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

Jump to bottom

Add missing save restore connector to eval scripts #6935

Merged
merged 3 commits into from
Jun 29, 2023
Merged

Add missing save restore connector to eval scripts #6935

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
7d32429
Add missing save restore connector to eval scripts
titu1994 Jun 28, 2023
ec86045
Merge branch 'main' into fix_eval
titu1994 Jun 29, 2023
5d28fb5
Merge branch 'main' into fix_eval
ericharper Jun 29, 2023
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
8 changes: 7 additions & 1 deletion examples/nlp/language_modeling/megatron_gpt_eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -174,8 +174,14 @@ def main(cfg) -> None:
or cfg.pipeline_model_parallel_size < 0
or cfg.get('pipeline_model_parallel_split_rank', -1) < 0
):
save_restore_connector = NLPSaveRestoreConnector()
if os.path.isdir(cfg.gpt_model_file):
save_restore_connector.model_extracted_dir = cfg.gpt_model_file
model_config = MegatronGPTModel.restore_from(
restore_path=cfg.gpt_model_file, trainer=trainer, return_config=True,
restore_path=cfg.gpt_model_file,
trainer=trainer,
return_config=True,
save_restore_connector=save_restore_connector,
)

with open_dict(cfg):
Expand Down
10 changes: 9 additions & 1 deletion examples/nlp/language_modeling/megatron_t5_eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,7 @@
# limitations under the License.


import os
from argparse import ArgumentParser

import torch
Expand Down Expand Up @@ -61,8 +62,15 @@ def main():
or args.pipeline_model_parallel_size < 0
or args.pipeline_model_parallel_split_rank < 0
):
save_restore_connector = NLPSaveRestoreConnector()
if os.path.isdir(args.model_file):
save_restore_connector.model_extracted_dir = args.model_file

model_config = MegatronT5Model.restore_from(
restore_path=args.model_file, trainer=Trainer(strategy=NLPDDPStrategy()), return_config=True,
restore_path=args.model_file,
trainer=Trainer(strategy=NLPDDPStrategy()),
return_config=True,
save_restore_connector=save_restore_connector,
)

args.tensor_model_parallel_size = model_config.get('tensor_model_parallel_size', 1)
Expand Down
2 changes: 1 addition & 1 deletion examples/nlp/language_modeling/tuning/megatron_gpt_ia3_eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@
# See the License for the specific language governing permissions and
# limitations under the License.


import os
import torch
import torch.multiprocessing as mp
from megatron.core import parallel_state
Expand Down