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

Bug fix for the "Link bit16 and fp32 parameters in partition" #5681

Merged
merged 4 commits into from
Jun 26, 2024
Merged

Bug fix for the "Link bit16 and fp32 parameters in partition" #5681

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
f8c4db9
Bug fix for the "Link bit16 and fp32 parameters in partition" for MoE…
U-rara Jun 18, 2024
3cca313
Merge branch 'master' into dev
U-rara Jun 18, 2024
e25fb52
Merge branch 'master' into dev
tjruwase Jun 19, 2024
8e3561f
Merge branch 'master' into dev
tjruwase Jun 21, 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
4 changes: 2 additions & 2 deletions deepspeed/runtime/zero/stage_1_and_2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -573,14 +573,14 @@ def _create_param_mapping(self):
return param_mapping

def _link_all_hp_params(self):
dp_world_size = dist.get_world_size(group=self.dp_process_group)
if self.cpu_offload:
self._get_offload_gradient_dict()

for i, _ in enumerate(self.optimizer.param_groups):
# Link bit16 and fp32 params in partition
partition_id = dist.get_rank(group=self.real_dp_process_group[i])
partition_size = self.bit16_groups_flat[i].numel() // dp_world_size
partition_size = self.bit16_groups_flat[i].numel() // dist.get_world_size(
group=self.real_dp_process_group[i])
flat_hp_partition = self.single_partition_of_fp32_groups[i]
link_hp_params(lp_param_list=self.bit16_groups[i],
flat_hp_partition=flat_hp_partition,
Expand Down
Loading