CI is failing due to benchmark failing at arithmetic.ApplyIndex.time_apply_index (0.65%)

[hasnain2808]

ASV benchmarks are failing on the benchmarks for arithmetic.ApplyIndex.time_apply_index
I believe the benchmarks are failing after #34499 was merged.

The latest benchmark failed link

Related errors

##[error][  0.65%] ··· arithmetic.ApplyIndex.time_apply_index                 3/10 failed
[  0.65%] ··· =================================== ==========
                             offset                         
              ----------------------------------- ----------
                      <YearEnd: month=12>          1.55±0ms 
                      <YearBegin: month=1>         1.37±0ms 
                 <QuarterEnd: startingMonth=3>     1.71±0ms 
                <QuarterBegin: startingMonth=3>    1.59±0ms 
                           <MonthEnd>              2.25±0ms 
                          <MonthBegin>             1.34±0ms 
                 <DateOffset: days=2, months=2>    3.16±0ms 
                         <BusinessDay>              failed  
                <SemiMonthEnd: day_of_month=15>     failed  
               <SemiMonthBegin: day_of_month=15>    failed  
              =================================== ==========

[  0.65%] ···· For parameters: <BusinessDay>
               Traceback (most recent call last):
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 1184, in main_run_server
                   main_run(run_args)
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 1058, in main_run
                   result = benchmark.do_run()
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 537, in do_run
                   return self.run(*self._current_params)
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 627, in run
                   samples, number = self.benchmark_timing(timer, min_repeat, max_repeat,
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 694, in benchmark_timing
                   timing = timer.timeit(number)
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/timeit.py", line 177, in timeit
                   timing = self.inner(it, self.timer)
                 File "<timeit-src>", line 6, in inner
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 599, in <lambda>
                   func = lambda: self.func(*param)
                 File "/home/runner/work/pandas/pandas/asv_bench/benchmarks/arithmetic.py", line 469, in time_apply_index
                   offset.apply_index(self.rng)
                 File "pandas/_libs/tslibs/offsets.pyx", line 87, in pandas._libs.tslibs.offsets.apply_index_wraps.wrapper
                 File "pandas/_libs/tslibs/offsets.pyx", line 1397, in pandas._libs.tslibs.offsets.BusinessDay.apply_index
               AttributeError: 'PeriodIndex' object has no attribute '_addsub_int_array'
               
               For parameters: <SemiMonthEnd: day_of_month=15>
               Traceback (most recent call last):
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 1184, in main_run_server
                   main_run(run_args)
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 1058, in main_run
                   result = benchmark.do_run()
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 537, in do_run
                   return self.run(*self._current_params)
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 627, in run
                   samples, number = self.benchmark_timing(timer, min_repeat, max_repeat,
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 694, in benchmark_timing
                   timing = timer.timeit(number)
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/timeit.py", line 177, in timeit
                   timing = self.inner(it, self.timer)
                 File "<timeit-src>", line 6, in inner
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 599, in <lambda>
                   func = lambda: self.func(*param)
                 File "/home/runner/work/pandas/pandas/asv_bench/benchmarks/arithmetic.py", line 469, in time_apply_index
                   offset.apply_index(self.rng)
                 File "pandas/_libs/tslibs/offsets.pyx", line 87, in pandas._libs.tslibs.offsets.apply_index_wraps.wrapper
                 File "pandas/_libs/tslibs/offsets.pyx", line 2319, in pandas._libs.tslibs.offsets.SemiMonthOffset.apply_index
               AttributeError: 'PeriodIndex' object has no attribute '_addsub_int_array'
               
               For parameters: <SemiMonthBegin: day_of_month=15>
               Traceback (most recent call last):
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 1184, in main_run_server
                   main_run(run_args)
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 1058, in main_run
                   result = benchmark.do_run()
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 537, in do_run
                   return self.run(*self._current_params)
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 627, in run
                   samples, number = self.benchmark_timing(timer, min_repeat, max_repeat,
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 694, in benchmark_timing
                   timing = timer.timeit(number)
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/timeit.py", line 177, in timeit
                   timing = self.inner(it, self.timer)
                 File "<timeit-src>", line 6, in inner
                 File "/home/runner/miniconda3/envs/pandas-dev/lib/python3.8/site-packages/asv/benchmark.py", line 599, in <lambda>
                   func = lambda: self.func(*param)
                 File "/home/runner/work/pandas/pandas/asv_bench/benchmarks/arithmetic.py", line 469, in time_apply_index
                   offset.apply_index(self.rng)
                 File "pandas/_libs/tslibs/offsets.pyx", line 87, in pandas._libs.tslibs.offsets.apply_index_wraps.wrapper
                 File "pandas/_libs/tslibs/offsets.pyx", line 2319, in pandas._libs.tslibs.offsets.SemiMonthOffset.apply_index
               AttributeError: 'PeriodIndex' object has no attribute '_addsub_int_array'

I am not sure what was the right way to report this thus created this issue. Do tell me if this is an incorrect place, incorrect way or incorrect language

[jorisvandenbossche]

cc @jbrockmendel

[jorisvandenbossche]

@hasnain2808 do you also see this locally? (you can run only this specific benchmark with asv dev -b arithmetic.ApplyIndex.time_apply_index)

[hasnain2808]

yes @jorisvandenbossche

Attaching the output for reference

(pandas-dev) xxxxxxxxx:~/xxxxxx/pandas-hasnain2808/asv_bench$ asv dev -b arithmetic.ApplyIndex.time_apply_index
· Discovering benchmarks
· Running 1 total benchmarks (1 commits * 1 environments * 1 benchmarks)
[  0.00%] ·· Benchmarking existing-py_home_moha_venv_pandas-dev_bin_python
[ 50.00%] ··· arithmetic.ApplyIndex.time_apply_index                                                                                       3/10 failed
[ 50.00%] ··· =================================== ==========
                             offset                         
              ----------------------------------- ----------
                      <YearEnd: month=12>          1.44±0ms 
                      <YearBegin: month=1>         1.51±0ms 
                 <QuarterEnd: startingMonth=3>     3.07±0ms 
                <QuarterBegin: startingMonth=3>    3.01±0ms 
                           <MonthEnd>              2.67±0ms 
                          <MonthBegin>             2.58±0ms 
                 <DateOffset: days=2, months=2>    4.74±0ms 
                         <BusinessDay>              failed  
                <SemiMonthEnd: day_of_month=15>     failed  
               <SemiMonthBegin: day_of_month=15>    failed  
              =================================== ==========

[ 50.00%] ···· For parameters: <BusinessDay>
               Traceback (most recent call last):
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 1184, in main_run_server
                   main_run(run_args)
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 1058, in main_run
                   result = benchmark.do_run()
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 537, in do_run
                   return self.run(*self._current_params)
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 631, in run
                   min_run_count=self.min_run_count)
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 694, in benchmark_timing
                   timing = timer.timeit(number)
                 File "/usr/lib/python3.6/timeit.py", line 178, in timeit
                   timing = self.inner(it, self.timer)
                 File "<timeit-src>", line 6, in inner
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 599, in <lambda>
                   func = lambda: self.func(*param)
                 File "/home/moha/projects/pandas-hasnain2808/asv_bench/benchmarks/arithmetic.py", line 469, in time_apply_index
                   offset.apply_index(self.rng)
                 File "pandas/_libs/tslibs/offsets.pyx", line 87, in pandas._libs.tslibs.offsets.apply_index_wraps.wrapper
                 File "pandas/_libs/tslibs/offsets.pyx", line 1397, in pandas._libs.tslibs.offsets.BusinessDay.apply_index
               AttributeError: 'PeriodIndex' object has no attribute '_addsub_int_array'
               
               For parameters: <SemiMonthEnd: day_of_month=15>
               Traceback (most recent call last):
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 1184, in main_run_server
                   main_run(run_args)
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 1058, in main_run
                   result = benchmark.do_run()
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 537, in do_run
                   return self.run(*self._current_params)
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 631, in run
                   min_run_count=self.min_run_count)
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 694, in benchmark_timing
                   timing = timer.timeit(number)
                 File "/usr/lib/python3.6/timeit.py", line 178, in timeit
                   timing = self.inner(it, self.timer)
                 File "<timeit-src>", line 6, in inner
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 599, in <lambda>
                   func = lambda: self.func(*param)
                 File "/home/moha/projects/pandas-hasnain2808/asv_bench/benchmarks/arithmetic.py", line 469, in time_apply_index
                   offset.apply_index(self.rng)
                 File "pandas/_libs/tslibs/offsets.pyx", line 87, in pandas._libs.tslibs.offsets.apply_index_wraps.wrapper
                 File "pandas/_libs/tslibs/offsets.pyx", line 2319, in pandas._libs.tslibs.offsets.SemiMonthOffset.apply_index
               AttributeError: 'PeriodIndex' object has no attribute '_addsub_int_array'
               
               For parameters: <SemiMonthBegin: day_of_month=15>
               Traceback (most recent call last):
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 1184, in main_run_server
                   main_run(run_args)
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 1058, in main_run
                   result = benchmark.do_run()
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 537, in do_run
                   return self.run(*self._current_params)
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 631, in run
                   min_run_count=self.min_run_count)
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 694, in benchmark_timing
                   timing = timer.timeit(number)
                 File "/usr/lib/python3.6/timeit.py", line 178, in timeit
                   timing = self.inner(it, self.timer)
                 File "<timeit-src>", line 6, in inner
                 File "/home/moha/venv/pandas-dev/lib/python3.6/site-packages/asv/benchmark.py", line 599, in <lambda>
                   func = lambda: self.func(*param)
                 File "/home/moha/projects/pandas-hasnain2808/asv_bench/benchmarks/arithmetic.py", line 469, in time_apply_index
                   offset.apply_index(self.rng)
                 File "pandas/_libs/tslibs/offsets.pyx", line 87, in pandas._libs.tslibs.offsets.apply_index_wraps.wrapper
                 File "pandas/_libs/tslibs/offsets.pyx", line 2319, in pandas._libs.tslibs.offsets.SemiMonthOffset.apply_index
               AttributeError: 'PeriodIndex' object has no attribute '_addsub_int_array'

[jorisvandenbossche]

Thanks, and I just updated to latest master and see the same error as well.

[hasnain2808]

Okay.

[jbrockmendel]

It's not #34499, its #34555. The solution is to change time_apply_index to just self.rng + offset and not call apply_index directly