torch 模型 gpu转cpu

发表于 | 阅读次数: 
1
2
3
4
5
6
7
8
9
10
11
require 'cutorch'
require 'cudnn'
require 'torch'
require 'nn'
require 'nnx'
require 'optim'
require 'nngraph'

m = torch.load('foo.t7')
m = m:float()
torch.save('foo.cpu.t8',m)
请作者喝一杯咖啡☕️