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

[ Misc ] Remove fp8_shard_indexer from Col/Row Parallel Linear (Simplify Weight Loading) #5928

Merged
merged 6 commits into from
Jun 28, 2024
Merged

[ Misc ] Remove fp8_shard_indexer from Col/Row Parallel Linear (Simplify Weight Loading) #5928

Changes from 3 commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
fed75ad
remove scales shard indexer fp8 in row parallel linear
Jun 27, 2024
7d8e9cd
removed from column parallel linear too
Jun 27, 2024
2803761
format
Jun 27, 2024
03cb10d
stash
Jun 27, 2024
1148d87
update comment
Jun 27, 2024
1bded5f
Merge branch 'main' into remove-fp8-shard-indexer-row
Jun 28, 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
26 changes: 6 additions & 20 deletions vllm/model_executor/layers/linear.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -269,10 +269,6 @@ def __init__(self,
self.register_parameter("bias", None)

def weight_loader(self, param: Parameter, loaded_weight: torch.Tensor):
# Special case for Fp8 scales.
fp8_scales_shard_indexer = getattr(param, "fp8_scales_shard_indexer",
None)

tp_rank = get_tensor_model_parallel_rank()
output_dim = getattr(param, "output_dim", None)
param_data = param.data
Expand All @@ -281,11 +277,11 @@ def weight_loader(self, param: Parameter, loaded_weight: torch.Tensor):
start_idx = tp_rank * shard_size
loaded_weight = loaded_weight.narrow(output_dim, start_idx,
shard_size)
# Special case for Fp8 scales.
elif fp8_scales_shard_indexer is not None:
param_data, loaded_weight = fp8_scales_shard_indexer(param_data,
loaded_weight,
shard_id=0)

# Special case for loading scales off disk, which often
# do not have a shape.
if len(loaded_weight.shape) == 0:
loaded_weight = loaded_weight.reshape(1)

assert param_data.shape == loaded_weight.shape
param_data.copy_(loaded_weight)
Expand Down Expand Up @@ -751,10 +747,6 @@ def __init__(self,
self.register_parameter("bias", None)

def weight_loader(self, param: Parameter, loaded_weight: torch.Tensor):
# Special case for Fp8 scales.
fp8_scales_shard_indexer = getattr(param, "fp8_scales_shard_indexer",
None)

tp_rank = get_tensor_model_parallel_rank()
input_dim = getattr(param, "input_dim", None)
param_data = param.data
Expand All @@ -764,13 +756,7 @@ def weight_loader(self, param: Parameter, loaded_weight: torch.Tensor):
loaded_weight = loaded_weight.narrow(input_dim, start_idx,
shard_size)

# Special case for Fp8 scales.
elif fp8_scales_shard_indexer is not None:
param_data, loaded_weight = fp8_scales_shard_indexer(param_data,
loaded_weight,
shard_id=0)

if fp8_scales_shard_indexer is None and len(loaded_weight.shape) == 0:
if len(loaded_weight.shape) == 0:
loaded_weight = loaded_weight.reshape(1)

assert param_data.shape == loaded_weight.shape
Expand Down
Loading