Skip to content

Compute model param count once #204

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
jaketae wants to merge 6 commits into
base: main
Choose a base branch
from
Open

Compute model param count once #204

Changes from 1 commit
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
544108d
refactor: compute model param count once
Nov 24, 2021
816b867
Update megatron/training.py
jaketae Nov 24, 2021
c2d6390
Update megatron/training.py
stas00 Nov 25, 2021
f4c7c67
fix: use deepspeed param count method
jaketae Dec 1, 2021
a7b10b7
refactor: replace filter w/ list comp, generator to list
jaketae Dec 1, 2021
ac3e138
refactor: use set for constant time lookup
jaketae Dec 1, 2021
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
10 changes: 6 additions & 4 deletions megatron/training.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -134,8 +134,10 @@ def pretrain(train_valid_test_dataset_provider,
# Model, optimizer, and learning rate.
timers('model-and-optimizer-setup').start()
model, optimizer, lr_scheduler = setup_model_and_optimizer(model_provider)
print_rank_0(f'estimated model parameters: {get_parameters_in_billions(model)}')
print_rank_0(f'estimated model parameters without embeddings: {get_parameters_in_billions(model, exclude_embeddings=True)}')
args.parameters_in_billions = get_parameters_in_billions(model)
args.parameters_in_billions_no_embedding = get_parameters_in_billions(model, exclude_embeddings=True)
print_rank_0(f'estimated model parameters: {args.parameters_in_billions}')
print_rank_0(f'estimated model parameters without embeddings: {args.parameters_in_billions_no_embedding}')
Comment thread
jaketae marked this conversation as resolved.
Outdated
timers('model-and-optimizer-setup').stop()
print_datetime('after model, optimizer, and learning rate '
'scheduler are built')
Expand Down Expand Up @@ -740,7 +742,7 @@ def train(forward_step_func, model, optimizer, lr_scheduler,
tp_rank = mpu.get_tensor_model_parallel_rank()
pp_rank = mpu.get_pipeline_model_parallel_rank()
preamble = f"[{tp_rank:0>3d}-{pp_rank:0>3d}]"
print(f"{preamble} {get_parameters_in_billions(model):.4f}B / {get_parameters_in_billions(model, exclude_embeddings=True):.4f}B")
print(f"{preamble} {args.parameters_in_billions:.4f}B / {args.parameters_in_billions_no_embedding:.4f}B")
Comment thread
jaketae marked this conversation as resolved.
Outdated
Comment thread
stas00 marked this conversation as resolved.
Outdated
torch.distributed.barrier()
else:
torch.distributed.barrier()
Expand Down Expand Up @@ -815,7 +817,7 @@ def train(forward_step_func, model, optimizer, lr_scheduler,
args.consumed_train_tokens += new_samples * args.curriculum_seqlen
else:
args.consumed_train_tokens += new_samples * args.seq_length
args.gigaflos_no_embeds += (6 * new_samples * args.seq_length * get_parameters_in_billions(model, exclude_embeddings=True))
args.gigaflos_no_embeds += (6 * new_samples * args.seq_length * args.parameters_in_billions_no_embedding)

# Logging.
if args.deepspeed:
Expand Down