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

fixes #131, module 'eole.utils' has no attribute 'distributed' error when training multi-gpu #132

Merged
merged 7 commits into from
Oct 25, 2024
Merged

fixes #131, module 'eole.utils' has no attribute 'distributed' error when training multi-gpu #132

Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
f72eb32
fix issue #131, module 'eole.utils' has no attribute 'distributed' er…
isanvicente Oct 22, 2024
f6b6fdd
fix issue #131, import only functions
isanvicente Oct 22, 2024
1250387
Merge branch 'eole-nlp:main' into main
isanvicente Oct 23, 2024
b733b73
apply black formatter to eole/trainer.py
isanvicente Oct 23, 2024
9bec8b1
Merge branch 'eole-nlp:main' into main
isanvicente Oct 23, 2024
7ebf696
make flake happy
francoishernandez Oct 24, 2024
ab6e151
make black happy
francoishernandez Oct 24, 2024
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
9 changes: 3 additions & 6 deletions eole/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,7 @@
import torch
import traceback
import eole.utils
from eole.utils.distributed import all_gather_list, all_reduce_and_rescale_tensors
from eole.utils.loss import LossCompute
from eole.utils.logging import logger
from eole.utils.misc import clear_gpu_cache, get_autocast
Expand Down Expand Up @@ -333,9 +334,7 @@ def train(
self._maybe_update_estim_lambda(step)

if self.n_gpu > 1 and self.parallel_mode == "data_parallel":
normalization = sum(
eole.utils.distributed.all_gather_list(normalization)
)
normalization = sum(all_gather_list(normalization))

self._gradient_accumulation(
batches, normalization, total_stats, report_stats
Expand Down Expand Up @@ -570,9 +569,7 @@ def _gradient_accumulation(
for p in self.model.parameters()
if p.requires_grad and p.grad is not None
]
eole.utils.distributed.all_reduce_and_rescale_tensors(
grads, float(self.n_gpu)
)
all_reduce_and_rescale_tensors(grads, float(self.n_gpu))

self.optim.step()

Expand Down
Loading