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

Updating Megatron LM conversion according to PTL 1.7 #5038

Merged
merged 3 commits into from
Oct 3, 2022
Merged

Updating Megatron LM conversion according to PTL 1.7 #5038

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
5b8148d
fixed megatron lm conversion bug (PTL related)
Davood-M Sep 29, 2022
4af7151
Merge branch 'r1.12.0' into davidm-3803163
Davood-M Sep 29, 2022
360cc22
Merge branch 'r1.12.0' into davidm-3803163
Davood-M Sep 30, 2022
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
7 changes: 4 additions & 3 deletions examples/nlp/language_modeling/megatron_lm_ckpt_to_nemo.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,6 +39,7 @@

import torch
from apex.transformer import parallel_state
from pytorch_lightning.core.saving import _load_state as ptl_load_state
from pytorch_lightning.core.saving import load_hparams_from_tags_csv, load_hparams_from_yaml
from pytorch_lightning.trainer.trainer import Trainer
from pytorch_lightning.utilities.cloud_io import load as pl_load
Expand Down Expand Up @@ -220,10 +221,10 @@ def add_optimizer_state(lm_checkpoint, new_checkpoint, megatron_amp_o2=True):
def load_model(cls, checkpoint, strict, **kwargs):
try:
if 'cfg' in kwargs:
model = cls._load_model_state(checkpoint, strict=strict, **kwargs)
model = ptl_load_state(cls, checkpoint, strict=strict, **kwargs)
else:
model = cls._load_model_state(
checkpoint, strict=strict, cfg=checkpoint[cls.CHECKPOINT_HYPER_PARAMS_KEY].cfg, **kwargs
model = ptl_load_state(
cls, checkpoint, strict=strict, cfg=checkpoint[cls.CHECKPOINT_HYPER_PARAMS_KEY].cfg, **kwargs
)
# register the artifacts
cfg = checkpoint[cls.CHECKPOINT_HYPER_PARAMS_KEY].cfg
Expand Down