AlenkaF commented on PR #41870:
URL: https://github.com/apache/arrow/pull/41870#issuecomment-2137751641
Benchmarks for `RecordBatch::ToTensor` after the changeing the
implementation to use `Table::ToTensor`:
```
(pyarrow-dev) alenkafrim@alenka-mac arrow % archery --quiet benchmark diff
--benchmark-filter=BatchToTensorSimple
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Non-regressions: (7)
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
benchmark baseline
contender change %
counters
BatchToTensorSimple<Int64Type>/size:4194304/num_columns:3 8.540 GiB/sec
8.826 GiB/sec 3.351 {'family_index': 3, 'per_family_instance_index': 3,
'run_name': 'BatchToTensorSimple<Int64Type>/size:4194304/num_columns:3',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 1545}
BatchToTensorSimple<Int16Type>/size:4194304/num_columns:3 4.515 GiB/sec
4.583 GiB/sec 1.516 {'family_index': 1, 'per_family_instance_index': 3,
'run_name': 'BatchToTensorSimple<Int16Type>/size:4194304/num_columns:3',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 787}
BatchToTensorSimple<Int64Type>/size:4194304/num_columns:30 5.355 GiB/sec
5.426 GiB/sec 1.320 {'family_index': 3, 'per_family_instance_index': 4,
'run_name': 'BatchToTensorSimple<Int64Type>/size:4194304/num_columns:30',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 971}
BatchToTensorSimple<Int16Type>/size:4194304/num_columns:30 2.113 GiB/sec
2.120 GiB/sec 0.331 {'family_index': 1, 'per_family_instance_index': 4,
'run_name': 'BatchToTensorSimple<Int16Type>/size:4194304/num_columns:30',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 380}
BatchToTensorSimple<Int16Type>/size:4194304/num_columns:300 2.009 GiB/sec
1.976 GiB/sec -1.620 {'family_index': 1, 'per_family_instance_index': 5,
'run_name': 'BatchToTensorSimple<Int16Type>/size:4194304/num_columns:300',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 363}
BatchToTensorSimple<Int16Type>/size:65536/num_columns:3 5.391 GiB/sec
5.141 GiB/sec -4.645 {'family_index': 1, 'per_family_instance_index': 0,
'run_name': 'BatchToTensorSimple<Int16Type>/size:65536/num_columns:3',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 61484}
BatchToTensorSimple<Int64Type>/size:4194304/num_columns:300 7.797 GiB/sec
7.429 GiB/sec -4.716 {'family_index': 3, 'per_family_instance_index': 5,
'run_name': 'BatchToTensorSimple<Int64Type>/size:4194304/num_columns:300',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 1374}
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Regressions: (17)
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
benchmark baseline
contender change %
counters
BatchToTensorSimple<Int8Type>/size:4194304/num_columns:300 698.025 MiB/sec
642.690 MiB/sec -7.927 {'family_index': 0, 'per_family_instance_index': 5,
'run_name': 'BatchToTensorSimple<Int8Type>/size:4194304/num_columns:300',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 117}
BatchToTensorSimple<Int8Type>/size:4194304/num_columns:30 936.761 MiB/sec
849.504 MiB/sec -9.315 {'family_index': 0, 'per_family_instance_index': 4,
'run_name': 'BatchToTensorSimple<Int8Type>/size:4194304/num_columns:30',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 164}
BatchToTensorSimple<Int32Type>/size:4194304/num_columns:30 2.943 GiB/sec
2.664 GiB/sec -9.484 {'family_index': 2, 'per_family_instance_index': 4,
'run_name': 'BatchToTensorSimple<Int32Type>/size:4194304/num_columns:30',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 530}
BatchToTensorSimple<Int8Type>/size:4194304/num_columns:3 1.220 GiB/sec
1.103 GiB/sec -9.540 {'family_index': 0, 'per_family_instance_index': 3,
'run_name': 'BatchToTensorSimple<Int8Type>/size:4194304/num_columns:3',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 226}
BatchToTensorSimple<Int32Type>/size:4194304/num_columns:300 3.350 GiB/sec
3.004 GiB/sec -10.308 {'family_index': 2, 'per_family_instance_index': 5,
'run_name': 'BatchToTensorSimple<Int32Type>/size:4194304/num_columns:300',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 603}
BatchToTensorSimple<Int8Type>/size:65536/num_columns:3 1.343 GiB/sec
1.193 GiB/sec -11.189 {'family_index': 0, 'per_family_instance_index': 0,
'run_name': 'BatchToTensorSimple<Int8Type>/size:65536/num_columns:3',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 15407}
BatchToTensorSimple<Int32Type>/size:4194304/num_columns:3 6.492 GiB/sec
5.679 GiB/sec -12.518 {'family_index': 2, 'per_family_instance_index': 3,
'run_name': 'BatchToTensorSimple<Int32Type>/size:4194304/num_columns:3',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 1170}
BatchToTensorSimple<Int32Type>/size:65536/num_columns:3 8.703 GiB/sec
7.530 GiB/sec -13.478 {'family_index': 2, 'per_family_instance_index': 0,
'run_name': 'BatchToTensorSimple<Int32Type>/size:65536/num_columns:3',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 99016}
BatchToTensorSimple<Int64Type>/size:65536/num_columns:3 17.419 GiB/sec
14.934 GiB/sec -14.269 {'family_index': 3, 'per_family_instance_index': 0,
'run_name': 'BatchToTensorSimple<Int64Type>/size:65536/num_columns:3',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 198847}
BatchToTensorSimple<Int8Type>/size:65536/num_columns:30 1.246 GiB/sec
1.013 GiB/sec -18.692 {'family_index': 0, 'per_family_instance_index': 1,
'run_name': 'BatchToTensorSimple<Int8Type>/size:65536/num_columns:30',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 14331}
BatchToTensorSimple<Int16Type>/size:65536/num_columns:30 3.813 GiB/sec
3.045 GiB/sec -20.148 {'family_index': 1, 'per_family_instance_index': 1,
'run_name': 'BatchToTensorSimple<Int16Type>/size:65536/num_columns:30',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 43240}
BatchToTensorSimple<Int32Type>/size:65536/num_columns:30 5.497 GiB/sec
3.822 GiB/sec -30.460 {'family_index': 2, 'per_family_instance_index': 1,
'run_name': 'BatchToTensorSimple<Int32Type>/size:65536/num_columns:30',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 63621}
BatchToTensorSimple<Int8Type>/size:65536/num_columns:300 665.489 MiB/sec
452.284 MiB/sec -32.037 {'family_index': 0, 'per_family_instance_index': 2,
'run_name': 'BatchToTensorSimple<Int8Type>/size:65536/num_columns:300',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 7122}
BatchToTensorSimple<Int64Type>/size:65536/num_columns:30 7.306 GiB/sec
4.883 GiB/sec -33.166 {'family_index': 3, 'per_family_instance_index': 1,
'run_name': 'BatchToTensorSimple<Int64Type>/size:65536/num_columns:30',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 83661}
BatchToTensorSimple<Int16Type>/size:65536/num_columns:300 1.024 GiB/sec
646.927 MiB/sec -38.317 {'family_index': 1, 'per_family_instance_index': 2,
'run_name': 'BatchToTensorSimple<Int16Type>/size:65536/num_columns:300',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 11642}
BatchToTensorSimple<Int64Type>/size:65536/num_columns:300 1.208 GiB/sec
711.915 MiB/sec -42.439 {'family_index': 3, 'per_family_instance_index': 2,
'run_name': 'BatchToTensorSimple<Int64Type>/size:65536/num_columns:300',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 13994}
BatchToTensorSimple<Int32Type>/size:65536/num_columns:300 1.158 GiB/sec
678.147 MiB/sec -42.812 {'family_index': 2, 'per_family_instance_index': 2,
'run_name': 'BatchToTensorSimple<Int32Type>/size:65536/num_columns:300',
'repetitions': 1, 'repetition_index': 0, 'threads': 1, 'iterations': 13406}
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
