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

TP/quantization/weight loading refactor part 1 - Simplify parallel linear logic (#1181)

Browse Source
This commit is contained in:
Zhuohan Li
2023-10-02 15:36:09 -07:00
committed by GitHub
parent 84e4e37d14
commit ba0bfd40e2
42 changed files with 819 additions and 1547 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
vllm/model_executor/utils.py
View File

@@ -4,9 +4,6 @@ import random
import numpy as np
import torch
from vllm.model_executor.parallel_utils.parallel_state import model_parallel_is_initialized
from vllm.model_executor.parallel_utils.tensor_parallel import model_parallel_cuda_manual_seed
def set_random_seed(seed: int) -> None:
random.seed(seed)
@@ -14,6 +11,3 @@ def set_random_seed(seed: int) -> None:
torch.manual_seed(seed)
if torch.cuda.is_available():
torch.cuda.manual_seed_all(seed)
if model_parallel_is_initialized():
model_parallel_cuda_manual_seed(seed)