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

Refactor system architecture (#82)

Browse Source
This commit is contained in:
Woosuk Kwon
2023-05-09 15:30:12 -07:00
committed by GitHub
parent 8917782af6
commit 7c041ab578
40 changed files with 194 additions and 446 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
cacheflow/utils.py
View File

@ -1,13 +1,8 @@
import enum
import random
import psutil
import numpy as np
import torch
from cacheflow.parallel_utils.parallel_state import model_parallel_is_initialized
from cacheflow.parallel_utils.tensor_parallel import model_parallel_cuda_manual_seed
class Device(enum.Enum):
GPU = enum.auto()
@ -28,17 +23,6 @@ class Counter:
self.counter = 0
def set_random_seed(seed: int):
random.seed(seed)
np.random.seed(seed)
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)
def get_gpu_memory(gpu: int = 0) -> int:
return torch.cuda.get_device_properties(gpu).total_memory