Skip to content

Rename submission modes: benchmark→private, leaderboard→public #100

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
msaroufim wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Rename submission modes: benchmark→private, leaderboard→public #100

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions problems/amd/eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -349,10 +349,10 @@ def main():
with mp_context.Pool(1) as pool:
if mode == "test":
return run_testing(logger, pool, tests)
if mode == "benchmark":
if mode == "private":
return run_benchmarking(logger, pool, tests)

if mode == "leaderboard":
if mode == "public":
# warmup
run_single_benchmark(pool, tests[0], False, 100, 1e7)
logger.log("benchmark-count", len(tests))
Expand Down
6 changes: 3 additions & 3 deletions problems/amd/mla-decode/eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -294,10 +294,10 @@ def main():
if mode == "test":
return run_testing(logger, tests)

if mode == "benchmark":
if mode == "private":
return run_benchmarking(logger, tests)
if mode == "leaderboard":

if mode == "public":
warm_up(tests[0])
result = benchmark(tests[-1], True, 100, 30e9)
if isinstance(result, Stats):
Expand Down
4 changes: 2 additions & 2 deletions problems/amd_distributed/eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -546,10 +546,10 @@ def main():
with mp_context.Pool(n_gpus) as pool:
if mode == "test":
return run_testing(logger, pool, tests)
if mode == "benchmark":
if mode == "private":
return run_benchmarking(logger, pool, tests)

if mode == "leaderboard":
if mode == "public":
# warmup
run_single_benchmark(pool, tests[0], False, 100, 1e7)
logger.log("benchmark-count", len(tests))
Expand Down
4 changes: 2 additions & 2 deletions problems/bioml/trimul/eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -352,10 +352,10 @@ def main():
with mp_context.Pool(1) as pool:
if mode == "test":
return run_testing(logger, pool, tests)
if mode == "benchmark":
if mode == "private":
return run_benchmarking(logger, pool, tests)

if mode == "leaderboard":
if mode == "public":
# warmup
run_single_benchmark(pool, tests[0], False, 100, 1e7)
logger.log("benchmark-count", len(tests))
Expand Down
4 changes: 2 additions & 2 deletions problems/nvidia/eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -449,10 +449,10 @@ def main():
with mp_context.Pool(1) as pool:
if mode == "test":
return run_testing(logger, pool, tests)
if mode == "benchmark":
if mode == "private":
return run_benchmarking(logger, pool, tests)

if mode == "leaderboard":
if mode == "public":
run_single_benchmark(pool, tests[0], False, 1000, 5e8)
logger.log("benchmark-count", len(tests))
passed = True
Expand Down
4 changes: 2 additions & 2 deletions problems/nvidia/eval_better_bench.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -472,10 +472,10 @@ def main():
with mp_context.Pool(1, initializer=_init_worker) as pool:
if mode == "test":
return run_testing(logger, pool, tests)
if mode == "benchmark":
if mode == "private":
return run_benchmarking(logger, pool, tests)

if mode == "leaderboard":
if mode == "public":
# Warmup all test shapes to ensure consistent benchmarking
for test in tests:
run_single_benchmark(pool, test, False, 1000, 5e8)
Expand Down
4 changes: 2 additions & 2 deletions problems/nvidia/eval_better_bench_grouped_gemm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -491,10 +491,10 @@ def main():
with mp_context.Pool(1, initializer=_init_worker) as pool:
if mode == "test":
return run_testing(logger, pool, tests)
if mode == "benchmark":
if mode == "private":
return run_benchmarking(logger, pool, tests)

if mode == "leaderboard":
if mode == "public":
# Warmup all test shapes to ensure consistent benchmarking
for test in tests:
run_single_benchmark(pool, test, False, 50, 5e8)
Expand Down
4 changes: 2 additions & 2 deletions problems/nvidia/nvfp4_gemm/eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -452,10 +452,10 @@ def build_test_string(tests: list[dict]):
with mp_context.Pool(1) as pool:
if mode == "test":
return run_testing(logger, pool, tests)
if mode == "benchmark":
if mode == "private":
return run_benchmarking(logger, pool, tests)

if mode == "leaderboard":
if mode == "public":
# Step 1: Compile kernel once (outside of timing)
logger.log("compile", "start")
compile_success, compile_error = pool.apply(_compile_kernel_once)
Expand Down
4 changes: 2 additions & 2 deletions problems/nvidia/nvfp4_group_gemm/eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -392,10 +392,10 @@ def main():
with mp_context.Pool(1) as pool:
if mode == "test":
return run_testing(logger, pool, tests)
if mode == "benchmark":
if mode == "private":
return run_benchmarking(logger, pool, tests)

if mode == "leaderboard":
if mode == "public":
# warmup
run_single_benchmark(pool, tests[0], False, 100, 1e7)
logger.log("benchmark-count", len(tests))
Expand Down
6 changes: 3 additions & 3 deletions problems/pmpp/eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -246,10 +246,10 @@ def main():
if mode == "test":
return run_testing(logger, tests)

if mode == "benchmark":
if mode == "private":
return run_benchmarking(logger, tests)
if mode == "leaderboard":

if mode == "public":
warm_up(tests[0])
result = benchmark(tests[-1], True, 100, 30e9)
if isinstance(result, Stats):
Expand Down
4 changes: 2 additions & 2 deletions problems/pmpp_v2/eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -343,10 +343,10 @@ def main():
with mp_context.Pool(1) as pool:
if mode == "test":
return run_testing(logger, pool, tests)
if mode == "benchmark":
if mode == "private":
return run_benchmarking(logger, pool, tests)

if mode == "leaderboard":
if mode == "public":
# warmup
run_single_benchmark(pool, tests[0], False, 100, 1e7)
logger.log("benchmark-count", len(tests))
Expand Down