332b730>, False)
+       246±0.8μs          723±2μs     2.94  groupby.GroupByMethods.time_dtype_as_group('float', 'cumprod', 'transformation')
+       151±0.4μs          442±5μs     2.93  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x1133397b8>, True)
+       106±0.8μs        310±0.3μs     2.93  period.PeriodIndexConstructor.time_from_date_range('D')
+         150±1μs          437±5μs     2.92  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x1133396a8>, True)
+       248±0.9μs          723±4μs     2.92  groupby.GroupByMethods.time_dtype_as_group('float', 'cumprod', 'direct')
+       150±0.7μs          437±3μs     2.91  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x11332b730>, True)
+       139±0.6μs          403±3μs     2.90  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x113339840>, False)
+     9.12±0.08μs       26.5±0.2μs     2.90  timestamp.TimestampProperties.time_is_quarter_end(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
+         151±2μs          435±4μs     2.89  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x113339840>, True)
+     9.30±0.09μs       26.9±0.7μs     2.89  timestamp.TimestampProperties.time_is_year_start(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
+         249±2μs          718±6μs     2.88  groupby.GroupByMethods.time_dtype_as_group('int', 'cumprod', 'direct')
+         251±5μs          721±3μs     2.87  groupby.GroupByMethods.time_dtype_as_group('int', 'cumprod', 'transformation')
+      9.24±0.1μs       26.3±0.1μs     2.85  timestamp.TimestampProperties.time_is_year_end(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
+         142±5ms          404±7ms     2.84  groupby.Groups.time_series_groups('object_large')
+      21.1±0.1ms         59.9±2ms     2.83  groupby.ApplyDictReturn.time_groupby_apply_dict_return
+      9.24±0.1μs       26.1±0.2μs     2.82  timestamp.TimestampProperties.time_is_month_start(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
+      24.6±0.1μs       69.1±0.8μs     2.80  indexing.NumericSeriesIndexing.time_iloc_list_like(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'nonunique_monotonic_inc')
+      9.28±0.1μs       25.9±0.1μs     2.79  timestamp.TimestampProperties.time_is_quarter_start(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
+     24.6±0.08μs       68.7±0.1μs     2.79  indexing.NumericSeriesIndexing.time_iloc_list_like(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'unique_monotonic_inc')
+      53.8±0.2μs        149±0.8μs     2.78  period.Indexing.time_series_loc
+        11.3±2ms       31.3±0.1ms     2.77  io.msgpack.MSGPack.time_read_msgpack
+      9.53±0.2μs       26.3±0.2μs     2.76  timestamp.TimestampProperties.time_is_leap_year(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
+     6.58±0.04μs       18.1±0.6μs     2.75  timestamp.TimestampAcrossDst.time_replace_across_dst
+      25.0±0.1μs         68.8±1μs     2.75  indexing.NumericSeriesIndexing.time_iloc_list_like(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
+      36.5±0.2ms         99.3±1ms     2.72  binary_ops.Ops.time_frame_multi_and(True, 'default')
+         725±3μs      1.95±0.02ms     2.68  io.csv.ReadCSVParseDates.time_multiple_date
+      25.7±0.2μs       68.9±0.3μs     2.68  indexing.NumericSeriesIndexing.time_iloc_list_like(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
+         330±2μs          882±2μs     2.68  period.Algorithms.time_value_counts('index')
+     9.77±0.05μs       25.9±0.2μs     2.65  timestamp.TimestampProperties.time_is_month_end(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
+      36.8±0.1ms        97.0±10ms     2.64  frame_methods.Interpolate.time_interpolate(None)
+        44.2±1ms          116±3ms     2.62  join_merge.MergeAsof.time_by_int
+      54.2±0.3μs          140±2μs     2.58  indexing.NumericSeriesIndexing.time_ix_slice(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
+       120±0.6ms          307±2ms     2.56  groupby.GroupByMethods.time_dtype_as_group('float', 'unique', 'transformation')
+      52.0±0.2ms          133±1ms     2.55  groupby.GroupByMethods.time_dtype_as_field('int', 'unique', 'direct')
+      76.5±0.7ms          195±2ms     2.55  groupby.GroupByMethods.time_dtype_as_group('int', 'unique', 'direct')
+       120±0.6ms          306±3ms     2.54  groupby.GroupByMethods.time_dtype_as_group('float', 'unique', 'direct')
+      77.2±0.4ms          196±3ms     2.54  groupby.GroupByMethods.time_dtype_as_group('int', 'unique', 'transformation')
+      53.8±0.4ms          137±3ms     2.54  groupby.GroupByMethods.time_dtype_as_field('float', 'unique', 'direct')
+      39.2±0.3μs       99.2±0.4μs     2.53  indexing.NumericSeriesIndexing.time_ix_slice(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
+      53.2±0.3ms          133±2ms     2.51  groupby.GroupByMethods.time_dtype_as_field('int', 'unique', 'transformation')
+        93.8±3ms        235±0.7ms     2.50  reshape.WideToLong.time_wide_to_long_big
+     7.76±0.04μs       19.4±0.1μs     2.49  timestamp.TimestampOps.time_replace_tz(None)
+       123±0.7ms          306±2ms     2.49  groupby.GroupByMethods.time_dtype_as_group('datetime', 'unique', 'transformation')
+      54.3±0.1ms          135±1ms     2.49  groupby.GroupByMethods.time_dtype_as_field('float', 'unique', 'transformation')
+         170±2μs          418±4μs     2.47  indexing.NonNumericSeriesIndexing.time_getitem_label_slice('datetime', 'nonunique_monotonic_inc')
+         125±1ms          305±2ms     2.45  groupby.GroupByMethods.time_dtype_as_group('datetime', 'unique', 'direct')
+      25.1±0.2μs       61.1±0.2μs     2.43  indexing.NumericSeriesIndexing.time_iloc_slice(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'unique_monotonic_inc')
+        1.90±0ms      4.56±0.03ms     2.40  binary_ops.Timeseries.time_timestamp_series_compare(None)
+     1.90±0.01ms      4.54±0.05ms     2.39  binary_ops.Timeseries.time_series_timestamp_compare(None)
+      25.6±0.5μs       60.8±0.2μs     2.38  indexing.NumericSeriesIndexing.time_iloc_slice(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'nonunique_monotonic_inc')
+      25.6±0.1μs       60.4±0.2μs     2.36  indexing.NumericSeriesIndexing.time_iloc_slice(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
+      25.7±0.1μs       60.2±0.6μs     2.34  indexing.NumericSeriesIndexing.time_iloc_slice(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
+      8.03±0.1μs       18.8±0.2μs     2.34  ctors.SeriesDtypesConstructors.time_dtindex_from_series
+        844±10ms       1.97±0.03s     2.34  stat_ops.FrameMultiIndexOps.time_op([0, 1], 'mad')
+      13.1±0.2μs       30.4±0.3μs     2.33  timestamp.TimestampOps.time_replace_tz('US/Eastern')
+      68.1±0.5ms        157±0.6ms     2.30  groupby.GroupByMethods.time_dtype_as_field('object', 'unique', 'direct')
+      68.4±0.5ms          157±1ms     2.30  groupby.GroupByMethods.time_dtype_as_field('object', 'unique', 'transformation')
+      66.2±0.8ms          149±1ms     2.25  groupby.GroupByMethods.time_dtype_as_field('datetime', 'unique', 'direct')
+      33.9±0.2μs       75.9±0.2μs     2.24  indexing.NumericSeriesIndexing.time_getitem_slice(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'unique_monotonic_inc')
+         171±3μs        381±0.8μs     2.23  multiindex_object.Values.time_datetime_level_values_sliced
+      65.8±0.6ms        145±0.5ms     2.21  groupby.GroupByMethods.time_dtype_as_field('datetime', 'unique', 'transformation')
+        98.4±1ms          217±2ms     2.20  stat_ops.FrameMultiIndexOps.time_op(1, 'mad')
+      34.7±0.2μs         76.3±1μs     2.20  indexing.NumericSeriesIndexing.time_getitem_slice(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'nonunique_monotonic_inc')
+         745±5μs      1.64±0.01ms     2.20  io.csv.ReadCSVParseDates.time_baseline
+      48.4±0.6μs          106±2μs     2.18  indexing.NumericSeriesIndexing.time_loc_slice(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
+      34.6±0.2μs       75.2±0.2μs     2.18  indexing.NumericSeriesIndexing.time_getitem_slice(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
+      34.8±0.3μs       75.6±0.7μs     2.17  indexing.NumericSeriesIndexing.time_getitem_slice(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
+         816±8μs      1.77±0.02ms     2.17  indexing.NonNumericSeriesIndexing.time_getitem_list_like('datetime', 'unique_monotonic_inc')
+         494±5μs      1.06±0.01ms     2.15  groupby.GroupByMethods.time_dtype_as_group('object', 'unique', 'transformation')
+        500±10μs      1.07±0.01ms     2.14  groupby.GroupByMethods.time_dtype_as_group('object', 'unique', 'direct')
+      36.7±0.1μs       78.0±0.5μs     2.12  indexing.NumericSeriesIndexing.time_ix_scalar(<class 'pandas.core.indexes.numeric.Float64Index'>, 'nonunique_monotonic_inc')
+      49.5±0.9μs          105±1μs     2.11  timeseries.AsOf.time_asof_single_early('DataFrame')
+      61.7±0.2ms          130±7ms     2.11  frame_methods.Interpolate.time_interpolate('infer')
+         880±6μs       1.84±0.1ms     2.09  frame_methods.Interpolate.time_interpolate_some_good(None)
+      43.1±0.3μs         89.6±1μs     2.08  indexing.NumericSeriesIndexing.time_iloc_array(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'nonunique_monotonic_inc')
+      36.4±0.2μs         75.4±2μs     2.07  indexing.NumericSeriesIndexing.time_loc_slice(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
+      43.0±0.2μs       88.2±0.3μs     2.05  indexing.NumericSeriesIndexing.time_iloc_array(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'unique_monotonic_inc')
+      43.9±0.3μs       88.6±0.4μs     2.02  indexing.NumericSeriesIndexing.time_iloc_array(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
+      28.4±0.5μs       57.0±0.6μs     2.01  indexing.NumericSeriesIndexing.time_ix_scalar(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
+      44.0±0.3μs       88.2±0.6μs     2.00  indexing.NumericSeriesIndexing.time_iloc_array(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
+      1.64±0.02s       3.28±0.08s     2.00  sparse.SparseDataFrameConstructor.time_constructor
+      80.4±0.9ms        160±0.6ms     1.99  join_merge.MergeAsof.time_by_object
+      32.5±0.2μs         64.6±1μs     1.98  indexing.NumericSeriesIndexing.time_ix_scalar(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
+      47.8±0.2μs       93.6±0.9μs     1.96  groupby.GroupByMethods.time_dtype_as_group('float', 'size', 'transformation')
+      47.8±0.7μs         93.5±1μs     1.96  groupby.GroupByMethods.time_dtype_as_field('int', 'size', 'transformation')
+      47.7±0.6μs       93.3±0.2μs     1.96  groupby.GroupByMethods.time_dtype_as_group('datetime', 'size', 'transformation')
+      8.76±0.1ms       17.0±0.2ms     1.94  frame_methods.Repr.time_repr_tall
+      1.19±0.01s       2.30±0.01s     1.94  timeseries.ToDatetimeNONISO8601.time_different_offset
+     1.66±0.01ms      3.23±0.01ms     1.94  reshape.SimpleReshape.time_stack
+      47.9±0.2μs       92.4±0.9μs     1.93  groupby.GroupByMethods.time_dtype_as_field('float', 'size', 'direct')
+      47.7±0.3μs         92.2±1μs     1.93  groupby.GroupByMethods.time_dtype_as_field('object', 'size', 'direct')
+      47.9±0.3μs         92.4±1μs     1.93  groupby.GroupByMethods.time_dtype_as_field('datetime', 'size', 'direct')
+      47.6±0.1μs       91.9±0.5μs     1.93  groupby.GroupByMethods.time_dtype_as_field('int', 'size', 'direct')
+      47.5±0.2μs       91.6±0.5μs     1.93  groupby.GroupByMethods.time_dtype_as_field('float', 'size', 'transformation')
+      48.0±0.5μs       92.2±0.8μs     1.92  groupby.GroupByMethods.time_dtype_as_group('float', 'size', 'direct')
+      48.0±0.3μs       92.2±0.5μs     1.92  groupby.GroupByMethods.time_dtype_as_group('int', 'size', 'direct')
+      47.9±0.6μs       91.7±0.5μs     1.92  groupby.GroupByMethods.time_dtype_as_field('datetime', 'size', 'transformation')
+      48.3±0.2μs       92.5±0.4μs     1.91  groupby.GroupByMethods.time_dtype_as_group('datetime', 'size', 'direct')
+      48.1±0.2μs       92.0±0.9μs     1.91  groupby.GroupByMethods.time_dtype_as_group('int', 'size', 'transformation')
+      48.3±0.2μs       91.8±0.7μs     1.90  groupby.GroupByMethods.time_dtype_as_group('object', 'size', 'transformation')
+      82.2±0.5μs          156±2μs     1.90  indexing.CategoricalIndexIndexing.time_getitem_bool_array('monotonic_decr')
+      48.1±0.3μs       91.4±0.5μs     1.90  groupby.GroupByMethods.time_dtype_as_field('object', 'size', 'transformation')
+      63.0±0.4μs          119±5μs     1.90  groupby.GroupByMethods.time_dtype_as_field('int', 'count', 'direct')
+      63.2±0.8μs          119±1μs     1.88  groupby.GroupByMethods.time_dtype_as_field('int', 'count', 'transformation')
+       230±0.8ms          432±3ms     1.87  groupby.Transform.time_transform_lambda_max
+      62.7±0.8μs          117±1μs     1.86  groupby.GroupByMethods.time_dtype_as_group('int', 'count', 'transformation')
+      62.1±0.5μs        115±0.6μs     1.86  groupby.GroupByMethods.time_dtype_as_field('float', 'count', 'transformation')
+      61.8±0.2μs        115±0.8μs     1.85  groupby.GroupByMethods.time_dtype_as_field('float', 'count', 'direct')
+      63.4±0.6μs        118±0.5μs     1.85  groupby.GroupByMethods.time_dtype_as_group('datetime', 'count', 'direct')
+      63.6±0.1μs          118±1μs     1.85  groupby.GroupByMethods.time_dtype_as_group('datetime', 'count', 'transformation')
+      25.7±0.2μs       47.5±0.4μs     1.85  indexing.NumericSeriesIndexing.time_ix_scalar(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
+      63.0±0.6μs          116±1μs     1.85  groupby.GroupByMethods.time_dtype_as_group('int', 'count', 'direct')
+     13.6±0.08μs       25.1±0.4μs     1.85  ctors.SeriesDtypesConstructors.time_index_from_array_floats
+     2.92±0.02ms         5.39±1ms     1.85  gil.ParallelRolling.time_rolling('var')
+      49.3±0.3μs       91.0±0.6μs     1.85  groupby.GroupByMethods.time_dtype_as_group('object', 'size', 'direct')
+      63.8±0.4μs        117±0.7μs     1.84  groupby.GroupByMethods.time_dtype_as_group('float', 'count', 'transformation')
+        63.5±1μs          116±3μs     1.83  groupby.GroupByMethods.time_dtype_as_group('float', 'count', 'direct')
+      95.6±0.9μs          174±6μs     1.83  frame_methods.GetDtypeCounts.time_frame_get_dtype_counts
+         972±3μs      1.77±0.01ms     1.82  io.csv.ReadCSVDInferDatetimeFormat.time_read_csv(False, 'ymd')
+      61.7±0.4μs        112±0.7μs     1.82  groupby.GroupByMethods.time_dtype_as_field('datetime', 'count', 'transformation')
+      63.4±0.7μs        115±0.3μs     1.82  groupby.GroupByMethods.time_dtype_as_group('object', 'count', 'transformation')
+      61.6±0.4μs        112±0.8μs     1.81  groupby.GroupByMethods.time_dtype_as_field('datetime', 'count', 'direct')
+      64.4±0.6μs          115±1μs     1.79  groupby.GroupByMethods.time_dtype_as_group('object', 'count', 'direct')
+        29.1±2ms         51.9±1ms     1.78  binary_ops.Ops.time_frame_comparison(False, 1)
+        28.7±2ms       51.1±0.6ms     1.78  binary_ops.Ops.time_frame_comparison(False, 'default')
+     18.6±0.09μs       33.0±0.4μs     1.77  ctors.SeriesDtypesConstructors.time_dtindex_from_index_with_series
+     2.48±0.03ms      4.40±0.02ms     1.77  reindex.DropDuplicates.time_frame_drop_dups_bool(True)
+        1.08±0ms      1.90±0.02ms     1.76  io.csv.ReadCSVDInferDatetimeFormat.time_read_csv(False, 'iso8601')
+      14.8±0.1μs       26.2±0.2μs     1.76  inference.ToNumeric.time_from_float('ignore')
+         378±3μs         660±10μs     1.74  indexing.NumericSeriesIndexing.time_getitem_list_like(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
+      15.0±0.1μs       25.9±0.4μs     1.73  inference.ToNumeric.time_from_float('coerce')
+     2.00±0.01ms       3.45±0.1ms     1.73  frame_methods.Interpolate.time_interpolate_some_good('infer')
+     3.44±0.01μs      5.94±0.02μs     1.72  inference.ToNumericDowncast.time_downcast('int32', None)
+      29.2±0.2μs       50.3±0.7μs     1.72  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x11332b7b8>, True)
+      84.8±0.4μs          145±1μs     1.71  indexing.NumericSeriesIndexing.time_ix_slice(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'unique_monotonic_inc')
+      19.9±0.3μs       33.9±0.2μs     1.71  ctors.SeriesDtypesConstructors.time_index_from_array_string
+         116±1ms          197±9ms     1.70  frame_methods.Iteration.time_iterrows
+       154±0.2μs          260±5μs     1.69  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x11332b840>, True)
+         178±1ms          301±5ms     1.69  sparse.SparseDataFrameConstructor.time_from_scipy
+         254±1μs          427±5μs     1.68  indexing.NumericSeriesIndexing.time_ix_list_like(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'unique_monotonic_inc')
+     1.00±0.01ms      1.69±0.01ms     1.68  timeseries.ResampleDataFrame.time_method('max')
+      80.1±0.9μs          134±2μs     1.67  indexing.NonNumericSeriesIndexing.time_getitem_pos_slice('datetime', 'nonunique_monotonic_inc')
+      70.6±0.6μs          117±1μs     1.66  indexing.NumericSeriesIndexing.time_ix_slice(<class 'pandas.core.indexes.numeric.Float64Index'>, 'nonunique_monotonic_inc')
+     3.02±0.02ms      5.02±0.01ms     1.66  reindex.DropDuplicates.time_frame_drop_dups_bool(False)
+         196±2ns          324±2ns     1.65  multiindex_object.Integer.time_is_monotonic
+     1.02±0.01ms      1.69±0.03ms     1.65  timeseries.ResampleDataFrame.time_method('min')
+         142±3ms          234±2ms     1.65  indexing.NumericSeriesIndexing.time_getitem_lists(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
+         140±2ms          231±3ms     1.65  indexing.NumericSeriesIndexing.time_loc_array(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
+         123±1μs         203±70μs     1.65  groupby.GroupByMethods.time_dtype_as_field('int', 'first', 'transformation')
+      24.9±0.2μs       40.9±0.5μs     1.64  indexing.NumericSeriesIndexing.time_iloc_list_like(<class 'pandas.core.indexes.numeric.Float64Index'>, 'nonunique_monotonic_inc')
+      72.4±0.3μs          119±2μs     1.64  series_methods.SeriesConstructor.time_constructor(None)
+         241±1μs          395±4μs     1.64  indexing.NumericSeriesIndexing.time_ix_list_like(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
+         141±3ms          231±3ms     1.64  indexing.NumericSeriesIndexing.time_getitem_array(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
+     1.71±0.04ms      2.79±0.01ms     1.63  reshape.Melt.time_melt_dataframe
+     3.36±0.02μs      5.47±0.02μs     1.63  offset.OnOffset.time_on_offset(<MonthBegin>)
+       140±0.6ms          228±2ms     1.63  indexing.NumericSeriesIndexing.time_loc_list_like(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
+       143±0.8μs          231±3μs     1.62  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x11332b840>, False)
+       141±0.9ms          229±3ms     1.62  indexing.NumericSeriesIndexing.time_getitem_list_like(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
+     2.90±0.01μs       4.68±0.2μs     1.62  categoricals.CategoricalSlicing.time_getitem_scalar('non_monotonic')
+         107±1μs          173±2μs     1.61  timeseries.DatetimeIndex.time_unique('dst')
+       123±0.9ms          198±3ms     1.61  groupby.GroupByMethods.time_dtype_as_group('int', 'skew', 'transformation')
+     10.3±0.03μs       16.5±0.2μs     1.61  offset.OffestDatetimeArithmetic.time_apply(<DateOffset: days=2, months=2>)
+         191±2ms          306±1ms     1.60  groupby.GroupByMethods.time_dtype_as_group('float', 'skew', 'direct')
+      24.9±0.2μs       39.9±0.2μs     1.60  indexing.NumericSeriesIndexing.time_iloc_list_like(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
+       124±0.4ms          198±1ms     1.59  groupby.GroupByMethods.time_dtype_as_group('int', 'skew', 'direct')
+      9.81±0.1ms       15.6±0.7ms     1.59  eval.Query.time_query_datetime_column
+      73.9±0.9ms        117±0.8ms     1.59  sparse.SparseDataFrameConstructor.time_from_dict
+       192±0.8ms          304±4ms     1.59  groupby.GroupByMethods.time_dtype_as_group('float', 'skew', 'transformation')
+      40.9±0.5μs       64.9±0.3μs     1.58  timeseries.SortIndex.time_get_slice(False)
+     4.25±0.03ms      6.72±0.03ms     1.58  categoricals.Rank.time_rank_int
+      81.2±0.2ms        128±0.9ms     1.57  groupby.GroupByMethods.time_dtype_as_field('float', 'skew', 'transformation')
+     1.43±0.01ms      2.26±0.01ms     1.57  io.csv.ReadCSVDInferDatetimeFormat.time_read_csv(True, 'iso8601')
+      51.6±0.5μs       81.1±0.8μs     1.57  indexing.NumericSeriesIndexing.time_ix_slice(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
+         448±3μs          703±9μs     1.57  indexing.MultiIndexing.time_series_ix
+        1.40±0ms      2.19±0.01ms     1.57  io.csv.ReadCSVDInferDatetimeFormat.time_read_csv(True, 'ymd')
+         227±2μs         354±20μs     1.56  frame_ctor.FromRecords.time_frame_from_records_generator(1000)
+        82.1±1ms          128±1ms     1.56  groupby.GroupByMethods.time_dtype_as_field('float', 'skew', 'direct')
+     3.20±0.01ms       4.99±0.2ms     1.56  frame_methods.Apply.time_apply_pass_thru
+      85.1±0.6ms        132±0.7ms     1.56  groupby.GroupByMethods.time_dtype_as_field('int', 'skew', 'direct')
+      85.1±0.6ms          132±1ms     1.55  groupby.GroupByMethods.time_dtype_as_field('int', 'skew', 'transformation')
+       111±0.9μs          171±2μs     1.55  indexing.DataFrameNumericIndexing.time_iloc_dups
+     4.41±0.05ms      6.82±0.03ms     1.55  categoricals.Rank.time_rank_int_cat_ordered
+     4.44±0.04ms      6.82±0.06ms     1.54  categoricals.Rank.time_rank_string_cat_ordered
+     3.86±0.02ms         5.93±2ms     1.53  gil.ParallelRolling.time_rolling('skew')
+        83.1±1ms          127±9ms     1.53  frame_methods.Apply.time_apply_axis_1
+     4.58±0.05ms      6.99±0.07ms     1.53  categoricals.Rank.time_rank_int_cat
+      6.00±0.1μs      9.17±0.04μs     1.53  timestamp.TimestampOps.time_replace_None('US/Eastern')
+        83.5±2μs        127±0.6μs     1.53  groupby.GroupByMethods.time_dtype_as_field('float', 'first', 'direct')
+         111±2ms          169±4ms     1.52  sparse.SparseSeriesToFrame.time_series_to_frame
+      22.9±0.2ms         34.7±1ms     1.52  frame_methods.Equals.time_frame_object_unequal
+     6.12±0.07ms       9.28±0.4ms     1.52  frame_methods.Apply.time_apply_lambda_mean
+        82.2±1μs          124±4μs     1.51  indexing.NumericSeriesIndexing.time_loc_slice(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'unique_monotonic_inc')
+      83.3±0.3μs          126±1μs     1.51  groupby.GroupByMethods.time_dtype_as_field('float', 'first', 'transformation')
+     2.33±0.01ms      3.52±0.05ms     1.51  rolling.VariableWindowMethods.time_rolling('DataFrame', '50s', 'float', 'max')
+         287±2ms         431±10ms     1.50  groupby.GroupByMethods.time_dtype_as_field('int', 'mad', 'direct')
+     2.37±0.03ms      3.57±0.04ms     1.50  rolling.VariableWindowMethods.time_rolling('DataFrame', '50s', 'int', 'max')
+     2.70±0.03ms      4.06±0.04ms     1.50  rolling.Methods.time_rolling('DataFrame', 10, 'float', 'std')
+         652±4ms          978±5ms     1.50  groupby.GroupByMethods.time_dtype_as_group('float', 'mad', 'transformation')
+     2.80±0.05ms      4.19±0.02ms     1.50  rolling.Methods.time_rolling('DataFrame', 10, 'int', 'std')
+      23.8±0.1ms         35.6±9ms     1.50  gil.ParallelFactorize.time_loop(2)
+     2.79±0.01ms      4.18±0.01ms     1.50  rolling.Methods.time_rolling('DataFrame', 1000, 'int', 'std')
+        529±10μs          793±7μs     1.50  indexing.MultiIndexing.time_frame_ix
+         159±1ms          238±1ms     1.50  timeseries.ToDatetimeISO8601.time_iso8601_tz_spaceformat
+     2.74±0.03ms      4.10±0.03ms     1.50  rolling.Methods.time_rolling('DataFrame', 1000, 'float', 'std')
+     2.35±0.01ms      3.50±0.02ms     1.49  rolling.VariableWindowMethods.time_rolling('DataFrame', '50s', 'float', 'min')
+        85.2±4μs          127±1μs     1.49  groupby.GroupByMethods.time_dtype_as_field('float', 'max', 'direct')
+         287±2ms         427±10ms     1.49  groupby.GroupByMethods.time_dtype_as_field('int', 'mad', 'transformation')
+        87.0±4μs          129±2μs     1.48  groupby.GroupByMethods.time_dtype_as_field('float', 'min', 'transformation')
+      58.3±0.1μs        86.4±10μs     1.48  frame_methods.Dtypes.time_frame_dtypes
+        87.9±4μs          130±1μs     1.48  groupby.GroupByMethods.time_dtype_as_field('float', 'min', 'direct')
+        87.8±4μs          130±1μs     1.48  groupby.GroupByMethods.time_dtype_as_field('float', 'max', 'transformation')
+     2.40±0.01ms      3.54±0.02ms     1.48  rolling.VariableWindowMethods.time_rolling('DataFrame', '50s', 'int', 'min')
+         324±1μs          478±1μs     1.48  groupby.GroupByMethods.time_dtype_as_field('float', 'sem', 'direct')
+       283±0.8ms          417±3ms     1.47  groupby.GroupByMethods.time_dtype_as_field('float', 'mad', 'direct')
+         663±2ms          978±5ms     1.47  groupby.GroupByMethods.time_dtype_as_group('float', 'mad', 'direct')
+         282±2ms          415±2ms     1.47  groupby.GroupByMethods.time_dtype_as_field('float', 'mad', 'transformation')
+         424±2ms          621±3ms     1.47  groupby.GroupByMethods.time_dtype_as_group('int', 'mad', 'direct')
+        84.4±5μs        124±0.5μs     1.47  groupby.GroupByMethods.time_dtype_as_field('float', 'prod', 'transformation')
+         176±1μs          258±3μs     1.47  groupby.GroupByMethods.time_dtype_as_field('object', 'last', 'direct')
+         425±4ms          621±4ms     1.46  groupby.GroupByMethods.time_dtype_as_group('int', 'mad', 'transformation')
+         325±5μs          476±8μs     1.46  groupby.GroupByMethods.time_dtype_as_field('float', 'sem', 'transformation')
+        84.2±4μs        123±0.6μs     1.46  groupby.GroupByMethods.time_dtype_as_field('float', 'prod', 'direct')
+      81.1±0.7μs          118±1μs     1.46  groupby.GroupByMethods.time_dtype_as_field('float', 'last', 'transformation')
+      25.1±0.3ms       36.5±0.8ms     1.46  strings.Methods.time_get
+       284±0.8μs          412±3μs     1.45  multiindex_object.Duplicates.time_remove_unused_levels
+      81.4±0.3μs          118±1μs     1.45  groupby.GroupByMethods.time_dtype_as_field('float', 'last', 'direct')
+         333±2μs          482±2μs     1.45  indexing.NumericSeriesIndexing.time_loc_list_like(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
+         177±1μs          256±3μs     1.45  groupby.GroupByMethods.time_dtype_as_field('object', 'last', 'transformation')
+         708±4μs      1.02±0.01ms     1.45  timeseries.ResampleDataFrame.time_method('mean')
+     1.08±0.01ms      1.56±0.02ms     1.45  sparse.FromCoo.time_sparse_series_from_coo
+       121±0.5μs          174±1μs     1.44  groupby.GroupByMethods.time_dtype_as_group('datetime', 'min', 'direct')
+        85.1±5μs        122±0.9μs     1.44  groupby.GroupByMethods.time_dtype_as_field('float', 'sum', 'transformation')
+         182±2μs        262±0.5μs     1.44  groupby.GroupByMethods.time_dtype_as_field('object', 'first', 'direct')
+         742±4μs      1.07±0.01ms     1.44  indexing.PanelIndexing.time_subset
+       112±0.3μs          161±8μs     1.43  groupby.GroupByMethods.time_dtype_as_field('int', 'last', 'direct')
+      9.18±0.1μs       13.2±0.1μs     1.43  timestamp.TimestampConstruction.time_parse_iso8601_tz
+       125±0.1μs        179±0.8μs     1.43  groupby.GroupByMethods.time_dtype_as_field('float', 'std', 'transformation')
+         107±1μs        154±0.7μs     1.43  groupby.GroupByMethods.time_dtype_as_group('object', 'last', 'transformation')
+       182±0.6μs          259±1μs     1.43  groupby.GroupByMethods.time_dtype_as_field('object', 'first', 'transformation')
+         376±2μs          537±4μs     1.43  groupby.GroupByMethods.time_dtype_as_group('float', 'sem', 'direct')
+     1.57±0.02ms      2.24±0.05ms     1.43  io.csv.ReadCSVFloatPrecision.time_read_csv(';', '_', None)
+     2.38±0.01ms      3.39±0.04ms     1.43  categoricals.Concat.time_union
+        91.2±5μs        130±0.6μs     1.42  groupby.GroupByMethods.time_dtype_as_field('float', 'var', 'direct')
+        85.4±5μs        121±0.9μs     1.42  groupby.GroupByMethods.time_dtype_as_field('float', 'sum', 'direct')
+     6.63±0.03ms       9.42±0.5ms     1.42  frame_methods.Apply.time_apply_np_mean
+       108±0.5μs        153±0.4μs     1.42  groupby.GroupByMethods.time_dtype_as_group('object', 'last', 'direct')
+        1.58±0ms      2.24±0.02ms     1.42  io.csv.ReadCSVFloatPrecision.time_read_csv(',', '_', 'high')
+       121±0.5μs        172±0.5μs     1.42  groupby.GroupByMethods.time_dtype_as_field('object', 'count', 'direct')
+       118±0.6μs          167±2μs     1.42  groupby.GroupByMethods.time_dtype_as_group('float', 'first', 'direct')
+         785±7μs      1.11±0.01ms     1.42  period.Indexing.time_align
+       121±0.4μs          172±1μs     1.42  groupby.GroupByMethods.time_dtype_as_group('datetime', 'max', 'transformation')
+         374±3μs          528±4μs     1.41  groupby.GroupByMethods.time_dtype_as_field('int', 'sem', 'direct')
+       115±0.7μs        163±0.9μs     1.41  groupby.GroupByMethods.time_dtype_as_group('float', 'last', 'direct')
+     2.66±0.02ms      3.76±0.05ms     1.41  rolling.VariableWindowMethods.time_rolling('Series', '50s', 'float', 'min')
+         376±4μs          532±6μs     1.41  groupby.GroupByMethods.time_dtype_as_field('int', 'sem', 'transformation')
+       109±0.7μs          154±2μs     1.41  groupby.GroupByMethods.time_dtype_as_group('object', 'first', 'transformation')
+        2.71±0ms      3.83±0.05ms     1.41  rolling.VariableWindowMethods.time_rolling('Series', '50s', 'int', 'max')
+     1.51±0.02ms      2.13±0.07ms     1.41  io.csv.ReadCSVFloatPrecision.time_read_csv(';', '.', 'high')
+         112±1μs          158±4μs     1.41  groupby.GroupByMethods.time_dtype_as_field('int', 'last', 'transformation')
+     2.99±0.05μs       4.21±0.4μs     1.41  categoricals.CategoricalSlicing.time_getitem_scalar('monotonic_incr')
+         122±1μs        171±0.5μs     1.41  groupby.GroupByMethods.time_dtype_as_group('datetime', 'max', 'direct')
+      15.5±0.2ms       21.8±0.3ms     1.41  io.msgpack.MSGPack.time_write_msgpack
+       120±0.8μs          168±1μs     1.41  groupby.GroupByMethods.time_dtype_as_group('datetime', 'first', 'direct')
+       116±0.7μs          163±1μs     1.41  groupby.GroupByMethods.time_dtype_as_group('float', 'last', 'transformation')
+         221±3ms        310±0.9ms     1.41  frame_methods.Duplicated.time_frame_duplicated_wide
+      47.4±0.6ms       66.6±0.8ms     1.41  index_object.IndexAppend.time_append_range_list
+     1.57±0.01ms      2.21±0.06ms     1.40  io.csv.ReadCSVFloatPrecision.time_read_csv(';', '_', 'high')
+         238±3μs          334±6μs     1.40  frame_ctor.FromRecords.time_frame_from_records_generator(None)
+       113±0.7μs          159±1μs     1.40  groupby.GroupByMethods.time_dtype_as_group('int', 'last', 'transformation')
+        1.57±0ms      2.21±0.05ms     1.40  io.csv.ReadCSVFloatPrecision.time_read_csv(',', '_', None)
+       120±0.8μs          168±3μs     1.40  groupby.GroupByMethods.time_dtype_as_group('datetime', 'first', 'transformation')
+         435±3μs         609±10μs     1.40  groupby.GroupByMethods.time_dtype_as_group('int', 'sem', 'transformation')
+         246±1μs         344±30μs     1.40  groupby.GroupByMethods.time_dtype_as_field('int', 'head', 'transformation')
+       114±0.6μs        160±0.5μs     1.40  groupby.GroupByMethods.time_dtype_as_group('int', 'last', 'direct')
+         127±4μs          177±4μs     1.40  groupby.GroupByMethods.time_dtype_as_field('int', 'max', 'transformation')
+         126±1μs        176±0.8μs     1.40  groupby.GroupByMethods.time_dtype_as_field('float', 'std', 'direct')
+         382±2μs          533±8μs     1.40  groupby.GroupByMethods.time_dtype_as_group('float', 'sem', 'transformation')
+        86.5±5μs        121±0.5μs     1.39  groupby.GroupByMethods.time_dtype_as_field('float', 'mean', 'direct')
+         121±4μs          168±1μs     1.39  groupby.GroupByMethods.time_dtype_as_field('object', 'count', 'transformation')
+         899±4μs      1.25±0.02ms     1.39  groupby.SumMultiLevel.time_groupby_sum_multiindex
+       110±0.4μs        153±0.8μs     1.39  groupby.GroupByMethods.time_dtype_as_field('float', 'median', 'direct')
+        92.7±4μs        129±0.8μs     1.39  groupby.GroupByMethods.time_dtype_as_field('float', 'var', 'transformation')
+       125±0.6μs          175±2μs     1.39  groupby.GroupByMethods.time_dtype_as_group('int', 'first', 'direct')
+         419±1μs         582±30μs     1.39  categoricals.CategoricalSlicing.time_getitem_list('non_monotonic')
+     1.60±0.03ms      2.22±0.07ms     1.39  io.csv.ReadCSVFloatPrecision.time_read_csv(',', '_', 'round_trip')
+      13.5±0.3μs      18.7±0.06μs     1.39  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<DateOffset: days=2, months=2>)
+       119±0.4μs          166±1μs     1.39  groupby.GroupByMethods.time_dtype_as_group('float', 'first', 'transformation')
+       111±0.8μs          154±3μs     1.39  groupby.GroupByMethods.time_dtype_as_group('object', 'first', 'direct')
+         128±4μs          177±7μs     1.38  groupby.GroupByMethods.time_dtype_as_field('int', 'max', 'direct')
+     2.70±0.04ms      3.74±0.04ms     1.38  rolling.VariableWindowMethods.time_rolling('Series', '50s', 'float', 'max')
+       124±0.5μs         172±40μs     1.38  groupby.GroupByMethods.time_dtype_as_field('int', 'first', 'direct')
+       171±0.8μs          236±2μs     1.38  groupby.GroupByMethods.time_dtype_as_field('int', 'std', 'direct')
+         123±1μs          171±1μs     1.38  groupby.GroupByMethods.time_dtype_as_group('datetime', 'min', 'transformation')
+         110±1μs          152±1μs     1.38  groupby.GroupByMethods.time_dtype_as_field('float', 'median', 'transformation')
+         127±4μs          175±1μs     1.38  groupby.GroupByMethods.time_dtype_as_group('int', 'max', 'transformation')
+         119±1μs          164±2μs     1.38  groupby.GroupByMethods.time_dtype_as_group('datetime', 'last', 'direct')
+      26.5±0.1ms       36.5±0.1ms     1.38  join_merge.Concat.time_concat_small_frames(0)
+     2.75±0.04ms      3.79±0.07ms     1.38  rolling.VariableWindowMethods.time_rolling('Series', '50s', 'int', 'min')
+         127±4μs          175±1μs     1.38  groupby.GroupByMethods.time_dtype_as_field('int', 'min', 'transformation')
+      18.0±0.4μs       24.7±0.4μs     1.37  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x11332b7b8>, False)
+       118±0.5μs        162±0.5μs     1.37  groupby.GroupByMethods.time_dtype_as_group('datetime', 'last', 'transformation')
+     1.55±0.03ms      2.13±0.01ms     1.37  io.csv.ReadCSVFloatPrecision.time_read_csv(';', '_', 'round_trip')
+         174±1μs          239±1μs     1.37  groupby.GroupByMethods.time_dtype_as_group('float', 'std', 'direct')
+        1.53±0ms      2.10±0.04ms     1.37  io.csv.ReadCSVFloatPrecision.time_read_csv(',', '.', 'high')
+         124±4μs          170±1μs     1.37  groupby.GroupByMethods.time_dtype_as_group('float', 'min', 'direct')
+         169±1μs          232±1μs     1.37  groupby.GroupByMethods.time_dtype_as_field('int', 'std', 'transformation')
+     1.60±0.01ms      2.19±0.05ms     1.37  io.csv.ReadCSVFloatPrecision.time_read_csv(',', '.', None)
+         440±2μs          602±4μs     1.37  groupby.GroupByMethods.time_dtype_as_group('int', 'sem', 'direct')
+         127±5μs          174±2μs     1.37  groupby.GroupByMethods.time_dtype_as_group('int', 'max', 'direct')
+     1.60±0.03ms      2.18±0.07ms     1.36  io.csv.ReadCSVFloatPrecision.time_read_csv(';', '.', None)
+      5.63±0.2ms      7.69±0.06ms     1.36  strings.Cat.time_cat(0, None, None, 0.001)
+         127±4μs        173±0.9μs     1.36  groupby.GroupByMethods.time_dtype_as_group('int', 'min', 'transformation')
+         124±5μs        169±0.9μs     1.36  groupby.GroupByMethods.time_dtype_as_group('float', 'min', 'transformation')
+         127±1μs        172±0.6μs     1.36  groupby.GroupByMethods.time_dtype_as_group('int', 'first', 'transformation')
+      43.8±0.2μs       59.6±0.5μs     1.36  indexing.NumericSeriesIndexing.time_iloc_array(<class 'pandas.core.indexes.numeric.Float64Index'>, 'nonunique_monotonic_inc')
+         134±8μs          183±4μs     1.36  groupby.GroupByMethods.time_dtype_as_field('int', 'var', 'direct')
+         205±2μs          279±5μs     1.36  timeseries.DatetimeIndex.time_normalize('dst')
+         420±2μs         572±40μs     1.36  categoricals.CategoricalSlicing.time_getitem_list('monotonic_incr')
+         126±4μs          172±4μs     1.36  groupby.GroupByMethods.time_dtype_as_group('float', 'max', 'transformation')
+     6.00±0.01ms      8.16±0.03ms     1.36  categoricals.Rank.time_rank_string_cat
+        89.3±5μs        122±0.4μs     1.36  groupby.GroupByMethods.time_dtype_as_field('float', 'mean', 'transformation')
+         136±8μs          185±4μs     1.36  groupby.GroupByMethods.time_dtype_as_group('float', 'var', 'transformation')
+     3.41±0.01ms      4.64±0.01ms     1.36  rolling.Methods.time_rolling('Series', 1000, 'float', 'std')
+      55.3±0.4ms       75.2±0.3ms     1.36  stat_ops.Correlation.time_corr('spearman')
+         129±4μs        175±0.3μs     1.36  groupby.GroupByMethods.time_dtype_as_field('int', 'min', 'direct')
+      17.6±0.2ms       23.9±0.2ms     1.35  stat_ops.FrameMultiIndexOps.time_op(0, 'kurt')
+     3.50±0.02ms      4.73±0.01ms     1.35  rolling.Methods.time_rolling('Series', 10, 'int', 'std')
+     3.49±0.01ms      4.72±0.02ms     1.35  rolling.Methods.time_rolling('Series', 1000, 'int', 'std')
+     7.50±0.03μs      10.1±0.05μs     1.35  offset.OnOffset.time_on_offset(<YearEnd: month=12>)
+         175±1μs          237±2μs     1.35  groupby.GroupByMethods.time_dtype_as_group('float', 'std', 'transformation')
+      3.45±0.1ms      4.65±0.03ms     1.35  rolling.Methods.time_rolling('Series', 10, 'float', 'std')
+      29.9±0.2μs       40.2±0.4μs     1.35  offset.OffestDatetimeArithmetic.time_subtract(<DateOffset: days=2, months=2>)
+         130±4μs        174±0.9μs     1.35  groupby.GroupByMethods.time_dtype_as_group('int', 'min', 'direct')
+     6.77±0.03μs      9.11±0.05μs     1.34  index_object.Indexing.time_get_loc('Int')
+      9.66±0.1ms       13.0±0.4ms     1.34  categoricals.CategoricalSlicing.time_getitem_bool_array('non_monotonic')
+      10.1±0.2ms      13.6±0.03ms     1.34  timedelta.TimedeltaOps.time_add_td_ts
+      44.3±0.3μs       59.3±0.5μs     1.34  indexing.NumericSeriesIndexing.time_iloc_array(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
+        76.2±2ms        102±0.9ms     1.34  stat_ops.FrameMultiIndexOps.time_op(1, 'kurt')
+     1.35±0.01ms      1.80±0.05ms     1.34  join_merge.Merge.time_merge_dataframe_integer_key(False)
+         127±4μs          170±1μs     1.33  groupby.GroupByMethods.time_dtype_as_group('float', 'max', 'direct')
+        1.33±0ms      1.78±0.06ms     1.33  groupby.Datelike.time_sum('date_range')
+     5.32±0.02ms       7.09±0.2ms     1.33  reindex.DropDuplicates.time_frame_drop_dups(True)
+         293±4ms         391±30ms     1.33  frame_methods.Nunique.time_frame_nunique
+      96.4±0.7μs        128±0.7μs     1.33  join_merge.Concat.time_concat_empty_right(0)
+      61.7±0.1μs       81.9±0.6μs     1.33  indexing.NumericSeriesIndexing.time_getitem_slice(<class 'pandas.core.indexes.numeric.Float64Index'>, 'nonunique_monotonic_inc')
+     10.8±0.08ms       14.3±0.2ms     1.33  categoricals.Constructor.time_regular
+         106±2ms        140±0.6ms     1.32  index_object.IndexAppend.time_append_obj_list
+      48.5±0.9μs       64.1±0.1μs     1.32  frame_ctor.FromNDArray.time_frame_from_ndarray
+         329±6μs          434±1μs     1.32  timeseries.ResetIndex.time_reest_datetimeindex(None)
+        98.5±2μs        130±0.4μs     1.32  join_merge.Concat.time_concat_empty_left(0)
+      12.7±0.1ms      16.7±0.07ms     1.32  reshape.PivotTable.time_pivot_table
+         139±6μs          184±3μs     1.32  groupby.GroupByMethods.time_dtype_as_group('float', 'var', 'direct')
+      62.9±0.4μs       82.8±0.5μs     1.32  inference.ToNumeric.time_from_str('ignore')
+      50.1±0.6μs       65.8±0.6μs     1.32  indexing.NonNumericSeriesIndexing.time_getitem_pos_slice('datetime', 'unique_monotonic_inc')
+         248±9μs          326±8μs     1.31  groupby.GroupByMethods.time_dtype_as_group('int', 'prod', 'direct')
+         218±2μs          286±2μs     1.31  groupby.GroupByMethods.time_dtype_as_group('object', 'head', 'transformation')
+     10.5±0.08μs       13.8±0.3μs     1.31  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<YearBegin: month=1>)
+      7.04±0.2μs       9.23±0.2μs     1.31  index_object.Indexing.time_get_loc_sorted('Int')
+       216±0.9μs          283±3μs     1.31  groupby.GroupByMethods.time_dtype_as_group('object', 'head', 'direct')
+     7.26±0.04ms       9.52±0.1ms     1.31  indexing.InsertColumns.time_assign_with_setitem
+      59.8±0.4μs         78.3±3μs     1.31  frame_ctor.FromSeries.time_mi_series
+      88.5±0.3μs          115±1μs     1.30  groupby.GroupByMethods.time_dtype_as_field('int', 'shift', 'direct')
+         598±2μs         780±50μs     1.30  frame_methods.Quantile.time_frame_quantile(1)
+       226±0.7μs          294±2μs     1.30  groupby.GroupByMethods.time_dtype_as_group('object', 'tail', 'transformation')
+     2.15±0.02ms      2.80±0.01ms     1.30  groupby.Transform.time_transform_multi_key4
+         226±9μs          294±5μs     1.30  groupby.GroupByMethods.time_dtype_as_field('int', 'prod', 'transformation')
+     3.10±0.02ms      4.04±0.03ms     1.30  io.sas.SAS.time_read_msgpack('xport')
+       241±0.7μs          313±2μs     1.30  groupby.GroupByMethods.time_dtype_as_group('int', 'std', 'transformation')
+      88.6±0.7μs          115±1μs     1.30  groupby.GroupByMethods.time_dtype_as_field('int', 'shift', 'transformation')
+      80.4±0.1μs        105±0.9μs     1.30  groupby.GroupByMethods.time_dtype_as_field('datetime', 'shift', 'direct')
+         229±2μs          298±5μs     1.30  groupby.GroupByMethods.time_dtype_as_group('object', 'tail', 'direct')
+         866±9μs      1.13±0.01ms     1.30  series_methods.ValueCounts.time_value_counts('int')
+         629±4μs          817±5μs     1.30  reindex.DropDuplicates.time_series_drop_dups_int(False)
+     3.75±0.01ms      4.86±0.04ms     1.30  rolling.Pairwise.time_pairwise(1000, 'corr', False)
+         226±1μs          294±2μs     1.30  groupby.GroupByMethods.time_dtype_as_group('float', 'head', 'direct')
+       237±0.3μs          308±3μs     1.30  groupby.GroupByMethods.time_dtype_as_group('datetime', 'tail', 'transformation')
+      89.5±0.2μs          116±1μs     1.30  groupby.GroupByMethods.time_dtype_as_group('int', 'shift', 'transformation')
+      13.0±0.2μs       16.8±0.1μs     1.29  offset.OffestDatetimeArithmetic.time_add(<DateOffset: days=2, months=2>)
+     5.16±0.09ms      6.68±0.03ms     1.29  groupby.Transform.time_transform_multi_key2
+      89.8±0.4μs          116±1μs     1.29  groupby.GroupByMethods.time_dtype_as_group('int', 'shift', 'direct')
+       114±0.5μs        147±0.8μs     1.29  inference.NumericInferOps.time_subtract(<class 'numpy.int8'>)
+      7.83±0.1ms       10.1±0.2ms     1.29  stat_ops.FrameOps.time_op('mad', 'float', 0, False)
+       228±0.6μs          294±3μs     1.29  groupby.GroupByMethods.time_dtype_as_group('float', 'head', 'transformation')
+        71.9±2μs       92.7±0.6μs     1.29  indexing.NumericSeriesIndexing.time_ix_scalar(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'unique_monotonic_inc')
+         399±8μs          515±2μs     1.29  timeseries.ResetIndex.time_reest_datetimeindex('US/Eastern')
+         237±1μs          306±2μs     1.29  groupby.GroupByMethods.time_dtype_as_group('datetime', 'tail', 'direct')
+         236±1μs          305±2μs     1.29  groupby.GroupByMethods.time_dtype_as_group('float', 'tail', 'transformation')
+      74.7±0.9μs       96.2±0.9μs     1.29  groupby.GroupByMethods.time_dtype_as_field('float', 'shift', 'direct')
+      80.4±0.8μs        104±0.6μs     1.29  groupby.GroupByMethods.time_dtype_as_field('datetime', 'shift', 'transformation')
+      75.5±0.9μs       97.1±0.6μs     1.29  groupby.GroupByMethods.time_dtype_as_field('float', 'shift', 'transformation')
+         242±3μs          311±2μs     1.29  groupby.GroupByMethods.time_dtype_as_group('int', 'std', 'direct')
+     2.06±0.01ms      2.64±0.06ms     1.28  io.csv.ReadCSVFloatPrecision.time_read_csv(';', '.', 'round_trip')
+         456±3ms          586±7ms     1.28  stat_ops.SeriesMultiIndexOps.time_op([0, 1], 'mad')
+     5.76±0.08ms      7.40±0.09ms     1.28  reindex.DropDuplicates.time_frame_drop_dups_na(True)
+      58.9±0.2μs       75.7±0.8μs     1.28  timeseries.SortIndex.time_sort_index(True)
+         229±1μs          294±2μs     1.28  groupby.GroupByMethods.time_dtype_as_group('datetime', 'head', 'direct')
+         249±3μs          320±4μs     1.28  groupby.GroupByMethods.time_dtype_as_group('int', 'head', 'direct')
+         142±8μs        182±0.8μs     1.28  groupby.GroupByMethods.time_dtype_as_field('int', 'var', 'transformation')
+      45.7±0.5μs       58.6±0.6μs     1.28  indexing.NumericSeriesIndexing.time_getitem_slice(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
+         228±2μs          292±3μs     1.28  groupby.GroupByMethods.time_dtype_as_group('datetime', 'head', 'transformation')
+         245±7μs          315±1μs     1.28  groupby.GroupByMethods.time_dtype_as_field('int', 'sum', 'transformation')
+     2.61±0.04ms      3.35±0.01ms     1.28  rolling.Pairwise.time_pairwise(1000, 'cov', False)
+       250±0.9μs          320±1μs     1.28  groupby.GroupByMethods.time_dtype_as_field('datetime', 'head', 'transformation')
+      71.5±0.6ms       91.7±0.6ms     1.28  join_merge.Concat.time_concat_series(1)
+         162±2ms          207±2ms     1.28  stat_ops.SeriesMultiIndexOps.time_op([0, 1], 'kurt')
+       114±0.5μs          147±1μs     1.28  inference.NumericInferOps.time_subtract(<class 'numpy.uint8'>)
+         117±2μs        150±0.5μs     1.28  inference.NumericInferOps.time_add(<class 'numpy.int8'>)
+     13.4±0.04ms       17.1±0.2ms     1.28  join_merge.Concat.time_concat_series(0)
+     7.82±0.05ms      10.0±0.05ms     1.28  stat_ops.FrameOps.time_op('mad', 'float', 0, True)
+     2.07±0.02ms      2.66±0.08ms     1.28  io.csv.ReadCSVFloatPrecision.time_read_csv(',', '.', 'round_trip')
+         239±1μs          305±3μs     1.28  groupby.GroupByMethods.time_dtype_as_group('float', 'tail', 'direct')
+         249±8μs          319±2μs     1.28  groupby.GroupByMethods.time_dtype_as_group('int', 'sum', 'direct')
+         204±9μs          260±4μs     1.28  groupby.GroupByMethods.time_dtype_as_group('int', 'var', 'direct')
+     1.58±0.01ms      2.02±0.01ms     1.28  join_merge.Merge.time_merge_dataframe_integer_key(True)
+         565±4μs         722±10μs     1.28  groupby.GroupByMethods.time_dtype_as_group('object', 'value_counts', 'transformation')
+         248±1μs          317±5μs     1.28  groupby.GroupByMethods.time_dtype_as_field('int', 'head', 'direct')
+       148±0.9ms          188±1ms     1.28  replace.Convert.time_replace('DataFrame', 'Timedelta')
+         568±1μs          724±4μs     1.28  groupby.GroupByMethods.time_dtype_as_group('object', 'value_counts', 'direct')
+     1.07±0.01ms      1.37±0.01ms     1.27  groupby.SumBools.time_groupby_sum_booleans
+         257±2μs          328±4μs     1.27  groupby.GroupByMethods.time_dtype_as_field('int', 'tail', 'transformation')
+      29.8±0.3ms       37.9±0.3ms     1.27  stat_ops.FrameMultiIndexOps.time_op(0, 'mad')
+         258±2μs          329±3μs     1.27  groupby.GroupByMethods.time_dtype_as_group('int', 'tail', 'direct')
+      3.62±0.03s       4.61±0.02s     1.27  period.DataFramePeriodColumn.time_set_index
+         649±2ms         826±10ms     1.27  stat_ops.FrameMultiIndexOps.time_op([0, 1], 'kurt')
+         251±8μs          319±4μs     1.27  groupby.GroupByMethods.time_dtype_as_group('int', 'sum', 'transformation')
+         149±1ms          190±1ms     1.27  replace.Convert.time_replace('DataFrame', 'Timestamp')
+         225±9μs        286±0.6μs     1.27  groupby.GroupByMethods.time_dtype_as_field('int', 'mean', 'transformation')
+         158±2ms          200±2ms     1.27  stat_ops.SeriesMultiIndexOps.time_op([0, 1], 'skew')
+         247±9μs          313±2μs     1.27  groupby.GroupByMethods.time_dtype_as_field('int', 'sum', 'direct')
+         255±9μs          323±4μs     1.27  groupby.GroupByMethods.time_dtype_as_group('float', 'mean', 'transformation')
+         259±3μs          328±1μs     1.27  groupby.GroupByMethods.time_dtype_as_field('int', 'tail', 'direct')
+     8.74±0.06μs      11.1±0.06μs     1.27  offset.OffestDatetimeArithmetic.time_apply(<YearEnd: month=12>)
+      64.7±0.5μs         82.0±2μs     1.27  indexing.NonNumericSeriesIndexing.time_get_value('datetime', 'nonunique_monotonic_inc')
+        51.4±1μs       65.2±0.3μs     1.27  timeseries.SortIndex.time_get_slice(True)
+       256±0.6μs          324±2μs     1.27  groupby.GroupByMethods.time_dtype_as_field('float', 'tail', 'transformation')
+        73.6±1μs       93.3±0.8μs     1.27  series_methods.Clip.time_clip
+         743±1μs          942±3μs     1.27  reindex.DropDuplicates.time_series_drop_dups_string(False)
+     2.57±0.02ms      3.26±0.01ms     1.27  rolling.Pairwise.time_pairwise(None, 'cov', False)
+         247±1μs        313±0.4μs     1.27  groupby.GroupByMethods.time_dtype_as_group('int', 'head', 'transformation')
+      10.6±0.1μs       13.4±0.3μs     1.27  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<YearEnd: month=12>)
+         687±3μs         870±10μs     1.27  groupby.GroupByMethods.time_dtype_as_group('datetime', 'value_counts', 'direct')
+      86.4±0.4μs          109±1μs     1.27  groupby.GroupByMethods.time_dtype_as_field('datetime', 'min', 'direct')
+     1.18±0.02μs      1.49±0.01μs     1.27  index_object.Indexing.time_get('Int')
+       116±0.5μs        147±0.5μs     1.26  inference.NumericInferOps.time_add(<class 'numpy.uint8'>)
+         698±6μs         882±20μs     1.26  groupby.GroupByMethods.time_dtype_as_field('int', 'value_counts', 'direct')
+        206±10μs          260±2μs     1.26  groupby.GroupByMethods.time_dtype_as_group('int', 'var', 'transformation')
+         251±7μs          317±5μs     1.26  groupby.GroupByMethods.time_dtype_as_group('int', 'prod', 'transformation')
+       108±0.4μs          136±1μs     1.26  groupby.GroupByMethods.time_dtype_as_field('object', 'shift', 'transformation')
+        32.1±1ms       40.5±0.5ms     1.26  io.csv.ReadCSVCategorical.time_convert_direct
+      82.7±0.5μs        104±0.8μs     1.26  groupby.GroupByMethods.time_dtype_as_group('object', 'shift', 'direct')
+         782±3μs          986±2μs     1.26  groupby.GroupByMethods.time_dtype_as_field('float', 'value_counts', 'direct')
+      83.4±0.4μs        105±0.6μs     1.26  groupby.GroupByMethods.time_dtype_as_group('object', 'shift', 'transformation')
+       107±0.4μs          135±2μs     1.26  groupby.GroupByMethods.time_dtype_as_field('object', 'shift', 'direct')
+         260±1μs          327±3μs     1.26  groupby.GroupByMethods.time_dtype_as_field('datetime', 'tail', 'direct')
+         252±1μs          317±2μs     1.26  groupby.GroupByMethods.time_dtype_as_field('datetime', 'head', 'direct')
+       247±0.9μs          311±3μs     1.26  groupby.GroupByMethods.time_dtype_as_field('float', 'head', 'direct')
+       259±0.9μs          326±3μs     1.26  groupby.GroupByMethods.time_dtype_as_group('int', 'tail', 'transformation')
+         150±5μs          188±4μs     1.26  indexing.AssignTimeseriesIndex.time_frame_assign_timeseries_index
+     2.52±0.01ms      3.17±0.03ms     1.26  io.csv.ReadCSVFloatPrecision.time_read_csv_python_engine(',', '_', 'high')
+        598±10ns          751±8ns     1.26  index_object.Indexing.time_get('String')
+         269±1μs          338±3μs     1.26  groupby.GroupByMethods.time_dtype_as_group('float', 'median', 'transformation')
+        226±10μs          283±2μs     1.26  groupby.GroupByMethods.time_dtype_as_field('int', 'mean', 'direct')
+      8.34±0.1ms      10.5±0.05ms     1.26  stat_ops.Rank.time_rank('Series', True)
+         257±2μs          322±2μs     1.25  groupby.GroupByMethods.time_dtype_as_field('float', 'tail', 'direct')
+     7.99±0.05ms       10.0±0.2ms     1.25  stat_ops.FrameOps.time_op('mad', 'int', 0, True)
+     2.62±0.03ms      3.28±0.01ms     1.25  rolling.Pairwise.time_pairwise(10, 'cov', False)
+         261±8μs          328±1μs     1.25  join_merge.Append.time_append_homogenous
+     8.01±0.05ms       10.0±0.1ms     1.25  stat_ops.FrameOps.time_op('mad', 'int', 0, False)
+      20.8±0.2ms       26.1±0.4ms     1.25  join_merge.MergeAsof.time_on_int
+         785±7μs          983±6μs     1.25  groupby.GroupByMethods.time_dtype_as_field('float', 'value_counts', 'transformation')
+         269±1μs          336±2μs     1.25  groupby.GroupByMethods.time_dtype_as_group('float', 'median', 'direct')
+       248±0.7μs          310±1μs     1.25  groupby.GroupByMethods.time_dtype_as_field('float', 'head', 'transformation')
+      40.6±0.1μs       50.8±0.5μs     1.25  indexing.NonNumericSeriesIndexing.time_getitem_pos_slice('string', 'nonunique_monotonic_inc')
+         260±8μs          326±1μs     1.25  groupby.GroupByMethods.time_dtype_as_group('float', 'sum', 'transformation')
+      57.7±0.6ms       72.1±0.3ms     1.25  io.sas.SAS.time_read_msgpack('sas7bdat')
+     7.36±0.08μs      9.19±0.06μs     1.25  index_object.Indexing.time_slice_step('Float')
+        88.5±1μs          111±2μs     1.25  groupby.GroupByMethods.time_dtype_as_field('datetime', 'min', 'transformation')
+        258±10μs          323±5μs     1.25  groupby.GroupByMethods.time_dtype_as_group('float', 'sum', 'direct')
+      8.31±0.1ms      10.4±0.04ms     1.25  stat_ops.Rank.time_rank('Series', False)
+     2.52±0.01ms      3.15±0.01ms     1.25  io.csv.ReadCSVFloatPrecision.time_read_csv_python_engine(',', '_', 'round_trip')
+     3.85±0.04ms      4.80±0.05ms     1.25  rolling.Pairwise.time_pairwise(None, 'corr', False)
+     2.51±0.02ms      3.13±0.01ms     1.25  io.csv.ReadCSVFloatPrecision.time_read_csv_python_engine(';', '_', 'high')
+     1.79±0.02ms      2.23±0.03ms     1.25  rolling.Methods.time_rolling('DataFrame', 10, 'float', 'count')
+      84.6±0.1μs        105±0.6μs     1.25  groupby.GroupByMethods.time_dtype_as_group('float', 'shift', 'transformation')
+         700±3μs          872±3μs     1.25  groupby.GroupByMethods.time_dtype_as_field('int', 'value_counts', 'transformation')
+       264±0.8μs          329±3μs     1.25  groupby.GroupByMethods.time_dtype_as_field('datetime', 'tail', 'transformation')
+      41.1±0.1μs       51.2±0.5μs     1.25  indexing.NonNumericSeriesIndexing.time_getitem_pos_slice('string', 'unique_monotonic_inc')
+      85.5±0.8μs        106±0.3μs     1.24  groupby.GroupByMethods.time_dtype_as_group('float', 'shift', 'direct')
+     2.53±0.01ms      3.15±0.02ms     1.24  io.csv.ReadCSVFloatPrecision.time_read_csv_python_engine(';', '_', None)
+     3.85±0.04ms      4.79±0.01ms     1.24  rolling.Pairwise.time_pairwise(10, 'corr', False)
+      85.4±0.5μs        106±0.4μs     1.24  groupby.GroupByMethods.time_dtype_as_group('datetime', 'shift', 'direct')
+         151±1μs          188±2μs     1.24  inference.NumericInferOps.time_add(<class 'numpy.int16'>)
+     2.53±0.02ms      3.14±0.01ms     1.24  io.csv.ReadCSVFloatPrecision.time_read_csv_python_engine(';', '_', 'round_trip')
+     1.20±0.05μs      1.49±0.02μs     1.24  index_object.Indexing.time_get('Float')
+        259±10μs          321±2μs     1.24  groupby.GroupByMethods.time_dtype_as_group('float', 'mean', 'direct')
+         232±8μs          288±2μs     1.24  groupby.GroupByMethods.time_dtype_as_group('int', 'mean', 'direct')
+         649±9μs          805±4μs     1.24  groupby.GroupByMethods.time_dtype_as_field('object', 'value_counts', 'transformation')
+      67.9±0.3μs       84.3±0.9μs     1.24  indexing.NonNumericSeriesIndexing.time_getitem_label_slice('datetime', 'unique_monotonic_inc')
+      6.25±0.1ms      7.75±0.02ms     1.24  groupby.Transform.time_transform_multi_key1
+         680±6μs         844±20μs     1.24  groupby.GroupByMethods.time_dtype_as_group('int', 'value_counts', 'direct')
+         649±2μs          806±1μs     1.24  groupby.GroupByMethods.time_dtype_as_field('object', 'value_counts', 'direct')
+         689±5μs          854±4μs     1.24  groupby.GroupByMethods.time_dtype_as_group('datetime', 'value_counts', 'transformation')
+         261±2μs          323±5μs     1.24  groupby.GroupByMethods.time_dtype_as_field('int', 'median', 'direct')
+     2.65±0.05ms       3.28±0.1ms     1.24  io.csv.ReadUint64Integers.time_read_uint64
+     9.29±0.07ms       11.5±0.2ms     1.24  frame_methods.MaskBool.time_frame_mask_floats
+     5.57±0.06ms      6.90±0.04ms     1.24  rolling.VariableWindowMethods.time_rolling('DataFrame', '50s', 'float', 'std')
+     1.88±0.01ms      2.33±0.07ms     1.24  stat_ops.FrameMultiIndexOps.time_op(1, 'prod')
+         258±2μs          319±2μs     1.24  groupby.GroupByMethods.time_dtype_as_field('int', 'median', 'transformation')
+     14.8±0.07ms      18.3±0.08ms     1.24  reindex.DropDuplicates.time_frame_drop_dups(False)
+      78.5±0.5μs       97.1±0.7μs     1.24  groupby.GroupByMethods.time_dtype_as_field('datetime', 'last', 'direct')
+      51.0±0.5ms       63.1±0.7ms     1.24  stat_ops.SeriesMultiIndexOps.time_op(1, 'mad')
+      66.3±0.5μs         81.9±1μs     1.24  indexing.NumericSeriesIndexing.time_loc_slice(<class 'pandas.core.indexes.numeric.Float64Index'>, 'nonunique_monotonic_inc')
+     1.43±0.01ms      1.76±0.03ms     1.23  stat_ops.SeriesMultiIndexOps.time_op(0, 'sum')
+      8.69±0.1μs      10.7±0.05μs     1.23  offset.OffestDatetimeArithmetic.time_apply(<YearBegin: month=1>)
+     1.00±0.01ms         1.23±0ms     1.23  rolling.Methods.time_rolling('DataFrame', 10, 'float', 'sum')
+        86.2±1μs        106±0.6μs     1.23  groupby.GroupByMethods.time_dtype_as_group('datetime', 'shift', 'transformation')
+      73.2±0.6ms       90.2±0.9ms     1.23  frame_methods.ToHTML.time_to_html_mixed
+     2.24±0.02ms      2.76±0.06ms     1.23  stat_ops.FrameMultiIndexOps.time_op(0, 'var')
+      7.63±0.4μs      9.38±0.09μs     1.23  index_object.Indexing.time_slice('Float')
+         996±5μs      1.22±0.01ms     1.23  rolling.Methods.time_rolling('DataFrame', 1000, 'float', 'sum')
+      85.5±0.7μs        105±0.9μs     1.23  groupby.GroupByMethods.time_dtype_as_field('datetime', 'first', 'direct')
+      83.6±0.8μs          103±2μs     1.23  groupby.GroupByMethods.time_dtype_as_field('datetime', 'max', 'transformation')
+         135±1μs          165±1μs     1.23  join_merge.Concat.time_concat_empty_right(1)
+         264±7μs          324±3μs     1.23  groupby.GroupByMethods.time_dtype_as_group('float', 'prod', 'direct')
+         152±2μs          187±4μs     1.23  inference.NumericInferOps.time_multiply(<class 'numpy.uint8'>)
+      83.2±0.6μs        102±0.4μs     1.23  groupby.GroupByMethods.time_dtype_as_field('datetime', 'max', 'direct')
+     4.27±0.05ms      5.24±0.04ms     1.23  stat_ops.FrameMultiIndexOps.time_op(0, 'sem')
+     2.57±0.03ms      3.15±0.03ms     1.23  io.csv.ReadCSVFloatPrecision.time_read_csv_python_engine(',', '_', None)
+         153±1μs          187±3μs     1.23  inference.NumericInferOps.time_multiply(<class 'numpy.uint16'>)
+         293±4μs          360±1μs     1.23  groupby.GroupByMethods.time_dtype_as_field('object', 'tail', 'transformation')
+         152±2μs          186±2μs     1.23  inference.NumericInferOps.time_add(<class 'numpy.uint16'>)
+      9.09±0.1ms       11.1±0.1ms     1.23  stat_ops.Rank.time_average_old('Series', True)
+       682±0.9μs          836±7μs     1.23  groupby.GroupByMethods.time_dtype_as_group('float', 'value_counts', 'transformation')
+         681±3μs          835±5μs     1.23  groupby.GroupByMethods.time_dtype_as_group('float', 'value_counts', 'direct')
+     22.6±0.05μs       27.7±0.1μs     1.22  indexing.NonNumericSeriesIndexing.time_getitem_scalar('datetime', 'nonunique_monotonic_inc')
+         286±2μs          350±2μs     1.22  groupby.GroupByMethods.time_dtype_as_field('object', 'head', 'transformation')
+         683±2μs         836±10μs     1.22  groupby.GroupByMethods.time_dtype_as_group('int', 'value_counts', 'transformation')
+      79.4±0.3μs         97.2±1μs     1.22  groupby.GroupByMethods.time_dtype_as_field('datetime', 'last', 'transformation')
+     10.2±0.03ms      12.4±0.03ms     1.22  gil.ParallelRolling.time_rolling('std')
+         155±2μs          190±3μs     1.22  inference.NumericInferOps.time_multiply(<class 'numpy.int16'>)
+      26.6±0.2μs      32.5±0.07μs     1.22  indexing.NumericSeriesIndexing.time_iloc_slice(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
+     1.81±0.01ms      2.22±0.02ms     1.22  rolling.Methods.time_rolling('DataFrame', 10, 'int', 'count')
+     3.29±0.02ms       4.02±0.2ms     1.22  binary_ops.Ops.time_frame_mult(False, 'default')
+         238±5μs          290±3μs     1.22  groupby.GroupByMethods.time_dtype_as_group('int', 'mean', 'transformation')
+     5.67±0.02ms      6.94±0.02ms     1.22  rolling.VariableWindowMethods.time_rolling('DataFrame', '50s', 'int', 'std')
+     9.03±0.07ms       11.0±0.2ms     1.22  stat_ops.Rank.time_average_old('Series', False)
+     1.84±0.02ms      2.24±0.02ms     1.22  rolling.Methods.time_rolling('DataFrame', 1000, 'int', 'count')
+       295±0.9μs          360±3μs     1.22  groupby.GroupByMethods.time_dtype_as_field('object', 'tail', 'direct')
+     1.85±0.02ms      2.25±0.04ms     1.22  stat_ops.FrameMultiIndexOps.time_op(1, 'mean')
+         263±3ms          321±3ms     1.22  groupby.Apply.time_copy_overhead_single_col
+         266±9μs          325±2μs     1.22  groupby.GroupByMethods.time_dtype_as_group('float', 'prod', 'transformation')
+     9.14±0.07μs       11.2±0.4μs     1.22  timestamp.TimestampProperties.time_is_year_start(None, 'B')
+     3.30±0.06ms       4.03±0.2ms     1.22  binary_ops.Ops.time_frame_add(False, 1)
+     4.28±0.04ms      5.21±0.06ms     1.22  stat_ops.FrameMultiIndexOps.time_op(1, 'sem')
+     1.79±0.03ms      2.18±0.01ms     1.22  rolling.Methods.time_rolling('DataFrame', 1000, 'float', 'count')
+      13.5±0.1μs       16.4±0.3μs     1.22  indexing.NumericSeriesIndexing.time_getitem_scalar(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
+      85.8±0.6μs          104±1μs     1.22  groupby.GroupByMethods.time_dtype_as_field('datetime', 'first', 'transformation')
+      43.6±0.7ms         53.1±1ms     1.22  frame_methods.Equals.time_frame_object_equal
+       135±0.7μs          165±2μs     1.22  join_merge.Concat.time_concat_empty_left(1)
+        1.04±0ms         1.27±0ms     1.22  rolling.Methods.time_rolling('DataFrame', 10, 'int', 'sum')
+      27.4±0.4ms         33.4±2ms     1.22  binary_ops.Timeseries.time_timestamp_ops_diff_with_shift('US/Eastern')
+         151±1μs          184±4μs     1.22  inference.NumericInferOps.time_subtract(<class 'numpy.uint16'>)
+     1.46±0.02ms      1.77±0.02ms     1.22  stat_ops.SeriesMultiIndexOps.time_op(1, 'mean')
+      5.72±0.1ms       6.95±0.1ms     1.22  strings.Cat.time_cat(0, ',', '-', 0.001)
+       153±0.7μs          186±2μs     1.21  inference.NumericInferOps.time_multiply(<class 'numpy.int8'>)
+        1.05±0ms         1.27±0ms     1.21  rolling.Methods.time_rolling('DataFrame', 1000, 'int', 'sum')
+     3.30±0.05ms       4.01±0.2ms     1.21  binary_ops.Ops.time_frame_add(False, 'default')
+     1.83±0.01ms      2.22±0.02ms     1.21  stat_ops.FrameMultiIndexOps.time_op(0, 'mean')
+     5.40±0.06ms      6.56±0.01ms     1.21  rolling.VariableWindowMethods.time_rolling('DataFrame', '1d', 'float', 'std')
+         157±2μs          191±2μs     1.21  indexing.DataFrameStringIndexing.time_boolean_rows
+       285±0.9μs          346±5μs     1.21  groupby.GroupByMethods.time_dtype_as_field('object', 'head', 'direct')
+         262±3μs          318±2μs     1.21  groupby.GroupByMethods.time_dtype_as_group('int', 'median', 'direct')
+      5.50±0.1ms      6.67±0.05ms     1.21  rolling.VariableWindowMethods.time_rolling('DataFrame', '1d', 'int', 'std')
+         228±4μs          276±2μs     1.21  join_merge.JoinNonUnique.time_join_non_unique_equal
+     9.06±0.04μs       11.0±0.1μs     1.21  timestamp.TimestampProperties.time_is_leap_year(None, 'B')
+      9.27±0.1μs       11.2±0.1μs     1.21  timestamp.TimestampProperties.time_is_quarter_start(None, 'B')
+      5.62±0.1ms      6.81±0.08ms     1.21  strings.Cat.time_cat(0, None, '-', 0.001)
+     1.45±0.02ms      1.75±0.01ms     1.21  stat_ops.SeriesMultiIndexOps.time_op(1, 'prod')
+     1.50±0.03ms      1.81±0.04ms     1.21  stat_ops.SeriesMultiIndexOps.time_op(0, 'mean')
+      5.69±0.1ms      6.88±0.07ms     1.21  rolling.VariableWindowMethods.time_rolling('DataFrame', '1h', 'float', 'std')
+      48.9±0.4μs       59.1±0.3μs     1.21  indexing.NumericSeriesIndexing.time_loc_slice(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
+      17.8±0.3μs       21.5±0.4μs     1.21  index_object.Indexing.time_get_loc('Float')
+     1.45±0.02ms      1.75±0.02ms     1.21  stat_ops.SeriesMultiIndexOps.time_op(1, 'sum')
+     5.80±0.02ms      7.00±0.08ms     1.21  frame_methods.MaskBool.time_frame_mask_bools
+         238±9μs          287±1μs     1.21  groupby.GroupByMethods.time_dtype_as_field('int', 'prod', 'direct')
+         151±1μs          183±3μs     1.21  inference.NumericInferOps.time_subtract(<class 'numpy.int16'>)
+      73.0±0.3μs         88.0±5μs     1.21  frame_methods.GetNumericData.time_frame_get_numeric_data
+        306±20ms          369±4ms     1.21  groupby.GroupStrings.time_multi_columns
+         138±1μs        167±0.8μs     1.21  groupby.GroupByMethods.time_dtype_as_group('float', 'cumcount', 'transformation')
+         264±4μs          318±2μs     1.21  groupby.GroupByMethods.time_dtype_as_group('int', 'median', 'transformation')
+     3.11±0.03ms      3.75±0.02ms     1.21  io.csv.ReadCSVFloatPrecision.time_read_csv_python_engine(',', '.', None)
+     9.17±0.09μs       11.1±0.1μs     1.20  timestamp.TimestampProperties.time_is_month_start(None, 'B')
+         192±2ms        231±0.9ms     1.20  strings.Split.time_split(True)
+        1.45±0ms      1.75±0.01ms     1.20  stat_ops.SeriesMultiIndexOps.time_op(0, 'prod')
+     3.63±0.05ms       4.37±0.3ms     1.20  binary_ops.Ops.time_frame_add(True, 1)
+       132±0.2μs          159±2μs     1.20  groupby.GroupByMethods.time_dtype_as_group('object', 'cumcount', 'transformation')
+     3.18±0.05μs      3.83±0.07μs     1.20  indexing.CategoricalIndexIndexing.time_getitem_scalar('monotonic_incr')
+     3.12±0.02ms      3.75±0.03ms     1.20  io.csv.ReadCSVFloatPrecision.time_read_csv_python_engine(';', '.', None)
+     3.63±0.03ms       4.36±0.3ms     1.20  binary_ops.Ops.time_frame_mult(True, 1)
+     4.13±0.02ms      4.96±0.06ms     1.20  groupby.Apply.time_scalar_function_single_col
+         407±2μs         488±30μs     1.20  frame_methods.Quantile.time_frame_quantile(0)
+     3.14±0.03ms      3.77±0.02ms     1.20  io.csv.ReadCSVFloatPrecision.time_read_csv_python_engine(',', '.', 'high')
+         151±1ms          181±3ms     1.20  binary_ops.Ops2.time_frame_float_div_by_zero
+         945±6μs         1.13±0ms     1.20  reshape.SparseIndex.time_unstack
+     3.11±0.04ms      3.73±0.02ms     1.20  io.csv.ReadCSVFloatPrecision.time_read_csv_python_engine(';', '.', 'high')
+     2.20±0.05ms      2.63±0.05ms     1.20  timeseries.AsOf.time_asof_nan_single('DataFrame')
+         138±1μs          166±1μs     1.20  groupby.GroupByMethods.time_dtype_as_group('float', 'cumcount', 'direct')
+      25.9±0.2ms         31.0±1ms     1.20  eval.Query.time_query_datetime_index
+     3.30±0.02ms       3.95±0.2ms     1.20  binary_ops.Ops.time_frame_mult(False, 1)
+         427±1μs         511±60μs     1.20  frame_methods.Isnull.time_isnull_floats_no_null
+         159±1μs          190±9μs     1.19  groupby.GroupByMethods.time_dtype_as_field('int', 'cumcount', 'transformation')
+      27.0±0.3μs       32.2±0.2μs     1.19  indexing.NumericSeriesIndexing.time_iloc_slice(<class 'pandas.core.indexes.numeric.Float64Index'>, 'nonunique_monotonic_inc')
+     4.29±0.03ms       5.12±0.7ms     1.19  groupby.Categories.time_groupby_extra_cat_nosort
+         133±1μs          159±1μs     1.19  groupby.GroupByMethods.time_dtype_as_group('object', 'cumcount', 'direct')
+     2.87±0.01ms      3.42±0.03ms     1.19  timeseries.ToDatetimeISO8601.time_iso8601_nosep
+     9.08±0.06ms       10.8±0.2ms     1.19  groupby.MultiColumn.time_cython_sum
+      9.17±0.1μs      10.9±0.08μs     1.19  timestamp.TimestampProperties.time_is_quarter_end(None, 'B')
+     4.52±0.02ms      5.39±0.02ms     1.19  io.csv.ReadCSVDInferDatetimeFormat.time_read_csv(True, 'custom')
+     3.14±0.01ms      3.74±0.02ms     1.19  io.csv.ReadCSVFloatPrecision.time_read_csv_python_engine(';', '.', 'round_trip')
+      38.8±0.2ms       46.2±0.4ms     1.19  algorithms.Factorize.time_factorize_float(True)
+         659±2μs          783±3μs     1.19  groupby.GroupByMethods.time_dtype_as_field('datetime', 'value_counts', 'direct')
+     3.15±0.03μs       3.74±0.1μs     1.19  indexing.CategoricalIndexIndexing.time_getitem_scalar('non_monotonic')
+     9.08±0.05μs      10.8±0.04μs     1.19  timestamp.TimestampProperties.time_is_month_end(None, 'B')
+      63.9±0.2ms       75.8±0.9ms     1.19  indexing.NumericSeriesIndexing.time_ix_array(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
+     3.12±0.01ms      3.71±0.04ms     1.19  io.csv.ReadCSVFloatPrecision.time_read_csv_python_engine(',', '.', 'round_trip')
+         138±1μs          164±3μs     1.19  groupby.GroupByMethods.time_dtype_as_group('datetime', 'cumcount', 'direct')
+     1.87±0.01ms      2.22±0.02ms     1.19  stat_ops.FrameMultiIndexOps.time_op(1, 'sum')
+      15.6±0.3ms      18.5±0.08ms     1.19  strings.Methods.time_len
+         159±1μs          189±1μs     1.18  groupby.GroupByMethods.time_dtype_as_field('datetime', 'cumcount', 'direct')
+     5.12±0.05ms       6.06±0.2ms     1.18  strings.Cat.time_cat(0, None, '-', 0.0)
+     2.91±0.04ms      3.44±0.02ms     1.18  timeseries.ToDatetimeISO8601.time_iso8601
+         662±4μs          782±4μs     1.18  groupby.GroupByMethods.time_dtype_as_field('datetime', 'value_counts', 'transformation')
+     2.28±0.02ms      2.70±0.03ms     1.18  stat_ops.FrameMultiIndexOps.time_op(1, 'var')
+      14.3±0.1ms       16.9±0.8ms     1.18  gil.ParallelReadCSV.time_read_csv('object')
+      8.09±0.1ms       9.55±0.1ms     1.18  groupby.MultiColumn.time_col_select_numpy_sum
+         984±4ns      1.16±0.01μs     1.18  timestamp.TimestampConstruction.time_parse_iso8601_no_tz
+     2.85±0.05ms      3.36±0.01ms     1.18  timeseries.ToDatetimeISO8601.time_iso8601_format_no_sep
+      86.7±0.4ms          102±2ms     1.18  groupby.DateAttributes.time_len_groupby_object
+     2.07±0.01ms      2.43±0.02ms     1.18  series_methods.IsIn.time_isin('object')
+       159±0.5μs        187±0.5μs     1.18  groupby.GroupByMethods.time_dtype_as_field('float', 'cumcount', 'direct')
+     18.4±0.06μs       21.6±0.3μs     1.18  index_object.Indexing.time_get_loc_sorted('Float')
+     13.1±0.08ms         15.4±2ms     1.18  eval.Eval.time_mult('python', 1)
+     5.89±0.09ms      6.92±0.03ms     1.18  rolling.VariableWindowMethods.time_rolling('DataFrame', '1h', 'int', 'std')
+         159±1μs          187±2μs     1.18  groupby.GroupByMethods.time_dtype_as_field('float', 'cumcount', 'transformation')
+         155±2ms          182±3ms     1.17  binary_ops.Ops2.time_frame_int_div_by_zero
+     6.31±0.01ms      7.41±0.06ms     1.17  rolling.VariableWindowMethods.time_rolling('Series', '50s', 'int', 'std')
+     5.06±0.05ms      5.94±0.08ms     1.17  strings.Cat.time_cat(0, None, None, 0.0)
+      63.0±0.2μs       74.0±0.4μs     1.17  offset.OffestDatetimeArithmetic.time_add_10(<DateOffset: days=2, months=2>)
+     1.16±0.01ms      1.36±0.01ms     1.17  algorithms.Hashing.time_series_int
+     2.37±0.02ms      2.78±0.01ms     1.17  stat_ops.FrameMultiIndexOps.time_op(1, 'std')
+     4.50±0.04ms      5.28±0.01ms     1.17  groupby.Transform.time_transform_multi_key3
+     1.90±0.01ms      2.22±0.01ms     1.17  groupby.TransformNaN.time_first
+       163±0.6μs          191±1μs     1.17  groupby.GroupByMethods.time_dtype_as_group('int', 'cumcount', 'direct')
+         412±2μs          483±4μs     1.17  reindex.Reindex.time_reindex_columns
+     1.81±0.04ms      2.12±0.02ms     1.17  reindex.DropDuplicates.time_frame_drop_dups_int(True)
+         544±3μs          637±3μs     1.17  ctors.SeriesConstructors.time_series_constructor(<class 'list'>, True)
+     2.24±0.02ms      2.63±0.05ms     1.17  groupby.CountMultiInt.time_multi_int_count
+         160±2μs        187±0.4μs     1.17  groupby.GroupByMethods.time_dtype_as_field('int', 'cumcount', 'direct')
+     1.22±0.01ms      1.43±0.04ms     1.17  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 1, 'nearest')
+      18.1±0.2ms       21.2±0.1ms     1.17  reindex.DropDuplicates.time_frame_drop_dups_na(False)
+      20.6±0.4ms       24.1±0.2ms     1.17  stat_ops.SeriesMultiIndexOps.time_op(1, 'kurt')
+         529±2μs          619±8μs     1.17  ctors.SeriesConstructors.time_series_constructor(<class 'list'>, False)
+      53.6±0.3μs       62.6±0.3μs     1.17  frame_methods.XS.time_frame_xs(0)
+     6.26±0.05ms      7.31±0.04ms     1.17  rolling.VariableWindowMethods.time_rolling('Series', '50s', 'float', 'std')
+     5.34±0.05ms       6.23±0.1ms     1.17  strings.Cat.time_cat(0, ',', '-', 0.0)
+     6.25±0.06ms      7.29±0.02ms     1.17  rolling.VariableWindowMethods.time_rolling('Series', '1h', 'float', 'std')
+       159±0.4μs          185±2μs     1.17  groupby.GroupByMethods.time_dtype_as_field('object', 'cumcount', 'direct')
+     2.97±0.03ms      3.46±0.01ms     1.17  timeseries.ToDatetimeISO8601.time_iso8601_format
+         360±3μs          419±3μs     1.17  reindex.ReindexMethod.time_reindex_method('pad')
+         184±1μs          214±3μs     1.17  indexing.DataFrameStringIndexing.time_boolean_rows_object
+     1.99±0.03ms      2.31±0.05ms     1.16  stat_ops.FrameMultiIndexOps.time_op(0, 'prod')
+     2.14±0.04ms      2.49±0.06ms     1.16  timeseries.AsOf.time_asof_single('DataFrame')
+         386±9μs          449±3μs     1.16  timeseries.DatetimeIndex.time_unique('repeated')
+     7.98±0.02ms       9.27±0.2ms     1.16  ctors.MultiIndexConstructor.time_multiindex_from_iterables
+     1.17±0.02ms      1.36±0.01ms     1.16  algorithms.Hashing.time_series_float
+     1.94±0.01ms      2.25±0.04ms     1.16  stat_ops.SeriesMultiIndexOps.time_op(1, 'std')
+      6.15±0.2ms      7.12±0.06ms     1.16  rolling.VariableWindowMethods.time_rolling('Series', '1d', 'int', 'std')
+     1.90±0.04ms      2.20±0.03ms     1.16  stat_ops.SeriesMultiIndexOps.time_op(0, 'var')
+         306±1ms          354±9ms     1.16  stat_ops.FrameMultiIndexOps.time_op([0, 1], 'skew')
+         217±5μs          251±2μs     1.16  timeseries.DatetimeIndex.time_add_timedelta('dst')
+     8.92±0.04ms       10.3±0.1ms     1.16  stat_ops.SeriesMultiIndexOps.time_op(0, 'mad')
+     1.94±0.01ms      2.25±0.03ms     1.16  stat_ops.SeriesMultiIndexOps.time_op(0, 'std')
+     9.14±0.04ms      10.6±0.08ms     1.16  join_merge.Merge.time_merge_2intkey(False)
+     5.50±0.09ms       6.35±0.1ms     1.16  strings.Cat.time_cat(0, ',', None, 0.0)
+     2.05±0.03ms      2.37±0.04ms     1.16  binary_ops.Ops.time_frame_comparison(True, 1)
+         162±2μs          187±2μs     1.16  groupby.GroupByMethods.time_dtype_as_group('int', 'cumcount', 'transformation')
+     1.16±0.02ms         1.34±0ms     1.16  algorithms.Hashing.time_series_timedeltas
+     1.90±0.03ms      2.20±0.02ms     1.16  stat_ops.FrameMultiIndexOps.time_op(0, 'sum')
+     2.38±0.01ms      2.75±0.02ms     1.15  stat_ops.FrameMultiIndexOps.time_op(0, 'std')
+         209±4μs          241±5μs     1.15  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<MonthEnd>)
+      35.1±0.5μs       40.5±0.3μs     1.15  indexing.NumericSeriesIndexing.time_loc_scalar(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
+       117±0.9ms        135±0.7ms     1.15  replace.Convert.time_replace('Series', 'Timedelta')
+     6.10±0.05ms      7.04±0.02ms     1.15  rolling.VariableWindowMethods.time_rolling('Series', '1d', 'float', 'std')
+      63.3±0.4ms         73.1±2ms     1.15  indexing.NumericSeriesIndexing.time_ix_array(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'nonunique_monotonic_inc')
+     1.54±0.01ms      1.77±0.01ms     1.15  timeseries.ResampleDatetetime64.time_resample
+         160±2μs          184±1μs     1.15  groupby.GroupByMethods.time_dtype_as_field('object', 'cumcount', 'transformation')
+     1.16±0.01ms      1.33±0.01ms     1.15  algorithms.Hashing.time_series_dates
+     4.90±0.05ms       5.64±0.1ms     1.15  io.csv.ReadUint64Integers.time_read_uint64_neg_values
+     1.91±0.01ms      2.20±0.05ms     1.15  stat_ops.SeriesMultiIndexOps.time_op(1, 'var')
+         687±3ms          790±4ms     1.15  join_merge.MergeAsof.time_multiby
+         251±1μs          288±4μs     1.15  inference.NumericInferOps.time_subtract(<class 'numpy.uint32'>)
+         214±4μs          246±2μs     1.15  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<Day>)
+         181±1μs          208±2μs     1.15  strings.Encode.time_encode_decode
+      15.0±0.3ms         17.2±2ms     1.15  eval.Eval.time_mult('numexpr', 'all')
+         163±3μs          188±1μs     1.15  groupby.GroupByMethods.time_dtype_as_field('datetime', 'cumcount', 'transformation')
+         117±1ms        134±0.7ms     1.15  replace.Convert.time_replace('Series', 'Timestamp')
+      54.9±0.8ms       62.9±0.7ms     1.15  join_merge.MergeOrdered.time_merge_ordered
+     2.62±0.01ms      3.00±0.04ms     1.14  stat_ops.SeriesMultiIndexOps.time_op(1, 'sem')
+         204±1μs        233±0.5μs     1.14  groupby.GroupByMethods.time_dtype_as_group('object', 'nunique', 'transformation')
+         203±2μs          232±3μs     1.14  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<MonthBegin>)
+        1.02±0ms      1.16±0.01ms     1.14  rolling.Methods.time_rolling('DataFrame', 1000, 'float', 'mean')
+      29.0±0.6ms         33.1±2ms     1.14  eval.Eval.time_chained_cmp('python', 'all')
+     1.48±0.04ms      1.69±0.05ms     1.14  timeseries.DatetimeIndex.time_add_timedelta('tz_aware')
+       203±0.5μs          232±2μs     1.14  groupby.GroupByMethods.time_dtype_as_group('object', 'nunique', 'direct')
+     4.67±0.03ms       5.32±0.1ms     1.14  stat_ops.SeriesMultiIndexOps.time_op(0, 'skew')
+     4.72±0.03ms      5.38±0.04ms     1.14  stat_ops.SeriesMultiIndexOps.time_op(0, 'kurt')
+     4.84±0.04ms      5.50±0.02ms     1.14  join_merge.Merge.time_merge_dataframe_integer_2key(False)
+      19.3±0.2μs       21.9±0.3μs     1.14  indexing.NumericSeriesIndexing.time_getitem_scalar(<class 'pandas.core.indexes.numeric.Float64Index'>, 'nonunique_monotonic_inc')
+     2.64±0.01ms      3.00±0.02ms     1.14  stat_ops.SeriesMultiIndexOps.time_op(0, 'sem')
+         449±2μs          509±6μs     1.13  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<DateOffset: days=2, months=2>)
+     1.47±0.01ms      1.67±0.02ms     1.13  series_methods.IsInForObjects.time_isin_long_series_short_values
+        1.23±0ms         1.39±0ms     1.13  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 0, 'nearest')
+         258±2μs          292±6μs     1.13  inference.NumericInferOps.time_add(<class 'numpy.uint32'>)
+         493±6μs          558±4μs     1.13  indexing.DataFrameNumericIndexing.time_bool_indexer
+     1.07±0.01ms      1.21±0.01ms     1.13  rolling.Methods.time_rolling('DataFrame', 10, 'int', 'mean')
+     1.82±0.02ms      2.05±0.02ms     1.13  rolling.VariableWindowMethods.time_rolling('DataFrame', '50s', 'float', 'sum')
+     5.86±0.03ms      6.62±0.02ms     1.13  stat_ops.FrameMultiIndexOps.time_op([0, 1], 'sem')
+       107±0.4μs          121±1μs     1.13  inference.ToNumericDowncast.time_downcast('int32', 'float')
+        846±20μs         956±20μs     1.13  series_methods.IsInForObjects.time_isin_short_series_long_values
+      65.6±0.5μs       74.1±0.3μs     1.13  indexing.DataFrameNumericIndexing.time_loc
+     1.23±0.01ms      1.39±0.01ms     1.13  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 0, 'nearest')
+      19.6±0.2ms       22.1±0.3ms     1.13  eval.Eval.time_chained_cmp('python', 1)
+     1.83±0.02ms      2.07±0.04ms     1.13  rolling.VariableWindowMethods.time_rolling('DataFrame', '1d', 'int', 'sum')
+      2.05±0.01s       2.31±0.02s     1.13  groupby.GroupByMethods.time_dtype_as_group('float', 'describe', 'direct')
+     1.81±0.02ms      2.04±0.03ms     1.13  rolling.VariableWindowMethods.time_rolling('DataFrame', '1h', 'float', 'sum')
+     1.61±0.05ms      1.82±0.01ms     1.13  groupby.GroupManyLabels.time_sum(1)
+     1.23±0.01ms      1.39±0.01ms     1.13  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 0, 'midpoint')
+      6.53±0.1ms      7.36±0.03ms     1.13  rolling.VariableWindowMethods.time_rolling('Series', '1h', 'int', 'std')
+     1.24±0.01ms         1.39±0ms     1.13  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 0, 'lower')
+     13.1±0.06ms       14.7±0.2ms     1.13  eval.Eval.time_add('python', 1)
+     1.39±0.02μs      1.57±0.01μs     1.13  timestamp.TimestampConstruction.time_parse_today
+         885±3ms         995±20ms     1.13  groupby.GroupByMethods.time_dtype_as_field('int', 'describe', 'transformation')
+     1.26±0.04ms      1.42±0.04ms     1.12  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 1, 'linear')
+        1.32±0ms      1.48±0.02ms     1.12  rolling.Methods.time_rolling('Series', 10, 'float', 'sum')
+      69.5±0.5ms       78.2±0.7ms     1.12  io.hdf.HDFStoreDataFrame.time_read_store_table_mixed
+     1.23±0.01ms         1.38±0ms     1.12  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 1, 'midpoint')
+        1.23±0ms      1.38±0.01ms     1.12  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 1, 'lower')
+     1.98±0.01ms      2.22±0.03ms     1.12  binary_ops.Timeseries.time_series_timestamp_compare('US/Eastern')
+     1.04±0.01ms      1.17±0.01ms     1.12  rolling.Methods.time_rolling('DataFrame', 10, 'float', 'mean')
+     1.86±0.01ms      2.09±0.02ms     1.12  rolling.VariableWindowMethods.time_rolling('DataFrame', '50s', 'int', 'sum')
+         1.31±0s       1.47±0.05s     1.12  groupby.GroupByMethods.time_dtype_as_group('int', 'describe', 'direct')
+     1.23±0.01ms         1.38±0ms     1.12  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 1, 'higher')
+     1.07±0.01ms         1.20±0ms     1.12  rolling.Methods.time_rolling('DataFrame', 1000, 'int', 'mean')
+        1.24±0ms      1.39±0.01ms     1.12  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 0, 'linear')
+     5.05±0.02ms      5.66±0.02ms     1.12  categoricals.Concat.time_concat
+       228±0.8μs          255±1μs     1.12  groupby.GroupByMethods.time_dtype_as_field('datetime', 'nunique', 'direct')
+     4.96±0.09ms       5.56±0.1ms     1.12  io.csv.ReadUint64Integers.time_read_uint64_na_values
+     2.78±0.03ms      3.11±0.04ms     1.12  gil.ParallelRolling.time_rolling('mean')
+      58.6±0.5ms       65.6±0.2ms     1.12  frame_ctor.FromDicts.time_nested_dict_int64
+     1.36±0.01ms      1.52±0.02ms     1.12  rolling.Methods.time_rolling('Series', 1000, 'int', 'sum')
+      19.7±0.2ms       22.0±0.8ms     1.12  frame_methods.Iteration.time_iteritems
+        1.15±0ms      1.28±0.06ms     1.12  index_object.Ops.time_divide('float')
+        1.32±0ms      1.48±0.01ms     1.12  rolling.Methods.time_rolling('Series', 1000, 'float', 'sum')
+     1.97±0.01ms      2.20±0.02ms     1.12  binary_ops.Timeseries.time_timestamp_series_compare('US/Eastern')
+      2.06±0.01s       2.30±0.01s     1.12  groupby.GroupByMethods.time_dtype_as_group('float', 'describe', 'transformation')
+     1.97±0.03ms      2.20±0.02ms     1.12  timeseries.ResampleSeries.time_resample('period', '5min', 'ohlc')
+     1.65±0.02ms      1.85±0.03ms     1.12  timeseries.ResampleSeries.time_resample('period', '1D', 'ohlc')
+     1.36±0.01ms      1.52±0.01ms     1.12  rolling.Methods.time_rolling('Series', 10, 'int', 'sum')
+     1.23±0.01ms      1.38±0.01ms     1.12  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 0, 'higher')
+     1.23±0.01ms      1.37±0.01ms     1.12  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 1, 'nearest')
+      22.3±0.2ms       24.9±0.1ms     1.12  join_merge.Merge.time_merge_2intkey(True)
+     1.80±0.02ms      2.01±0.03ms     1.12  rolling.VariableWindowMethods.time_rolling('DataFrame', '1d', 'float', 'sum')
+         989±6μs      1.10±0.01ms     1.12  replace.FillNa.time_replace(True)
+      2.30±0.02s       2.56±0.01s     1.11  io.json.ReadJSON.time_read_json('index', 'int')
+     2.20±0.01ms      2.45±0.06ms     1.11  groupby.TransformBools.time_transform_mean
+        1.24±0ms      1.38±0.01ms     1.11  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 1, 'lower')
+     1.87±0.02ms      2.08±0.01ms     1.11  rolling.VariableWindowMethods.time_rolling('DataFrame', '1h', 'int', 'sum')
+        1.23±0ms      1.37±0.01ms     1.11  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 1, 'higher')
+      12.0±0.3μs      13.4±0.07μs     1.11  indexing.NumericSeriesIndexing.time_getitem_scalar(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
+      2.30±0.03s       2.56±0.03s     1.11  io.json.ReadJSON.time_read_json('index', 'datetime')
+      6.92±0.1ms       7.69±0.2ms     1.11  binary_ops.Ops.time_frame_mult(True, 'default')
+     1.25±0.01ms      1.39±0.01ms     1.11  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 0, 'midpoint')
+      1.32±0.01s       1.46±0.01s     1.11  groupby.GroupByMethods.time_dtype_as_group('int', 'describe', 'transformation')
+     10.0±0.06ms       11.1±0.5ms     1.11  eval.Eval.time_mult('numexpr', 1)
+       303±0.6μs          337±2μs     1.11  groupby.GroupByMethods.time_dtype_as_group('int', 'nunique', 'transformation')
+     1.24±0.01ms         1.38±0ms     1.11  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 0, 'lower')
+     1.24±0.01ms         1.37±0ms     1.11  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 1, 'midpoint')
+     1.75±0.03ms      1.94±0.01ms     1.11  rolling.Quantile.time_quantile('DataFrame', 1000, 'float', 1, 'midpoint')
+         250±4μs          277±2μs     1.11  reindex.Reindex.time_reindex_dates
+         897±4ms         993±10ms     1.11  groupby.GroupByMethods.time_dtype_as_field('int', 'describe', 'direct')
+       161±0.5μs        178±0.5μs     1.11  reindex.Fillna.time_float_32('backfill')
+      17.3±0.4ms       19.1±0.1ms     1.11  frame_ctor.FromDictwithTimestamp.time_dict_with_timestamp_offsets(<Nano>)
+      45.0±0.5ms       49.7±0.7ms     1.10  stat_ops.FrameMultiIndexOps.time_op(1, 'skew')
+      20.7±0.3ms       22.9±0.1ms     1.10  frame_ctor.FromDictwithTimestamp.time_dict_with_timestamp_offsets(<Hour>)
+     1.60±0.01ms      1.76±0.04ms     1.10  rolling.Quantile.time_quantile('DataFrame', 10, 'float', 0, 'lower')
+      22.3±0.3ms       24.6±0.4ms     1.10  frame_methods.SortIndexByColumns.time_frame_sort_values_by_columns
+         303±2μs          334±3μs     1.10  groupby.GroupByMethods.time_dtype_as_group('int', 'nunique', 'direct')
+         238±4ms         262±10ms     1.10  frame_methods.GetDtypeCounts.time_info
+         298±1μs          328±5μs     1.10  groupby.GroupByMethods.time_dtype_as_group('datetime', 'nunique', 'direct')
+         894±6ms          984±8ms     1.10  groupby.GroupByMethods.time_dtype_as_field('float', 'describe', 'direct')


SOME BENCHMARKS HAVE CHANGED SIGNIFICANTLY.

Speedups

-         246±2ns        224±0.8ns     0.91  timedelta.TimedeltaProperties.time_timedelta_nanoseconds
-      42.2±0.1ms       38.3±0.2ms     0.91  io.sql.WriteSQLDtypes.time_to_sql_dataframe_column('sqlalchemy', 'bool')
-        795±10μs          722±7μs     0.91  indexing.NumericSeriesIndexing.time_ix_array(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'unique_monotonic_inc')
-     7.47±0.05μs      6.78±0.06μs     0.91  offset.OnOffset.time_on_offset(<BusinessMonthBegin>)
-        70.4±3ms       63.8±0.2ms     0.91  join_merge.ConcatDataFrames.time_f_ordered(1, False)
-      33.9±0.3μs       30.7±0.9μs     0.91  offset.OffestDatetimeArithmetic.time_subtract_10(<Day>)
-     2.14±0.01ms      1.94±0.04ms     0.91  groupby.Datelike.time_sum('date_range_tz')
-      8.59±0.1μs       7.74±0.1μs     0.90  indexing.NumericSeriesIndexing.time_iloc_scalar(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
-      40.2±0.2ms      36.0±0.08ms     0.90  io.sql.WriteSQLDtypes.time_to_sql_dataframe_column('sqlalchemy', 'string')
-     6.35±0.09μs      5.69±0.06μs     0.90  timedelta.TimedeltaConstructor.time_from_datetime_timedelta
-     39.3±0.07ms      35.2±0.05ms     0.90  io.sql.WriteSQLDtypes.time_to_sql_dataframe_column('sqlalchemy', 'float')
-         133±3ms          119±4ms     0.89  gil.ParallelGroupbyMethods.time_loop(4, 'var')
-      39.0±0.1ms       34.8±0.2ms     0.89  io.sql.WriteSQLDtypes.time_to_sql_dataframe_column('sqlalchemy', 'int')
-       143±0.9ms          128±1ms     0.89  categoricals.Constructor.time_with_nan
-     2.71±0.02ms      2.42±0.04ms     0.89  frame_methods.NSort.time_nlargest_one_column('last')
-      28.4±0.3ms       25.3±0.2ms     0.89  stat_ops.FrameOps.time_op('kurt', 'int', 1, False)
-      7.94±0.2μs      7.06±0.07μs     0.89  offset.OnOffset.time_on_offset(<SemiMonthEnd: day_of_month=15>)
-        97.7±1μs       86.7±0.5μs     0.89  index_object.SetOperations.time_operation('datetime', 'union')
-     8.58±0.04μs      7.59±0.05μs     0.88  indexing.NumericSeriesIndexing.time_iloc_scalar(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'nonunique_monotonic_inc')
-     8.56±0.04μs      7.54±0.03μs     0.88  indexing.NumericSeriesIndexing.time_iloc_scalar(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'unique_monotonic_inc')
-        1.13±0ms          994±4μs     0.88  indexing.CategoricalIndexIndexing.time_getitem_bool_array('non_monotonic')
-      5.51±0.1ms      4.84±0.03ms     0.88  timeseries.Factorize.time_factorize(None)
-      8.63±0.1μs      7.59±0.04μs     0.88  indexing.NumericSeriesIndexing.time_iloc_scalar(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
-        72.2±2ms       63.4±0.3ms     0.88  join_merge.ConcatDataFrames.time_f_ordered(1, True)
-      16.4±0.5μs       14.4±0.2μs     0.88  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<BusinessDay>)
-      20.5±0.3ms       17.8±0.2ms     0.87  timeseries.DatetimeIndex.time_normalize('tz_aware')
-      18.3±0.4μs       15.9±0.2μs     0.87  offset.OffestDatetimeArithmetic.time_add_10(<BusinessDay>)
-         114±3ms         98.5±3ms     0.86  gil.ParallelGroupbyMethods.time_loop(4, 'prod')
-       125±0.9μs          107±1μs     0.86  offset.OffestDatetimeArithmetic.time_add(<CustomBusinessMonthBegin>)
-     14.1±0.06μs      12.1±0.05μs     0.86  offset.OffestDatetimeArithmetic.time_apply(<BusinessDay>)
-         139±2μs          119±2μs     0.86  offset.OffestDatetimeArithmetic.time_add_10(<CustomBusinessMonthBegin>)
-      18.6±0.1μs       15.9±0.4μs     0.86  timeseries.AsOf.time_asof_single('Series')
-      9.41±0.1ms       8.03±0.2ms     0.85  strings.Cat.time_cat(0, ',', None, 0.001)
-       124±0.2μs        105±0.5μs     0.85  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<CustomBusinessMonthBegin>)
-         192±1ms          163±1ms     0.85  io.stata.Stata.time_read_stata('tc')
-         193±1ms          163±3ms     0.84  io.stata.Stata.time_read_stata('td')
-      9.00±0.3μs      7.57±0.05μs     0.84  timeseries.AsOf.time_asof_single_early('Series')
-         105±1ms         88.6±1ms     0.84  indexing.NumericSeriesIndexing.time_loc_list_like(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'unique_monotonic_inc')
-      36.2±0.2ms       30.4±0.2ms     0.84  plotting.TimeseriesPlotting.time_plot_irregular
-      5.95±0.1ms      4.97±0.09ms     0.84  timeseries.Factorize.time_factorize('Asia/Tokyo')
-       122±0.8μs        101±0.3μs     0.83  offset.OffestDatetimeArithmetic.time_apply(<CustomBusinessMonthBegin>)
-      37.4±0.3μs       30.9±0.1μs     0.83  indexing.NumericSeriesIndexing.time_loc_scalar(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
-       129±0.9ms          107±2ms     0.82  offset.OffsetSeriesArithmetic.time_add_offset(<CustomBusinessMonthBegin>)
-      35.4±0.5ms       28.9±0.5ms     0.82  plotting.TimeseriesPlotting.time_plot_regular_compat
-         130±1ms        106±0.8ms     0.81  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<CustomBusinessMonthBegin>)
-      15.1±0.3μs       12.3±0.2μs     0.81  timedelta.TimedeltaConstructor.time_from_components
-      23.5±0.1μs      19.0±0.03μs     0.81  offset.OffestDatetimeArithmetic.time_subtract_10(<BusinessDay>)
-     4.69±0.03ms       3.79±0.2ms     0.81  frame_methods.NSort.time_nsmallest_two_columns('first')
-        11.1±2ms       8.93±0.2ms     0.80  strings.Cat.time_cat(0, ',', None, 0.15)
-         120±1μs         95.9±2μs     0.80  offset.OffestDatetimeArithmetic.time_subtract_10(<CustomBusinessMonthBegin>)
-         116±1μs       92.1±0.6μs     0.79  offset.OffestDatetimeArithmetic.time_subtract(<CustomBusinessMonthBegin>)
-      16.3±0.4μs       12.9±0.1μs     0.79  offset.OffestDatetimeArithmetic.time_add(<BusinessDay>)
-      9.86±0.2ms      7.81±0.03ms     0.79  series_methods.ValueCounts.time_value_counts('object')
-     4.80±0.02ms      3.76±0.07ms     0.78  frame_methods.NSort.time_nsmallest_two_columns('last')
-         365±5ns          284±2ns     0.78  indexing.MethodLookup.time_lookup_iloc
-     1.83±0.05ms      1.42±0.02ms     0.77  series_methods.NSort.time_nlargest('last')
-         140±2ms          108±3ms     0.77  reshape.Unstack.time_without_last_row
-         373±3ns          284±2ns     0.76  indexing.MethodLookup.time_lookup_loc
-     1.25±0.01ms         946±20μs     0.76  stat_ops.SeriesOps.time_op('median', 'int', True)
-     4.99±0.04μs      3.78±0.05μs     0.76  timeseries.DatetimeIndex.time_get('dst')
-      19.0±0.1ms       14.3±0.4ms     0.76  algorithms.Factorize.time_factorize_int(True)
-     1.25±0.01ms          941±8μs     0.75  stat_ops.SeriesOps.time_op('median', 'int', False)
-     5.04±0.09μs      3.79±0.04μs     0.75  timeseries.DatetimeIndex.time_get('tz_naive')
-      20.9±0.2ms       15.6±0.6ms     0.75  index_object.SetOperations.time_operation('strings', 'symmetric_difference')
-        10.8±1ms       8.05±0.2ms     0.74  timeseries.AsOf.time_asof_nan('DataFrame')
-     1.78±0.01ms      1.33±0.02ms     0.74  series_methods.NSort.time_nlargest('first')
-        94.6±1ms         69.9±4ms     0.74  frame_methods.Describe.time_series_describe
-         319±1ms          235±6ms     0.74  frame_methods.Describe.time_dataframe_describe
-      11.9±0.1ms      8.78±0.02ms     0.74  io.hdf.HDFStoreDataFrame.time_store_info
-       103±0.2μs       74.6±0.3μs     0.73  indexing.NumericSeriesIndexing.time_loc_scalar(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'unique_monotonic_inc')
-      21.7±0.2μs       15.8±0.2μs     0.73  offset.OffestDatetimeArithmetic.time_subtract(<YearBegin: month=1>)
-      54.1±0.8μs       39.1±0.3μs     0.72  indexing.NumericSeriesIndexing.time_loc_scalar(<class 'pandas.core.indexes.numeric.Float64Index'>, 'unique_monotonic_inc')
-       736±200μs          527±1μs     0.72  timeseries.InferFreq.time_infer_freq(None)
-     5.11±0.09ms      3.66±0.02ms     0.72  offset.OnOffset.time_on_offset(<CustomBusinessMonthBegin>)
-         188±2ms        133±0.3ms     0.71  timeseries.DatetimeIndex.time_to_pydatetime('tz_aware')
-     1.08±0.01ms         765±10μs     0.71  indexing.NumericSeriesIndexing.time_getitem_array(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
-      10.9±0.1ms      7.59±0.02ms     0.70  inference.DateInferOps.time_subtract_datetimes
-      22.2±0.1μs       15.5±0.2μs     0.70  offset.OffestDatetimeArithmetic.time_subtract(<YearEnd: month=12>)
-     2.37±0.02ms      1.64±0.03ms     0.69  groupby.RankWithTies.time_rank_ties('int64', 'dense')
-      21.2±0.5μs      14.6±0.09μs     0.69  offset.OffestDatetimeArithmetic.time_add_10(<YearBegin: month=1>)
-         282±3ns          194±1ns     0.69  timedelta.TimedeltaProperties.time_timedelta_days
-     2.37±0.01ms      1.63±0.01ms     0.69  groupby.RankWithTies.time_rank_ties('int64', 'min')
-        1.74±0ms      1.19±0.01ms     0.69  series_methods.NSort.time_nsmallest('first')
-         145±2ms       99.2±0.5ms     0.68  indexing.NumericSeriesIndexing.time_getitem_lists(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
-     11.1±0.06ms       7.55±0.1ms     0.68  frame_methods.ToString.time_to_string_floats
-      1.14±0.3ms          777±4μs     0.68  timeseries.InferFreq.time_infer_freq('B')
-     2.37±0.03ms      1.61±0.01ms     0.68  groupby.RankWithTies.time_rank_ties('int64', 'average')
-     2.38±0.01ms      1.62±0.02ms     0.68  groupby.RankWithTies.time_rank_ties('int64', 'max')
-     2.40±0.01ms      1.63±0.01ms     0.68  groupby.RankWithTies.time_rank_ties('float32', 'first')
-        17.0±1ms       11.4±0.1ms     0.67  categoricals.ValueCounts.time_value_counts(False)
-     2.41±0.02ms      1.61±0.01ms     0.67  groupby.RankWithTies.time_rank_ties('int64', 'first')
-     2.41±0.02ms      1.61±0.01ms     0.67  groupby.RankWithTies.time_rank_ties('float64', 'first')
-     1.54±0.01ms      1.03±0.01ms     0.67  series_methods.NSort.time_nsmallest('last')
-     6.44±0.03ms      4.28±0.05ms     0.67  stat_ops.FrameOps.time_op('median', 'int', 0, False)
-         173±1μs        115±0.7μs     0.66  offset.OffestDatetimeArithmetic.time_subtract(<CustomBusinessMonthEnd>)
-     2.47±0.02ms      1.63±0.01ms     0.66  groupby.RankWithTies.time_rank_ties('float64', 'dense')
-     6.53±0.03ms      4.30±0.04ms     0.66  stat_ops.FrameOps.time_op('median', 'int', 0, True)
-     2.46±0.01ms      1.62±0.02ms     0.66  groupby.RankWithTies.time_rank_ties('float64', 'max')
-     2.46±0.02ms      1.61±0.01ms     0.66  groupby.RankWithTies.time_rank_ties('float64', 'min')
-     2.44±0.01ms      1.59±0.01ms     0.65  groupby.RankWithTies.time_rank_ties('datetime64', 'dense')
-     2.48±0.01ms      1.62±0.01ms     0.65  groupby.RankWithTies.time_rank_ties('float64', 'average')
-     2.48±0.02ms      1.61±0.01ms     0.65  groupby.RankWithTies.time_rank_ties('float32', 'max')
-     2.46±0.02ms      1.60±0.01ms     0.65  groupby.RankWithTies.time_rank_ties('datetime64', 'average')
-     2.45±0.02ms      1.59±0.02ms     0.65  groupby.RankWithTies.time_rank_ties('datetime64', 'min')
-        2.49±0ms      1.61±0.01ms     0.65  groupby.RankWithTies.time_rank_ties('float32', 'min')
-     2.50±0.02ms      1.61±0.01ms     0.64  groupby.RankWithTies.time_rank_ties('float32', 'dense')
-     2.49±0.03ms      1.60±0.02ms     0.64  groupby.RankWithTies.time_rank_ties('datetime64', 'first')
-     2.51±0.03ms         1.61±0ms     0.64  groupby.RankWithTies.time_rank_ties('float32', 'average')
-         304±2ns        195±0.9ns     0.64  timedelta.TimedeltaProperties.time_timedelta_microseconds
-     5.25±0.05ms      3.35±0.02ms     0.64  offset.OnOffset.time_on_offset(<CustomBusinessMonthEnd>)
-        948±20μs         605±10μs     0.64  indexing.NumericSeriesIndexing.time_loc_array(<class 'pandas.core.indexes.numeric.Int64Index'>, 'unique_monotonic_inc')
-     2.53±0.07ms      1.61±0.02ms     0.64  groupby.RankWithTies.time_rank_ties('datetime64', 'max')
-     3.32±0.01ms      2.10±0.01ms     0.63  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x11332b6a8>, True)
-         105±1ms       66.0±0.6ms     0.63  index_object.IndexAppend.time_append_int_list
-      25.7±0.2μs      16.1±0.08μs     0.62  offset.OffestDatetimeArithmetic.time_subtract(<QuarterEnd: startingMonth=3>)
-     2.38±0.01ms      1.47±0.01ms     0.62  period.Algorithms.time_value_counts('series')
-      29.1±0.5μs       17.9±0.1μs     0.62  offset.OffestDatetimeArithmetic.time_subtract_10(<QuarterEnd: startingMonth=3>)
-         471±3ms          284±2ms     0.60  indexing.NonNumericSeriesIndexing.time_getitem_list_like('datetime', 'nonunique_monotonic_inc')
-       141±0.7μs       84.5±0.3μs     0.60  offset.OffestDatetimeArithmetic.time_subtract_10(<CustomBusinessMonthEnd>)
-      50.1±0.5μs       30.0±0.5μs     0.60  categoricals.IsMonotonic.time_categorical_series_is_monotonic_increasing
-      26.8±0.2μs       15.9±0.3μs     0.59  offset.OffestDatetimeArithmetic.time_subtract(<BusinessQuarterEnd: startingMonth=3>)
-      49.7±0.6μs       29.0±0.2μs     0.58  categoricals.IsMonotonic.time_categorical_series_is_monotonic_decreasing
-     30.7±0.08μs       17.8±0.1μs     0.58  offset.OffestDatetimeArithmetic.time_subtract_10(<BusinessQuarterEnd: startingMonth=3>)
-     3.13±0.03ms      1.81±0.02ms     0.58  indexing.DataFrameNumericIndexing.time_loc_dups
-      23.9±0.7μs       13.8±0.1μs     0.58  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<BusinessQuarterEnd: startingMonth=3>)
-      29.7±0.4μs      17.2±0.09μs     0.58  offset.OffestDatetimeArithmetic.time_subtract_10(<YearBegin: month=1>)
-       137±0.5μs       78.2±0.9μs     0.57  offset.OffestDatetimeArithmetic.time_add_10(<CustomBusinessMonthEnd>)
-      26.6±0.3μs       15.2±0.3μs     0.57  offset.OffestDatetimeArithmetic.time_add_10(<QuarterEnd: startingMonth=3>)
-       359±0.9ms          205±2ms     0.57  reindex.Reindex.time_reindex_multiindex
-      23.8±0.2μs       13.6±0.5μs     0.57  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<QuarterEnd: startingMonth=3>)
-        20.2±1ms       11.5±0.1ms     0.57  categoricals.ValueCounts.time_value_counts(True)
-     20.2±0.07μs      11.5±0.05μs     0.57  offset.OffestDatetimeArithmetic.time_apply(<QuarterEnd: startingMonth=3>)
-         346±2ns          196±1ns     0.57  timedelta.TimedeltaProperties.time_timedelta_seconds
-      86.2±0.8ms         48.6±3ms     0.56  reshape.Unstack.time_full_product
-      41.5±0.2ms       23.3±0.2ms     0.56  categoricals.Constructor.time_all_nan
-         379±1ms          211±2ms     0.56  series_methods.SeriesConstructor.time_constructor('dict')
-         544±7ns          303±2ns     0.56  timestamp.TimestampProperties.time_freqstr(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
-     13.6±0.07ms       7.37±0.4ms     0.54  binary_ops.Timeseries.time_timestamp_ops_diff('US/Eastern')
-       141±0.6ms         75.9±1ms     0.54  indexing.NumericSeriesIndexing.time_getitem_array(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
-     22.7±0.08μs      12.2±0.06μs     0.54  offset.OffestDatetimeArithmetic.time_add(<QuarterEnd: startingMonth=3>)
-      21.4±0.2μs       11.4±0.2μs     0.53  offset.OffestDatetimeArithmetic.time_apply(<BusinessQuarterEnd: startingMonth=3>)
-       141±0.6ms       75.4±0.6ms     0.53  indexing.NumericSeriesIndexing.time_loc_array(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
-         294±3μs          156±2μs     0.53  indexing.CategoricalIndexIndexing.time_getitem_bool_array('monotonic_incr')
-      23.8±0.1μs      12.5±0.05μs     0.53  offset.OffestDatetimeArithmetic.time_add(<BusinessQuarterEnd: startingMonth=3>)
-       128±0.9ms         66.8±2ms     0.52  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<CustomBusinessMonthEnd>)
-        582±50ns          303±4ns     0.52  timestamp.TimestampProperties.time_freqstr(None, 'B')
-       126±0.8μs       65.1±0.3μs     0.52  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<CustomBusinessMonthEnd>)
-      31.3±0.2μs      16.0±0.04μs     0.51  offset.OffestDatetimeArithmetic.time_subtract(<QuarterBegin: startingMonth=3>)
-      35.3±0.2μs       18.0±0.3μs     0.51  offset.OffestDatetimeArithmetic.time_subtract_10(<QuarterBegin: startingMonth=3>)
-         129±1ms       64.8±0.8ms     0.50  offset.OffsetSeriesArithmetic.time_add_offset(<CustomBusinessMonthEnd>)
-         122±1μs       61.3±0.9μs     0.50  offset.OffestDatetimeArithmetic.time_apply(<CustomBusinessMonthEnd>)
-       124±0.6μs       62.1±0.4μs     0.50  offset.OffestDatetimeArithmetic.time_add(<CustomBusinessMonthEnd>)
-      30.2±0.1μs      14.9±0.05μs     0.49  offset.OffestDatetimeArithmetic.time_add_10(<YearEnd: month=12>)
-      32.7±0.6μs       16.2±0.1μs     0.49  offset.OffestDatetimeArithmetic.time_subtract(<BusinessQuarterBegin: startingMonth=3>)
-      36.2±0.2μs       17.7±0.1μs     0.49  offset.OffestDatetimeArithmetic.time_subtract_10(<BusinessQuarterBegin: startingMonth=3>)
-      31.7±0.1μs       15.4±0.2μs     0.49  offset.OffestDatetimeArithmetic.time_add_10(<QuarterBegin: startingMonth=3>)
-     1.39±0.01ms          671±8μs     0.48  series_methods.Map.time_map('dict')
-      95.7±0.5ms       45.8±0.3ms     0.48  rolling.Quantile.time_quantile('Series', 10, 'float', 0.5, 'linear')
-      95.6±0.8ms       45.4±0.4ms     0.48  rolling.Quantile.time_quantile('Series', 10, 'float', 0.5, 'midpoint')
-      28.3±0.3μs       13.4±0.1μs     0.47  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<MonthBegin>)
-      28.7±0.2μs      13.6±0.06μs     0.47  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<QuarterBegin: startingMonth=3>)
-      36.9±0.1μs       17.5±0.1μs     0.47  offset.OffestDatetimeArithmetic.time_subtract_10(<YearEnd: month=12>)
-      95.2±0.4ms       44.4±0.2ms     0.47  rolling.Quantile.time_quantile('DataFrame', 10, 'float', 0.5, 'midpoint')
-      96.3±0.4ms       44.8±0.6ms     0.47  rolling.Quantile.time_quantile('DataFrame', 10, 'float', 0.5, 'linear')
-      95.2±0.7ms       44.1±0.5ms     0.46  rolling.Quantile.time_quantile('Series', 10, 'float', 0.5, 'higher')
-      28.8±0.3μs      13.3±0.07μs     0.46  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<BusinessYearBegin: month=1>)
-      28.7±0.3μs      13.2±0.06μs     0.46  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<BusinessMonthEnd>)
-        95.3±1ms       43.9±0.3ms     0.46  rolling.Quantile.time_quantile('Series', 10, 'float', 0.5, 'lower')
-      30.0±0.2μs      13.7±0.08μs     0.46  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<BusinessQuarterBegin: startingMonth=3>)
-      33.7±0.3μs       15.4±0.6μs     0.46  offset.OffestDatetimeArithmetic.time_add_10(<BusinessQuarterBegin: startingMonth=3>)
-      95.3±0.1ms       43.1±0.3ms     0.45  rolling.Quantile.time_quantile('DataFrame', 10, 'float', 0.5, 'higher')
-      96.0±0.3ms      43.4±0.09ms     0.45  rolling.Quantile.time_quantile('DataFrame', 10, 'float', 0.5, 'nearest')
-      95.4±0.5ms       43.0±0.5ms     0.45  rolling.Quantile.time_quantile('Series', 10, 'float', 0.5, 'nearest')
-      57.8±0.8ms         26.1±1ms     0.45  binary_ops.Timeseries.time_timestamp_ops_diff_with_shift(None)
-      95.4±0.3ms       42.9±0.4ms     0.45  rolling.Quantile.time_quantile('DataFrame', 10, 'float', 0.5, 'lower')
-     1.52±0.02ms          683±3μs     0.45  offset.OffsetSeriesArithmetic.time_add_offset(<QuarterBegin: startingMonth=3>)
-      95.1±0.3ms       42.6±0.5ms     0.45  rolling.Quantile.time_quantile('Series', 10, 'int', 0.5, 'linear')
-     29.0±0.06μs       12.9±0.2μs     0.45  offset.OffestDatetimeArithmetic.time_add(<BusinessQuarterBegin: startingMonth=3>)
-     4.48±0.03ms      1.99±0.03ms     0.44  ctors.SeriesConstructors.time_series_constructor(<function SeriesConstructors.<lambda> at 0x11332b6a8>, False)
-      32.3±0.1μs       14.3±0.3μs     0.44  offset.OffestDatetimeArithmetic.time_add_10(<MonthBegin>)
-      95.9±0.7ms       42.3±0.2ms     0.44  rolling.Quantile.time_quantile('Series', 10, 'int', 0.5, 'midpoint')
-      95.2±0.6ms       41.9±0.6ms     0.44  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 0.5, 'linear')
-     25.3±0.05μs      11.1±0.06μs     0.44  offset.OffestDatetimeArithmetic.time_apply(<QuarterBegin: startingMonth=3>)
-      36.4±0.6μs       16.0±0.5μs     0.44  offset.OffestDatetimeArithmetic.time_subtract_10(<MonthBegin>)
-     28.1±0.05μs       12.3±0.2μs     0.44  offset.OffestDatetimeArithmetic.time_add(<QuarterBegin: startingMonth=3>)
-      25.2±0.4μs       11.0±0.2μs     0.44  offset.OffestDatetimeArithmetic.time_apply(<BusinessMonthEnd>)
-      95.9±0.3ms       41.5±0.1ms     0.43  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 0.5, 'midpoint')
-      26.3±0.2μs       11.4±0.1μs     0.43  offset.OffestDatetimeArithmetic.time_apply(<BusinessQuarterBegin: startingMonth=3>)
-     1.56±0.02ms          675±3μs     0.43  offset.OffsetSeriesArithmetic.time_add_offset(<YearBegin: month=1>)
-      33.5±0.3μs       14.5±0.3μs     0.43  offset.OffestDatetimeArithmetic.time_subtract(<MonthBegin>)
-      25.0±0.3μs      10.8±0.03μs     0.43  offset.OffestDatetimeArithmetic.time_apply(<MonthBegin>)
-      94.4±0.8ms       40.6±0.4ms     0.43  rolling.Quantile.time_quantile('Series', 10, 'int', 0.5, 'nearest')
-      33.8±0.2μs       14.5±0.4μs     0.43  offset.OffestDatetimeArithmetic.time_add_10(<BusinessMonthEnd>)
-      95.3±0.8ms       40.7±0.4ms     0.43  rolling.Quantile.time_quantile('Series', 10, 'int', 0.5, 'lower')
-      95.4±0.3ms       40.6±0.9ms     0.43  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 0.5, 'higher')
-      95.9±0.9ms       40.7±0.1ms     0.42  rolling.Quantile.time_quantile('Series', 10, 'int', 0.5, 'higher')
-      28.1±0.2μs       11.9±0.1μs     0.42  offset.OffestDatetimeArithmetic.time_add(<MonthBegin>)
-      34.9±0.4μs       14.7±0.3μs     0.42  offset.OffestDatetimeArithmetic.time_subtract(<BusinessMonthEnd>)
-     28.8±0.08μs      12.1±0.04μs     0.42  offset.OffestDatetimeArithmetic.time_add(<BusinessYearBegin: month=1>)
-      95.3±0.8ms      39.9±0.08ms     0.42  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 0.5, 'lower')
-     31.9±0.04μs       13.4±0.1μs     0.42  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<BusinessMonthBegin>)
-      28.6±0.5μs       11.9±0.2μs     0.42  offset.OffestDatetimeArithmetic.time_add(<BusinessMonthEnd>)
-      26.5±0.2μs      11.0±0.07μs     0.42  offset.OffestDatetimeArithmetic.time_apply(<BusinessYearBegin: month=1>)
-      38.9±0.3μs      16.1±0.08μs     0.41  offset.OffestDatetimeArithmetic.time_subtract_10(<BusinessMonthBegin>)
-      95.7±0.3ms       39.5±0.4ms     0.41  rolling.Quantile.time_quantile('DataFrame', 10, 'int', 0.5, 'nearest')
-      38.7±0.5μs       15.9±0.1μs     0.41  offset.OffestDatetimeArithmetic.time_subtract_10(<BusinessMonthEnd>)
-      35.5±0.3μs       14.5±0.3μs     0.41  offset.OffestDatetimeArithmetic.time_subtract(<BusinessMonthBegin>)
-      48.9±0.5μs       19.3±0.2μs     0.39  offset.OffestDatetimeArithmetic.time_subtract_10(<SemiMonthBegin: day_of_month=15>)
-      50.9±0.6ms       20.0±0.2ms     0.39  indexing.CategoricalIndexIndexing.time_get_indexer_list('monotonic_incr')
-         197±2ms       77.1±0.8ms     0.39  indexing.NumericSeriesIndexing.time_getitem_lists(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'nonunique_monotonic_inc')
-        49.7±1μs       19.3±0.1μs     0.39  offset.OffestDatetimeArithmetic.time_subtract_10(<SemiMonthEnd: day_of_month=15>)
-         193±3ms       74.1±0.6ms     0.38  indexing.NumericSeriesIndexing.time_loc_array(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'nonunique_monotonic_inc')
-      32.0±0.7μs       12.2±0.2μs     0.38  offset.OffestDatetimeArithmetic.time_add(<BusinessMonthBegin>)
-         196±2ms       74.5±0.7ms     0.38  indexing.NumericSeriesIndexing.time_getitem_array(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'nonunique_monotonic_inc')
-     1.86±0.01ms          708±6μs     0.38  period.Algorithms.time_drop_duplicates('series')
-      29.2±0.2μs      11.0±0.06μs     0.38  offset.OffestDatetimeArithmetic.time_apply(<BusinessMonthBegin>)
-        43.3±2μs       16.3±0.1μs     0.38  offset.OffestDatetimeArithmetic.time_subtract(<BusinessYearBegin: month=1>)
-      40.0±0.5μs       15.1±0.2μs     0.38  offset.OffestDatetimeArithmetic.time_add_10(<BusinessYearBegin: month=1>)
-      37.5±0.4μs      14.1±0.07μs     0.38  offset.OffestDatetimeArithmetic.time_add_10(<BusinessMonthBegin>)
-      39.7±0.2μs       14.9±0.2μs     0.38  offset.OffestDatetimeArithmetic.time_add_10(<BusinessYearEnd: month=12>)
-        42.1±2μs       15.7±0.3μs     0.37  offset.OffestDatetimeArithmetic.time_subtract(<BusinessYearEnd: month=12>)
-      46.8±0.6μs       17.2±0.4μs     0.37  offset.OffestDatetimeArithmetic.time_subtract(<SemiMonthBegin: day_of_month=15>)
-      44.9±0.1μs       16.3±0.1μs     0.36  offset.OffestDatetimeArithmetic.time_add_10(<SemiMonthBegin: day_of_month=15>)
-         1.75±0s          630±3ms     0.36  reshape.GetDummies.time_get_dummies_1d_sparse
-      47.7±0.5μs       17.0±0.2μs     0.36  offset.OffestDatetimeArithmetic.time_subtract(<SemiMonthEnd: day_of_month=15>)
-      50.0±0.5μs       17.8±0.3μs     0.36  offset.OffestDatetimeArithmetic.time_subtract_10(<BusinessYearEnd: month=12>)
-      46.0±0.2μs       16.2±0.1μs     0.35  offset.OffestDatetimeArithmetic.time_add_10(<SemiMonthEnd: day_of_month=15>)
-      50.1±0.4μs       17.4±0.4μs     0.35  offset.OffestDatetimeArithmetic.time_subtract_10(<BusinessYearBegin: month=1>)
-        43.8±2μs      14.2±0.04μs     0.32  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<SemiMonthEnd: day_of_month=15>)
-        44.2±3μs      14.2±0.09μs     0.32  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<SemiMonthBegin: day_of_month=15>)
-      41.8±0.2μs       13.2±0.1μs     0.32  offset.OffestDatetimeArithmetic.time_add(<SemiMonthBegin: day_of_month=15>)
-      39.3±0.5μs      12.0±0.07μs     0.31  offset.OffestDatetimeArithmetic.time_apply(<SemiMonthBegin: day_of_month=15>)
-      52.7±0.6μs       15.9±0.1μs     0.30  offset.OffestDatetimeArithmetic.time_subtract_10(<MonthEnd>)
-        43.3±1μs       13.0±0.2μs     0.30  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<MonthEnd>)
-      43.1±0.6μs      12.9±0.08μs     0.30  offset.OffestDatetimeArithmetic.time_add(<SemiMonthEnd: day_of_month=15>)
-         233±3ms         69.4±2ms     0.30  rolling.Quantile.time_quantile('Series', 1000, 'float', 0.5, 'linear')
-      49.2±0.2μs      14.7±0.07μs     0.30  offset.OffestDatetimeArithmetic.time_subtract(<MonthEnd>)
-         233±3ms         69.3±1ms     0.30  rolling.Quantile.time_quantile('Series', 1000, 'float', 0.5, 'midpoint')
-         232±2ms       67.6±0.5ms     0.29  rolling.Quantile.time_quantile('DataFrame', 1000, 'float', 0.5, 'linear')
-      40.4±0.3μs      11.8±0.09μs     0.29  offset.OffestDatetimeArithmetic.time_apply(<SemiMonthEnd: day_of_month=15>)
-      48.1±0.8μs       14.0±0.3μs     0.29  offset.OffestDatetimeArithmetic.time_add_10(<MonthEnd>)
-        52.4±1μs       15.2±0.3μs     0.29  offset.OffestDatetimeArithmetic.time_add_10(<BusinessQuarterEnd: startingMonth=3>)
-     71.1±0.09μs       20.6±0.2μs     0.29  indexing.CategoricalIndexIndexing.time_getitem_list_like('non_monotonic')
-         235±3ms       67.0±0.1ms     0.29  rolling.Quantile.time_quantile('DataFrame', 1000, 'float', 0.5, 'midpoint')
-      70.4±0.3μs       20.1±0.1μs     0.29  indexing.CategoricalIndexIndexing.time_getitem_list_like('monotonic_incr')
-         230±2ms         65.5±4ms     0.28  rolling.Quantile.time_quantile('Series', 1000, 'float', 0.5, 'nearest')
-      42.2±0.1μs       11.9±0.1μs     0.28  offset.OffestDatetimeArithmetic.time_add(<MonthEnd>)
-         205±1ms       57.3±0.2ms     0.28  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 0.5, 'midpoint')
-         234±2ms       65.2±0.3ms     0.28  rolling.Quantile.time_quantile('Series', 1000, 'float', 0.5, 'lower')
-         233±3ms         64.7±2ms     0.28  rolling.Quantile.time_quantile('Series', 1000, 'float', 0.5, 'higher')
-         208±2ms       57.5±0.5ms     0.28  rolling.Quantile.time_quantile('Series', 1000, 'int', 0.5, 'midpoint')
-        233±10ms       64.4±0.8ms     0.28  rolling.Quantile.time_quantile('DataFrame', 1000, 'float', 0.5, 'nearest')
-      39.3±0.3μs      10.8±0.05μs     0.28  offset.OffestDatetimeArithmetic.time_apply(<MonthEnd>)
-       206±0.9ms       56.7±0.6ms     0.27  rolling.Quantile.time_quantile('Series', 1000, 'int', 0.5, 'linear')
-         205±1ms       55.6±0.4ms     0.27  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 0.5, 'linear')
-         236±3ms       63.0±0.3ms     0.27  rolling.Quantile.time_quantile('DataFrame', 1000, 'float', 0.5, 'lower')
-      50.6±0.5μs      13.3±0.03μs     0.26  offset.OffestDatetimeArithmetic.time_apply_np_dt64(<BusinessYearEnd: month=12>)
-        242±10ms         63.7±1ms     0.26  rolling.Quantile.time_quantile('DataFrame', 1000, 'float', 0.5, 'higher')
-         625±4μs          157±1μs     0.25  groupby.GroupByMethods.time_dtype_as_group('object', 'bfill', 'direct')
-         810±3μs          203±3μs     0.25  series_methods.Map.time_map('Series')
-         625±6μs          156±1μs     0.25  groupby.GroupByMethods.time_dtype_as_group('object', 'ffill', 'direct')
-         628±2μs        156±0.6μs     0.25  groupby.GroupByMethods.time_dtype_as_group('object', 'ffill', 'transformation')
-        636±10μs          156±2μs     0.25  groupby.GroupByMethods.time_dtype_as_group('object', 'bfill', 'transformation')
-      49.1±0.5μs       12.0±0.1μs     0.24  offset.OffestDatetimeArithmetic.time_add(<BusinessYearEnd: month=12>)
-         206±1ms       49.8±0.5ms     0.24  rolling.Quantile.time_quantile('Series', 1000, 'int', 0.5, 'lower')
-         206±1ms       49.5±0.7ms     0.24  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 0.5, 'nearest')
-     5.56±0.03ms      1.34±0.01ms     0.24  series_methods.Dir.time_dir_strings
-         206±2ms       49.5±0.3ms     0.24  rolling.Quantile.time_quantile('Series', 1000, 'int', 0.5, 'nearest')
-       206±0.7ms       49.2±0.3ms     0.24  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 0.5, 'higher')
-         206±2ms       49.1±0.6ms     0.24  rolling.Quantile.time_quantile('Series', 1000, 'int', 0.5, 'higher')
-      46.7±0.6μs      11.1±0.09μs     0.24  offset.OffestDatetimeArithmetic.time_apply(<BusinessYearEnd: month=12>)
-       206±0.9ms       48.7±0.5ms     0.24  rolling.Quantile.time_quantile('DataFrame', 1000, 'int', 0.5, 'lower')
-     1.04±0.02ms          242±2μs     0.23  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<QuarterBegin: startingMonth=3>)
-     1.06±0.01ms          239±1μs     0.22  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<YearBegin: month=1>)
-         302±1μs       67.4±0.4μs     0.22  groupby.GroupByMethods.time_dtype_as_group('object', 'all', 'transformation')
-         302±1μs       66.9±0.9μs     0.22  groupby.GroupByMethods.time_dtype_as_group('object', 'any', 'direct')
-         302±3μs       66.8±0.6μs     0.22  groupby.GroupByMethods.time_dtype_as_group('object', 'all', 'direct')
-         305±2μs       67.0±0.4μs     0.22  groupby.GroupByMethods.time_dtype_as_group('object', 'any', 'transformation')
-         512±9ms          111±2ms     0.22  timeseries.DatetimeIndex.time_to_date('tz_aware')
-         528±5ms          111±2ms     0.21  timeseries.DatetimeIndex.time_to_time('tz_aware')
-      17.2±0.2ms      3.49±0.09ms     0.20  groupby.Datelike.time_sum('period_range')
-         457±3ms         92.0±1ms     0.20  multiindex_object.GetLoc.time_large_get_loc_warm
-      63.2±0.3μs       12.5±0.1μs     0.20  indexing.CategoricalIndexIndexing.time_getitem_slice('monotonic_decr')
-     2.47±0.06ms        460±0.6μs     0.19  offset.ApplyIndex.time_apply_index(<QuarterBegin: startingMonth=3>)
-     2.37±0.02ms          416±2μs     0.18  offset.ApplyIndex.time_apply_index(<YearBegin: month=1>)
-      76.8±0.4μs       12.2±0.1μs     0.16  indexing.CategoricalIndexIndexing.time_getitem_slice('monotonic_incr')
-      76.7±0.3μs      11.9±0.04μs     0.16  indexing.CategoricalIndexIndexing.time_getitem_slice('non_monotonic')
-     3.64±0.01ms         540±10μs     0.15  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.UInt64Engine'>, <class 'numpy.float64'>, 'non_monotonic')
-        58.2±2ms       8.51±0.1ms     0.15  categoricals.Isin.time_isin_categorical('object')
-     3.65±0.01ms         527±10μs     0.14  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.UInt64Engine'>, <class 'numpy.float64'>, 'monotonic_decr')
-      45.7±0.5μs      5.97±0.04μs     0.13  offset.OnOffset.time_on_offset(<QuarterEnd: startingMonth=3>)
-     3.54±0.01ms          414±3μs     0.12  indexing.CategoricalIndexIndexing.time_get_loc_scalar('non_monotonic')
-     3.59±0.02ms          417±5μs     0.12  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.Int64Engine'>, <class 'numpy.float64'>, 'non_monotonic')
-     3.59±0.01ms          415±8μs     0.12  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.Int64Engine'>, <class 'numpy.float64'>, 'monotonic_decr')
-     5.11±0.04ms          588±4μs     0.12  timeseries.DatetimeIndex.time_to_time('dst')
-     6.05±0.06ms          678±1μs     0.11  offset.OffsetSeriesArithmetic.time_add_offset(<QuarterEnd: startingMonth=3>)
-        3.51±0ms          392±1μs     0.11  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.UInt64Engine'>, <class 'numpy.int64'>, 'monotonic_decr')
-         145±1ms      15.9±0.03ms     0.11  timeseries.DatetimeIndex.time_to_time('repeated')
-       145±0.7ms      15.9±0.03ms     0.11  timeseries.DatetimeIndex.time_to_time('tz_naive')
-     3.52±0.04ms          385±5μs     0.11  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.UInt64Engine'>, <class 'numpy.int64'>, 'non_monotonic')
-     3.57±0.01ms         390±10μs     0.11  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.Int64Engine'>, <class 'numpy.uint64'>, 'non_monotonic')
-     3.56±0.01ms          388±3μs     0.11  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.Int64Engine'>, <class 'numpy.uint64'>, 'monotonic_decr')
-     6.30±0.08ms          682±3μs     0.11  offset.OffsetSeriesArithmetic.time_add_offset(<YearEnd: month=12>)
-      5.07±0.1ms          515±2μs     0.10  timeseries.DatetimeIndex.time_to_date('dst')
-         142±1ms      13.9±0.02ms     0.10  timeseries.DatetimeIndex.time_to_date('repeated')
-         143±2ms      13.9±0.05ms     0.10  timeseries.DatetimeIndex.time_to_date('tz_naive')
-      91.7±0.4ms      8.06±0.05ms     0.09  inference.DateInferOps.time_timedelta_plus_datetime
-     6.15±0.01ms         510±10μs     0.08  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.Float64Engine'>, <class 'numpy.uint64'>, 'monotonic_decr')
-      67.0±0.8ms      5.01±0.09ms     0.07  sparse.Arithmetic.time_divide(0.1, nan)
-     3.36±0.01ms          245±3μs     0.07  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.UInt64Engine'>, <class 'numpy.uint64'>, 'monotonic_decr')
-     3.37±0.01ms          244±3μs     0.07  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.UInt64Engine'>, <class 'numpy.uint64'>, 'non_monotonic')
-     3.42±0.01ms          248±8μs     0.07  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.Int64Engine'>, <class 'numpy.int64'>, 'monotonic_decr')
-     7.20±0.08ms          516±7μs     0.07  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.Float64Engine'>, <class 'numpy.uint64'>, 'non_monotonic')
-        66.7±1ms      4.73±0.02ms     0.07  sparse.Arithmetic.time_divide(0.01, nan)
-        3.42±0ms          240±2μs     0.07  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.Int64Engine'>, <class 'numpy.int64'>, 'non_monotonic')
-     6.07±0.02ms          415±5μs     0.07  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.Float64Engine'>, <class 'numpy.int64'>, 'monotonic_decr')
-      7.17±0.2ms         417±10μs     0.06  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.Float64Engine'>, <class 'numpy.int64'>, 'non_monotonic')
-      74.1±0.5ms      3.71±0.09ms     0.05  rolling.VariableWindowMethods.time_rolling('Series', '1h', 'float', 'max')
-      74.2±0.2ms      3.65±0.05ms     0.05  rolling.VariableWindowMethods.time_rolling('Series', '1h', 'float', 'min')
-      65.4±0.9ms      3.18±0.07ms     0.05  sparse.Arithmetic.time_add(0.01, nan)
-      65.6±0.7ms      3.17±0.03ms     0.05  sparse.Arithmetic.time_add(0.1, nan)
-     74.3±0.09ms      3.55±0.02ms     0.05  rolling.VariableWindowMethods.time_rolling('Series', '1h', 'int', 'min')
-      74.6±0.4ms      3.54±0.03ms     0.05  rolling.VariableWindowMethods.time_rolling('Series', '1h', 'int', 'max')
-      2.88±0.02s          136±2ms     0.05  plotting.Plotting.time_frame_plot
-      73.8±0.5ms      3.49±0.05ms     0.05  rolling.VariableWindowMethods.time_rolling('DataFrame', '1h', 'float', 'max')
-      2.82±0.01s         131±10ms     0.05  plotting.Plotting.time_series_plot
-      73.9±0.4ms      3.38±0.03ms     0.05  rolling.VariableWindowMethods.time_rolling('DataFrame', '1h', 'float', 'min')
-      73.6±0.1ms      3.33±0.01ms     0.05  rolling.VariableWindowMethods.time_rolling('DataFrame', '1h', 'int', 'max')
-      74.0±0.4ms      3.33±0.05ms     0.04  rolling.VariableWindowMethods.time_rolling('DataFrame', '1h', 'int', 'min')
-     5.50±0.05ms          243±2μs     0.04  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<QuarterEnd: startingMonth=3>)
-     8.19±0.07μs          352±3ns     0.04  timestamp.TimestampProperties.time_is_month_end(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, None)
-        5.89±0ms          252±3μs     0.04  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.Float64Engine'>, <class 'numpy.float64'>, 'monotonic_decr')
-     8.11±0.07μs          346±3ns     0.04  timestamp.TimestampProperties.time_is_month_end(None, None)
-     5.80±0.04ms          243±3μs     0.04  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<YearEnd: month=12>)
-     7.98±0.09μs          298±1ns     0.04  timestamp.TimestampProperties.time_week(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
-      8.06±0.1μs        301±0.9ns     0.04  timestamp.TimestampProperties.time_week(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, None)
-     7.99±0.08μs          298±1ns     0.04  timestamp.TimestampProperties.time_week(None, None)
-      6.77±0.3ms          250±3μs     0.04  indexing_engines.NumericEngineIndexing.time_get_loc(<class 'pandas._libs.index.Float64Engine'>, <class 'numpy.float64'>, 'non_monotonic')
-     7.88±0.06μs          290±2ns     0.04  timestamp.TimestampProperties.time_dayofyear(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
-     8.10±0.04μs        297±0.8ns     0.04  timestamp.TimestampProperties.time_week(None, 'B')
-     8.03±0.08μs          293±1ns     0.04  timestamp.TimestampProperties.time_dayofyear(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, None)
-     7.97±0.03μs        290±0.7ns     0.04  timestamp.TimestampProperties.time_dayofyear(None, 'B')
-     7.89±0.04μs          287±1ns     0.04  timestamp.TimestampProperties.time_dayofyear(None, None)
-     8.06±0.04μs          275±7ns     0.03  timestamp.TimestampProperties.time_days_in_month(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, None)
-     8.05±0.07μs          274±5ns     0.03  timestamp.TimestampProperties.time_days_in_month(None, 'B')
-     8.23±0.09μs         276±10ns     0.03  timestamp.TimestampProperties.time_is_year_end(None, None)
-     8.14±0.07μs        269±0.7ns     0.03  timestamp.TimestampProperties.time_days_in_month(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
-     8.24±0.09μs          272±4ns     0.03  timestamp.TimestampProperties.time_days_in_month(None, None)
-     8.04±0.02μs          260±6ns     0.03  timestamp.TimestampProperties.time_quarter(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, 'B')
-     7.92±0.03μs          255±2ns     0.03  timestamp.TimestampProperties.time_quarter(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, None)
-     8.18±0.06μs          260±1ns     0.03  timestamp.TimestampProperties.time_is_year_start(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, None)
-     8.22±0.05μs          261±5ns     0.03  timestamp.TimestampProperties.time_is_quarter_end(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, None)
-      8.29±0.2μs        262±0.7ns     0.03  timestamp.TimestampProperties.time_is_year_end(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, None)
-     8.13±0.06μs        257±0.6ns     0.03  timestamp.TimestampProperties.time_quarter(None, 'B')
-     8.32±0.04μs          261±5ns     0.03  timestamp.TimestampProperties.time_is_quarter_end(None, None)
-      7.97±0.1μs          250±2ns     0.03  timestamp.TimestampProperties.time_quarter(None, None)
-      8.39±0.1μs          263±4ns     0.03  timestamp.TimestampProperties.time_is_leap_year(None, None)
-      8.18±0.2μs          256±6ns     0.03  timestamp.TimestampProperties.time_is_year_start(None, None)
-     8.28±0.05μs        258±0.2ns     0.03  timestamp.TimestampProperties.time_is_leap_year(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, None)
-     8.15±0.07μs          253±3ns     0.03  timestamp.TimestampProperties.time_is_quarter_start(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, None)
-     8.14±0.04μs          253±1ns     0.03  timestamp.TimestampProperties.time_is_quarter_start(None, None)
-     8.21±0.07μs          254±3ns     0.03  timestamp.TimestampProperties.time_is_month_start(None, None)
-     8.40±0.09μs          257±1ns     0.03  timestamp.TimestampProperties.time_is_month_start(<DstTzInfo 'Europe/Amsterdam' LMT+0:20:00 STD>, None)
-      25.7±0.1ms          694±5μs     0.03  offset.OffsetSeriesArithmetic.time_add_offset(<BusinessQuarterEnd: startingMonth=3>)
-      31.2±0.2ms          694±2μs     0.02  offset.OffsetSeriesArithmetic.time_add_offset(<BusinessMonthEnd>)
-      31.4±0.1ms          685±2μs     0.02  offset.OffsetSeriesArithmetic.time_add_offset(<BusinessYearBegin: month=1>)
-      31.5±0.2ms          688±7μs     0.02  offset.OffsetSeriesArithmetic.time_add_offset(<BusinessQuarterBegin: startingMonth=3>)
-      1.82±0.01s       37.6±0.2ms     0.02  stat_ops.FrameOps.time_op('median', 'float', 1, False)
-      1.83±0.01s       37.4±0.2ms     0.02  stat_ops.FrameOps.time_op('median', 'float', 1, True)
-      1.82±0.01s      36.1±0.08ms     0.02  stat_ops.FrameOps.time_op('median', 'int', 1, False)
-     34.6±0.06ms          686±5μs     0.02  offset.OffsetSeriesArithmetic.time_add_offset(<BusinessMonthBegin>)
-      1.83±0.07s       36.1±0.2ms     0.02  stat_ops.FrameOps.time_op('median', 'int', 1, True)
-        55.7±1ms          702±5μs     0.01  offset.OffsetSeriesArithmetic.time_add_offset(<BusinessYearEnd: month=12>)
-      45.9±0.7ms          529±1μs     0.01  offset.ApplyIndex.time_apply_index(<YearEnd: month=12>)
-      43.4±0.6ms          490±1μs     0.01  offset.ApplyIndex.time_apply_index(<QuarterEnd: startingMonth=3>)
-      70.4±0.3ms          749±2μs     0.01  groupby.GroupByMethods.time_dtype_as_field('float', 'rank', 'transformation')
-      70.5±0.5ms          750±3μs     0.01  groupby.GroupByMethods.time_dtype_as_field('float', 'rank', 'direct')
-        73.1±1ms         768±10μs     0.01  groupby.GroupByMethods.time_dtype_as_field('int', 'rank', 'direct')
-      72.6±0.9ms          760±8μs     0.01  groupby.GroupByMethods.time_dtype_as_field('int', 'rank', 'transformation')
-      19.6±0.2μs          201±1ns     0.01  timedelta.DatetimeAccessor.time_dt_accessor
-     25.8±0.05ms        258±0.9μs     0.01  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<BusinessQuarterEnd: startingMonth=3>)
-      55.3±0.4ms          541±8μs     0.01  indexing.NumericSeriesIndexing.time_getitem_list_like(<class 'pandas.core.indexes.numeric.Float64Index'>, 'nonunique_monotonic_inc')
-         183±7ms      1.77±0.03ms     0.01  index_object.Indexing.time_get_loc_non_unique('Float')
-      1.66±0.01s       14.9±0.1ms     0.01  rolling.Pairwise.time_pairwise(1000, 'corr', True)
-      1.67±0.01s      14.5±0.08ms     0.01  rolling.Pairwise.time_pairwise(10, 'corr', True)
-      1.68±0.01s      14.5±0.04ms     0.01  rolling.Pairwise.time_pairwise(None, 'corr', True)
-         169±2ms         1.45±0ms     0.01  index_object.Indexing.time_get_loc_non_unique_sorted('Float')
-      30.6±0.2ms          258±2μs     0.01  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<BusinessMonthEnd>)
-      1.65±0.02s       13.4±0.1ms     0.01  rolling.Pairwise.time_pairwise(1000, 'cov', True)
-      31.7±0.5ms          252±1μs     0.01  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<BusinessYearBegin: month=1>)
-         1.66±0s       13.1±0.1ms     0.01  rolling.Pairwise.time_pairwise(10, 'cov', True)
-      31.5±0.5ms          248±3μs     0.01  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<BusinessQuarterBegin: startingMonth=3>)
-      1.67±0.03s      12.9±0.06ms     0.01  rolling.Pairwise.time_pairwise(None, 'cov', True)
-      54.8±0.5ms          419±3μs     0.01  indexing.NumericSeriesIndexing.time_ix_list_like(<class 'pandas.core.indexes.numeric.Float64Index'>, 'nonunique_monotonic_inc')
-         268±4ms      2.03±0.02ms     0.01  multiindex_object.Integer.time_get_indexer
-      57.7±0.9ms          430±5μs     0.01  indexing.NumericSeriesIndexing.time_ix_list_like(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'nonunique_monotonic_inc')
-      31.8±0.6μs          234±4ns     0.01  categoricals.IsMonotonic.time_categorical_index_is_monotonic_decreasing
-      58.4±0.3ms          425±5μs     0.01  indexing.NumericSeriesIndexing.time_ix_list_like(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
-       106±0.7ms          766±3μs     0.01  groupby.GroupByMethods.time_dtype_as_group('int', 'rank', 'direct')
-      31.8±0.3μs          230±2ns     0.01  categoricals.IsMonotonic.time_categorical_index_is_monotonic_increasing
-       106±0.6ms          759±2μs     0.01  groupby.GroupByMethods.time_dtype_as_group('int', 'rank', 'transformation')
-      15.9±0.1μs          114±1ns     0.01  timeseries.DatetimeAccessor.time_dt_accessor
-      34.3±0.3ms          241±2μs     0.01  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<BusinessMonthBegin>)
-      54.6±0.3ms          377±3μs     0.01  indexing.NumericSeriesIndexing.time_loc_list_like(<class 'pandas.core.indexes.numeric.Float64Index'>, 'nonunique_monotonic_inc')
-      57.5±0.3ms        388±100μs     0.01  groupby.GroupByMethods.time_dtype_as_field('int', 'ffill', 'direct')
-      97.3±0.3ms          657±4μs     0.01  groupby.GroupByMethods.time_dtype_as_field('datetime', 'rank', 'transformation')
-      98.2±0.9ms          661±4μs     0.01  groupby.GroupByMethods.time_dtype_as_field('datetime', 'rank', 'direct')
-     1.87±0.01ms      11.1±0.07μs     0.01  categoricals.Contains.time_categorical_contains
-      57.2±0.2ms         328±90μs     0.01  groupby.GroupByMethods.time_dtype_as_field('int', 'bfill', 'direct')
-       257±0.7ms         1.47±0ms     0.01  index_object.Indexing.time_get_loc_non_unique_sorted('Int')
-         259±8ms      1.47±0.03ms     0.01  index_object.Indexing.time_get_loc_non_unique('Int')
-      57.5±0.6ms         323±90μs     0.01  groupby.GroupByMethods.time_dtype_as_field('int', 'ffill', 'transformation')
-     1.91±0.01ms      10.2±0.08μs     0.01  offset.OnOffset.time_on_offset(<BusinessYearEnd: month=12>)
-     1.28±0.01ms      6.08±0.06μs     0.00  offset.OnOffset.time_on_offset(<BusinessQuarterBegin: startingMonth=3>)
-         130±2ms          606±9μs     0.00  indexing.NumericSeriesIndexing.time_getitem_list_like(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'nonunique_monotonic_inc')
-       164±0.3ms          763±5μs     0.00  groupby.GroupByMethods.time_dtype_as_group('float', 'rank', 'transformation')
-        55.5±1ms          257±2μs     0.00  offset.OffsetDatetimeIndexArithmetic.time_add_offset(<BusinessYearEnd: month=12>)
-     1.23±0.01ms       5.66±0.1μs     0.00  offset.OnOffset.time_on_offset(<QuarterBegin: startingMonth=3>)
-       165±0.8ms          761±2μs     0.00  groupby.GroupByMethods.time_dtype_as_group('float', 'rank', 'direct')
-         167±2ms          767±2μs     0.00  groupby.GroupByMethods.time_dtype_as_group('datetime', 'rank', 'transformation')
-         169±1ms         773±10μs     0.00  groupby.GroupByMethods.time_dtype_as_group('datetime', 'rank', 'direct')
-     1.51±0.03ms       6.76±0.1μs     0.00  offset.OnOffset.time_on_offset(<BusinessMonthEnd>)
-       132±0.8ms          587±6μs     0.00  indexing.NumericSeriesIndexing.time_getitem_list_like(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
-      57.6±0.4ms          252±8μs     0.00  groupby.GroupByMethods.time_dtype_as_field('int', 'bfill', 'transformation')
-     1.55±0.01ms      6.00±0.02μs     0.00  offset.OnOffset.time_on_offset(<BusinessQuarterEnd: startingMonth=3>)
-       129±0.3ms          401±6μs     0.00  indexing.NumericSeriesIndexing.time_loc_list_like(<class 'pandas.core.indexes.numeric.UInt64Index'>, 'nonunique_monotonic_inc')
-      84.1±0.2ms          253±2μs     0.00  groupby.GroupByMethods.time_dtype_as_group('int', 'bfill', 'direct')
-      84.4±0.6ms          252±3μs     0.00  groupby.GroupByMethods.time_dtype_as_group('int', 'bfill', 'transformation')
-      83.9±0.5ms          250±3μs     0.00  groupby.GroupByMethods.time_dtype_as_group('int', 'ffill', 'transformation')
-      85.5±0.7ms          252±1μs     0.00  groupby.GroupByMethods.time_dtype_as_group('int', 'ffill', 'direct')
-         131±3ms          384±2μs     0.00  indexing.NumericSeriesIndexing.time_loc_list_like(<class 'pandas.core.indexes.numeric.Int64Index'>, 'nonunique_monotonic_inc')
-        1.46±0ms      3.77±0.02μs     0.00  offset.OnOffset.time_on_offset(<BusinessYearBegin: month=1>)
-      58.7±0.8ms          143±4μs     0.00  groupby.GroupByMethods.time_dtype_as_field('object', 'any', 'transformation')
-      58.0±0.4ms          135±3μs     0.00  groupby.GroupByMethods.time_dtype_as_field('object', 'all', 'direct')
-         250±2ms          579±6μs     0.00  groupby.GroupByMethods.time_dtype_as_field('int', 'pct_change', 'direct')
-      58.3±0.4ms        134±0.7μs     0.00  groupby.GroupByMethods.time_dtype_as_field('object', 'any', 'direct')
-       248±0.8ms          572±3μs     0.00  groupby.GroupByMethods.time_dtype_as_field('int', 'pct_change', 'transformation')
-      58.2±0.1ms        132±0.7μs     0.00  groupby.GroupByMethods.time_dtype_as_field('object', 'all', 'transformation')
-      1.61±0.01s      3.52±0.03ms     0.00  rolling.VariableWindowMethods.time_rolling('Series', '1d', 'int', 'max')
-         1.61±0s      3.51±0.07ms     0.00  rolling.VariableWindowMethods.time_rolling('Series', '1d', 'int', 'min')
-         1.62±0s      3.47±0.03ms     0.00  rolling.VariableWindowMethods.time_rolling('Series', '1d', 'float', 'max')
-         1.62±0s      3.45±0.04ms     0.00  rolling.VariableWindowMethods.time_rolling('Series', '1d', 'float', 'min')
-      98.5±0.5ms          207±2μs     0.00  groupby.GroupByMethods.time_dtype_as_field('float', 'bfill', 'transformation')
-        99.5±1ms          208±2μs     0.00  groupby.GroupByMethods.time_dtype_as_field('float', 'bfill', 'direct')
-      98.9±0.6ms          206±1μs     0.00  groupby.GroupByMethods.time_dtype_as_field('float', 'ffill', 'direct')
-      1.61±0.01s      3.33±0.04ms     0.00  rolling.VariableWindowMethods.time_rolling('DataFrame', '1d', 'int', 'min')
-      99.8±0.6ms          205±2μs     0.00  groupby.GroupByMethods.time_dtype_as_field('float', 'ffill', 'transformation')
-      1.61±0.01s      3.30±0.03ms     0.00  rolling.VariableWindowMethods.time_rolling('DataFrame', '1d', 'int', 'max')
-      1.61±0.01s      3.25±0.04ms     0.00  rolling.VariableWindowMethods.time_rolling('DataFrame', '1d', 'float', 'max')
-         1.61±0s      3.23±0.03ms     0.00  rolling.VariableWindowMethods.time_rolling('DataFrame', '1d', 'float', 'min')
-         281±2ms          521±3μs     0.00  groupby.GroupByMethods.time_dtype_as_field('float', 'pct_change', 'direct')
-       158±0.6ms          291±1μs     0.00  groupby.GroupByMethods.time_dtype_as_field('object', 'ffill', 'direct')
-       159±0.9ms          294±5μs     0.00  groupby.GroupByMethods.time_dtype_as_field('object', 'bfill', 'transformation')
-       160±0.8ms          292±2μs     0.00  groupby.GroupByMethods.time_dtype_as_field('object', 'bfill', 'direct')
-         280±2ms          510±3μs     0.00  groupby.GroupByMethods.time_dtype_as_field('float', 'pct_change', 'transformation')
-         610±4ms         1.10±0ms     0.00  timedelta.DatetimeAccessor.time_timedelta_days
-         161±2ms          290±3μs     0.00  groupby.GroupByMethods.time_dtype_as_field('object', 'ffill', 'transformation')
-       132±0.7ms          235±2μs     0.00  groupby.GroupByMethods.time_dtype_as_group('float', 'ffill', 'transformation')
-       133±0.7ms          237±2μs     0.00  groupby.GroupByMethods.time_dtype_as_group('float', 'bfill', 'transformation')
-         626±2ms      1.11±0.01ms     0.00  timedelta.DatetimeAccessor.time_timedelta_nanoseconds
-       132±0.5ms          234±3μs     0.00  groupby.GroupByMethods.time_dtype_as_group('float', 'ffill', 'direct')
-       133±0.8ms        236±0.9μs     0.00  groupby.GroupByMethods.time_dtype_as_group('float', 'bfill', 'direct')
-         639±6ms      1.12±0.03ms     0.00  timedelta.DatetimeAccessor.time_timedelta_seconds
-        643±20ms      1.10±0.01ms     0.00  timedelta.DatetimeAccessor.time_timedelta_microseconds
-         363±3ms          578±5μs     0.00  groupby.GroupByMethods.time_dtype_as_group('int', 'pct_change', 'direct')
-         366±2ms          570±2μs     0.00  groupby.GroupByMethods.time_dtype_as_group('int', 'pct_change', 'transformation')
-      48.4±0.2ms         67.9±1μs     0.00  groupby.GroupByMethods.time_dtype_as_field('int', 'all', 'transformation')
-       151±0.7ms          211±1μs     0.00  groupby.GroupByMethods.time_dtype_as_field('datetime', 'bfill', 'direct')
-         152±2ms          212±2μs     0.00  groupby.GroupByMethods.time_dtype_as_field('datetime', 'ffill', 'direct')
-       152±0.9ms        211±0.9μs     0.00  groupby.GroupByMethods.time_dtype_as_field('datetime', 'bfill', 'transformation')
-      49.1±0.3ms       68.0±0.4μs     0.00  groupby.GroupByMethods.time_dtype_as_field('int', 'any', 'transformation')
-         151±2ms          210±1μs     0.00  groupby.GroupByMethods.time_dtype_as_field('datetime', 'ffill', 'transformation')
-      48.0±0.2ms       66.4±0.3μs     0.00  groupby.GroupByMethods.time_dtype_as_field('float', 'any', 'transformation')
-      48.0±0.4ms       66.1±0.2μs     0.00  groupby.GroupByMethods.time_dtype_as_field('float', 'all', 'direct')
-      49.1±0.4ms       67.2±0.6μs     0.00  groupby.GroupByMethods.time_dtype_as_field('int', 'all', 'direct')
-      49.2±0.5ms       67.2±0.5μs     0.00  groupby.GroupByMethods.time_dtype_as_field('int', 'any', 'direct')
-      48.9±0.6ms       66.8±0.7μs     0.00  groupby.GroupByMethods.time_dtype_as_field('float', 'any', 'direct')
-      48.1±0.2ms         65.7±1μs     0.00  groupby.GroupByMethods.time_dtype_as_field('float', 'all', 'transformation')
-      52.9±0.2ms       65.5±0.4μs     0.00  groupby.GroupByMethods.time_dtype_as_field('datetime', 'any', 'transformation')
-      53.3±0.1ms       65.0±0.3μs     0.00  groupby.GroupByMethods.time_dtype_as_field('datetime', 'any', 'direct')
-      53.5±0.3ms       64.6±0.7μs     0.00  groupby.GroupByMethods.time_dtype_as_field('datetime', 'all', 'transformation')
-      53.5±0.2ms       64.4±0.4μs     0.00  groupby.GroupByMethods.time_dtype_as_field('datetime', 'all', 'direct')
-       137±0.2ms          161±1μs     0.00  groupby.GroupByMethods.time_dtype_as_group('datetime', 'bfill', 'direct')
-       135±0.6ms          158±1μs     0.00  groupby.GroupByMethods.time_dtype_as_group('datetime', 'ffill', 'direct')
-         135±1ms        157±0.6μs     0.00  groupby.GroupByMethods.time_dtype_as_group('datetime', 'ffill', 'transformation')
-         137±2ms          157±1μs     0.00  groupby.GroupByMethods.time_dtype_as_group('datetime', 'bfill', 'transformation')
-     1.87±0.01ms      2.08±0.07μs     0.00  categoricals.Contains.time_categorical_index_contains
-         565±1ms          562±5μs     0.00  groupby.GroupByMethods.time_dtype_as_group('float', 'pct_change', 'direct')
-         565±2ms          555±3μs     0.00  groupby.GroupByMethods.time_dtype_as_group('float', 'pct_change', 'transformation')
-      70.9±0.2ms         68.7±1μs     0.00  groupby.GroupByMethods.time_dtype_as_group('int', 'any', 'transformation')
-      71.6±0.6ms       68.5±0.5μs     0.00  groupby.GroupByMethods.time_dtype_as_group('int', 'any', 'direct')
-      72.1±0.3ms       68.0±0.5μs     0.00  groupby.GroupByMethods.time_dtype_as_group('int', 'all', 'direct')
-      72.7±0.6ms       68.1±0.6μs     0.00  groupby.GroupByMethods.time_dtype_as_group('int', 'all', 'transformation')
-       111±0.8ms       69.5±0.4μs     0.00  groupby.GroupByMethods.time_dtype_as_group('float', 'all', 'direct')
-         112±1ms       68.8±0.2μs     0.00  groupby.GroupByMethods.time_dtype_as_group('float', 'any', 'transformation')
-       112±0.8ms       69.0±0.9μs     0.00  groupby.GroupByMethods.time_dtype_as_group('float', 'any', 'direct')
-         112±1ms       68.2±0.3μs     0.00  groupby.GroupByMethods.time_dtype_as_group('float', 'all', 'transformation')
-       114±0.4ms       68.9±0.2μs     0.00  groupby.GroupByMethods.time_dtype_as_group('datetime', 'all', 'transformation')
-       115±0.4ms       69.7±0.3μs     0.00  groupby.GroupByMethods.time_dtype_as_group('datetime', 'any', 'direct')
-       115±0.9ms       69.4±0.7μs     0.00  groupby.GroupByMethods.time_dtype_as_group('datetime', 'any', 'transformation')
-         116±1ms       69.4±0.4μs     0.00  groupby.GroupByMethods.time_dtype_as_group('datetime', 'all', 'direct')
-      1.49±0.01s          471±5μs     0.00  series_methods.IsInForObjects.time_isin_nans
-       290±0.6ms       19.7±0.2μs     0.00  multiindex_object.GetLoc.time_large_get_loc
-        937±30ms      2.28±0.01μs     0.00  series_methods.SeriesGetattr.time_series_datetimeindex_repr

Metadata

Metadata

Assignees

No one assigned

    Labels

    PerformanceMemory or execution speed performance

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions