apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 08:37:06 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Sync] Update to latest code (#2679)

Browse Source 
* [Sync] Update to latest code

* Add new code files

* Add new code files

* update code

* Try to fix build.sh

* Try to fix build.sh

* Update code

* Update requirements.txt

* Update code

---------

Co-authored-by: Jiang-Jia-Jun <jiangjiajun@baidu.com>
This commit is contained in:
Jiang-Jia-Jun
2025-07-03 15:43:53 +08:00
committed by GitHub
parent d222248d00
commit 05c670e593
95 changed files with 9916 additions and 1312 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
fastdeploy/worker/eplb.py
View File

@@ -84,9 +84,10 @@ def replicate_experts(
return phy2log, rank, logcnt
def rebalance_experts_hierarchical(weight: np.ndarray,
num_physical_experts: int, num_groups: int,
num_nodes: int, num_gpus: int):
def rebalance_experts_hierarchical(
weight: np.ndarray, num_physical_experts: int, num_groups: int,
num_nodes: int,
num_gpus: int) -> Tuple[np.ndarray, np.ndarray, np.ndarray]:
"""
Parameters:
weight: [num_moe_layers, num_logical_experts]