tensorflow-yolov3 icon indicating copy to clipboard operation
tensorflow-yolov3 copied to clipboard

求救!这个框架支持多卡训练吗,为什么我用服务器训练只能用一个卡啊,需要怎么设置啊,求告知!感激不尽!

Open lukaisheng1203 opened this issue 4 years ago • 3 comments

@YunYang1994

lukaisheng1203 avatar Jul 18 '19 14:07 lukaisheng1203

如果你只是简单需要多张显卡训练,那么 CUDA_VISIBLE_DEVICES=0, 1, 2, 3 python train.py;如果你是要多张显卡分别读取数据进行训练,然后它们之间还可以通信进行共享参数,那么你得自己写代码了。

YunYang1994 avatar Jul 19 '19 07:07 YunYang1994

https://stackoverflow.com/questions/39649102/how-do-i-select-which-gpu-to-run-a-job-on

参考这个

import os os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID" os.environ["CUDA_VISIBLE_DEVICES"] = "0" ... import pycuda.autoinit import tensorflow as tf ...

kalayan avatar Feb 25 '20 04:02 kalayan

目前这样设置, CUDA_VISIBLE_DEVICES=0, 1, 2, 3 python train.py,发现只是卡0上占用显存,这是什么问题呢?@YunYang1994

Wuqiman avatar Apr 14 '22 05:04 Wuqiman