Add inline evaluation results for summary evaluators #757
isahers1Annotations
1 warning and 2 notices
|
benchmark
ubuntu-latest pipelines will use ubuntu-24.04 soon. For more details, see https://github.com/actions/runner-images/issues/10636
|
|
Benchmark results:
python/langsmith/evaluation/_arunner.py#L1
...........
create_5_000_run_trees: Mean +- std dev: 681 ms +- 45 ms
...........
create_10_000_run_trees: Mean +- std dev: 1.44 sec +- 0.11 sec
...........
create_20_000_run_trees: Mean +- std dev: 1.43 sec +- 0.13 sec
...........
dumps_class_nested_py_branch_and_leaf_200x400: Mean +- std dev: 724 us +- 14 us
...........
dumps_class_nested_py_leaf_50x100: Mean +- std dev: 26.9 ms +- 0.6 ms
...........
dumps_class_nested_py_leaf_100x200: Mean +- std dev: 111 ms +- 4 ms
...........
dumps_dataclass_nested_50x100: Mean +- std dev: 27.8 ms +- 1.0 ms
...........
WARNING: the benchmark result may be unstable
* the standard deviation (27.9 ms) is 30% of the mean (93.4 ms)
Try to rerun the benchmark with more runs, values and/or loops.
Run 'python -m pyperf system tune' command to reduce the system jitter.
Use pyperf stats, pyperf dump and pyperf hist to analyze results.
Use --quiet option to hide these warnings.
dumps_pydantic_nested_50x100: Mean +- std dev: 93.4 ms +- 27.9 ms
...........
dumps_pydanticv1_nested_50x100: Mean +- std dev: 209 ms +- 4 ms
|
|
Comparison against main:
python/langsmith/evaluation/_arunner.py#L1
+-----------------------------------------------+----------+------------------------+
| Benchmark | main | changes |
+===============================================+==========+========================+
| create_5_000_run_trees | 739 ms | 681 ms: 1.09x faster |
+-----------------------------------------------+----------+------------------------+
| dumps_pydanticv1_nested_50x100 | 222 ms | 209 ms: 1.06x faster |
+-----------------------------------------------+----------+------------------------+
| create_10_000_run_trees | 1.46 sec | 1.44 sec: 1.01x faster |
+-----------------------------------------------+----------+------------------------+
| create_20_000_run_trees | 1.44 sec | 1.43 sec: 1.01x faster |
+-----------------------------------------------+----------+------------------------+
| dumps_class_nested_py_branch_and_leaf_200x400 | 690 us | 724 us: 1.05x slower |
+-----------------------------------------------+----------+------------------------+
| dumps_class_nested_py_leaf_50x100 | 25.2 ms | 26.9 ms: 1.07x slower |
+-----------------------------------------------+----------+------------------------+
| dumps_class_nested_py_leaf_100x200 | 104 ms | 111 ms: 1.07x slower |
+-----------------------------------------------+----------+------------------------+
| dumps_dataclass_nested_50x100 | 25.4 ms | 27.8 ms: 1.09x slower |
+-----------------------------------------------+----------+------------------------+
| dumps_pydantic_nested_50x100 | 67.6 ms | 93.4 ms: 1.38x slower |
+-----------------------------------------------+----------+------------------------+
| Geometric mean | (ref) | 1.05x slower |
+-----------------------------------------------+----------+------------------------+
|