biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Fix per file ruff ignores related to simplification (#26259)

Browse Source 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-10-05 21:31:53 +01:00
committed by GitHub
parent 6b6e98775f
commit b893d661b1
32 changed files with 47 additions and 191 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
vllm/model_executor/layers/fused_moe/layer.py
View File

@@ -886,10 +886,7 @@ def determine_expert_map(
# Distribute experts as evenly as possible to each rank.
base_experts = global_num_experts // ep_size
remainder = global_num_experts % ep_size
if ep_rank < remainder:
local_num_experts = base_experts + 1
else:
local_num_experts = base_experts
local_num_experts = base_experts + 1 if ep_rank < remainder else base_experts
# Create a tensor of size num_experts filled with -1
expert_map = torch.full((global_num_experts,), -1, dtype=torch.int32)