Skip to content

Commit 3e61e24

Browse files
authored
Rename benchmark folder (#258)
1 parent 9f7158a commit 3e61e24

File tree

3 files changed

+10
-10
lines changed

3 files changed

+10
-10
lines changed

.gitignore

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -88,4 +88,4 @@ venv
8888
CLAUDE.md
8989
triton
9090
torch
91-
benchmark/tritonbench
91+
benchmarks/tritonbench

benchmark/README.md renamed to benchmarks/README.md

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -2,12 +2,12 @@
22

33
Performance comparison between Helion, torch.compile, Triton, and PyTorch eager is done by leveraging [TritonBench](https://github.com/pytorch-labs/tritonbench).
44

5-
Currently supported kernels for performance comparison are listed in `KERNEL_MAPPINGS` in `benchmark/run.py`.
5+
Currently supported kernels for performance comparison are listed in `KERNEL_MAPPINGS` in `benchmarks/run.py`.
66

77
To run the benchmark:
88

9-
`$ python benchmark/run.py --metrics speedup,accuracy --kernel <kernel_name>`
9+
`$ python benchmarks/run.py --metrics speedup,accuracy --kernel <kernel_name>`
1010

1111
e.g. for `vector_add` kernel:
1212

13-
`$ python benchmark/run.py --metrics speedup,accuracy --kernel vector_add`
13+
`$ python benchmarks/run.py --metrics speedup,accuracy --kernel vector_add`

benchmark/run.py renamed to benchmarks/run.py

Lines changed: 6 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -1,12 +1,12 @@
11
"""Performance comparison between Helion, torch.compile, Triton, and PyTorch eager by leveraging TritonBench.
22
3-
Currently supported kernels are listed in `KERNEL_MAPPINGS` in `benchmark/run.py`.
3+
Currently supported kernels are listed in `KERNEL_MAPPINGS` in `benchmarks/run.py`.
44
55
Usage:
6-
$ python benchmark/run.py [tritonbench args...] --kernel <kernel_name>
6+
$ python benchmarks/run.py [tritonbench args...] --kernel <kernel_name>
77
88
Example usage:
9-
$ python benchmark/run.py --metrics speedup,accuracy --kernel vector_add
9+
$ python benchmarks/run.py --metrics speedup,accuracy --kernel vector_add
1010
"""
1111

1212
from __future__ import annotations
@@ -71,9 +71,9 @@ def check_and_setup_tritonbench() -> None:
7171

7272
print("Tritonbench not found. Installing...", file=sys.stderr)
7373

74-
# Clone to benchmark/tritonbench
75-
benchmark_dir = Path(__file__).parent
76-
tritonbench_path = benchmark_dir / "tritonbench"
74+
# Clone to benchmarks/tritonbench
75+
benchmarks_dir = Path(__file__).parent
76+
tritonbench_path = benchmarks_dir / "tritonbench"
7777

7878
try:
7979
# Clone the repository if it doesn't exist

0 commit comments

Comments
 (0)