Skip to content
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

Update maxtext checkpoitin test to add async #571

Merged
merged 7 commits into from
Jan 27, 2025
Merged

Update maxtext checkpoitin test to add async #571

Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
9aca39e
Update maxtext checkpoitin test to add async
anfals Jan 18, 2025
f6f71d2
Merge branch 'master' into anfals/maxtext_checkpointing_async
anfals Jan 24, 2025
65a25f9
Minor tweaks to test name and to improve readability
anfals Jan 24, 2025
966b82d
formatter
anfals Jan 24, 2025
22895ce
formatter
anfals Jan 24, 2025
773f13e
explicitly pass args
anfals Jan 26, 2025
0679e02
Merge branch 'master' into anfals/maxtext_checkpointing_async
anfals Jan 26, 2025
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
30 changes: 16 additions & 14 deletions dags/multipod/maxtext_checkpointing.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,17 +55,19 @@
for accelerator, slices in test_configs.items():
cores = accelerator.rsplit("-", maxsplit=1)[-1]
for slice_num in slices:
command = (
"bash end_to_end/test_checkpointing.sh"
f" checkpointing-{mode.value}-{slice_num}x-{accelerator}"
f" {base_output_directory} {dataset_path} true",
)
maxtext_v4_configs_test = gke_config.get_gke_config(
num_slices=slice_num,
cluster=clusters[accelerator],
time_out_in_min=60,
test_name=f"maxtext-checkpointing-{mode.value}",
run_model_cmds=command,
docker_image=image.value,
test_owner=test_owner.SURBHI_J,
).run()
for chkpt_mode in ["sync", "async"]:
async_checkpointing = chkpt_mode == "async"
command = (
"bash end_to_end/test_checkpointing.sh"
f" checkpointing-{mode.value}-{slice_num}x-{accelerator}-{chkpt_mode}"
f" {base_output_directory} {dataset_path} true tfds autoselected {async_checkpointing}"
)
maxtext_v4_configs_test = gke_config.get_gke_config(
num_slices=slice_num,
cluster=clusters[accelerator],
time_out_in_min=60,
test_name=f"maxtext-checkpointing-{mode.value}-{chkpt_mode}",
run_model_cmds=command,
docker_image=image.value,
test_owner=test_owner.SURBHI_J,
).run()
Loading