Skip to content

[Fix] rank0 sample.get_state_dict dont set sample.step #1233

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
wants to merge 1 commit into
base: main
Choose a base branch
from
+6 −3
Open

[Fix] rank0 sample.get_state_dict dont set sample.step #1233

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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: 6 additions & 3 deletions xtuner/v1/datasets/sampler.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -107,6 +107,7 @@ def __iter__(self) -> Iterator[int]:

def __len__(self) -> int:
"""The number of samples in this rank."""
# TODO: not same with LengthGroupedSampler?
return self.num_samples - self.step

def set_epoch(self, epoch: int) -> None:
Expand Down Expand Up @@ -137,10 +138,11 @@ def load_state_dict(self, state_dict) -> None:
)

def get_state_dict(self, step: int):
self.step = step % self.total_size
# Attention! Do not set self.step here, or it will cause the next __iter__ to get less samples.
# self.step = step % self.total_size
return {
"epoch": self.epoch,
"step": self.step,
"step": step,
"world_size": self.world_size,
"shuffle": self.shuffle,
"round_up": self.round_up,
Expand Down Expand Up @@ -291,7 +293,8 @@ def get_state_dict(self, step: int):
Returns:
dict: The state of the sampler.
"""
self.step = step % self.total_size
# Attention! Do not set self.step here, or it will cause the next __iter__ to get less samples.
# self.step = step % self.total_size
return {
"epoch": self.epoch,
"step": self.step,
Expand Down
Loading