Skip to content

Add liger smoketest #591

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
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Add liger smoketest #591

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
3967f3a
add Liger smoketest; parameterize LoRA test
JamesKunstle Jun 3, 2025
a3e0fde
Revert changes to test parameters
booxter Jun 20, 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
16 changes: 13 additions & 3 deletions tests/smoke/test_train.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,7 @@
from instructlab.training.config import (
DataProcessArgs,
DistributedBackend,
LoraOptions,
TorchrunArgs,
TrainingArgs,
)
Expand Down Expand Up @@ -231,25 +232,36 @@ def cached_training_data(
@pytest.mark.parametrize(
"dist_backend", [DistributedBackend.FSDP, DistributedBackend.DEEPSPEED]
)
@pytest.mark.parametrize("cpu_offload", [True, False])
@pytest.mark.parametrize("cpu_offload", [False, True])
@pytest.mark.parametrize("lora_rank", [0])
@pytest.mark.parametrize("use_liger", [False, True])
def test_training_feature_matrix(
cached_test_model: pathlib.Path,
cached_training_data: pathlib.Path,
checkpoint_dir: pathlib.Path,
prepared_data_dir: pathlib.Path,
use_liger: bool,
lora_rank: int,
cpu_offload: bool,
dist_backend: DistributedBackend,
) -> None:
torch_args = TorchrunArgs(**DEFAULT_TORCHRUN_ARGS)
train_args = TrainingArgs(
model_path=str(cached_test_model),
data_path=str(cached_training_data),
data_output_dir=str(prepared_data_dir),
ckpt_output_dir=str(checkpoint_dir),
lora=LoraOptions(rank=lora_rank),
use_liger=use_liger,
**MINIMAL_TRAINING_ARGS,
)

train_args.distributed_backend = dist_backend

if lora_rank > 0:
# LoRA doesn't support full state saving.
train_args.accelerate_full_state_at_epoch = False

if dist_backend == DistributedBackend.FSDP:
train_args.fsdp_options.cpu_offload_params = cpu_offload
else:
Expand All @@ -258,6 +270,4 @@ def test_training_feature_matrix(
pytest.xfail("DeepSpeed CPU Adam isn't currently building correctly")
train_args.deepspeed_options.cpu_offload_optimizer = cpu_offload

torch_args = TorchrunArgs(**DEFAULT_TORCHRUN_ARGS)

run_training(torch_args=torch_args, train_args=train_args)
239 changes: 0 additions & 239 deletions tests/smoketest.sh
View file
Open in desktop

This file was deleted.

Loading