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

Disable nvfuser for gpt #3845

Merged
merged 4 commits into from
Mar 16, 2022
Merged

Disable nvfuser for gpt #3845

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
1c0d303
disable nvfuser
ericharper Mar 15, 2022
2de2ee3
disable nvfuser
ericharper Mar 15, 2022
795cbbc
delete line
ericharper Mar 15, 2022
89c3241
Merge branch 'main' into disable_nvfuser_gpt
ericharper Mar 16, 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
2 changes: 1 addition & 1 deletion examples/nlp/language_modeling/conf/megatron_gpt_config.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -106,7 +106,7 @@ model:
splits_string: 900,50,50
seq_length: ${model.encoder_seq_length}
skip_warmup: True
num_workers: 0
num_workers: 2
dataloader_type: single # cyclic
reset_position_ids: False # Reset position ids after end-of-document token
reset_attention_mask: False # Reset attention mask after end-of-document token
Expand Down
7 changes: 3 additions & 4 deletions nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -106,7 +106,7 @@ def __init__(self, cfg: DictConfig, trainer: Trainer):
self._validate_trainer()

# used in NVIDIA NGC PyTorch containers
self._enable_nvidia_optimizations()
# self._enable_nvidia_optimizations()

if self.cfg.get('use_cpu_initialization', False) is False:
torch.cuda.set_device(trainer.local_rank)
Expand Down Expand Up @@ -558,9 +558,8 @@ def build_train_valid_test_datasets(self):
return

logging.info('Building GPT datasets.')
global_batch_size = self.trainer.world_size * self.cfg.micro_batch_size / self.cfg.tensor_model_parallel_size
# Compute trianing micro-batch steps: total_global_batch_steps x grad_acumms_per_global_batch
max_train_steps = self.trainer.max_steps * self.trainer.accumulate_grad_batches
global_batch_size = self.cfg.global_batch_size
max_train_steps = self.trainer.max_steps
eval_iters = (max_train_steps // self.trainer.val_check_interval + 1) * self.trainer.limit_val_batches
test_iters = self.trainer.limit_test_batches

Expand Down