计算flops和add

Magic_o

于 2021-12-30 15:10:50 发布

阅读量285

点赞数

CC 4.0 BY-SA版权

文章标签：深度学习 pytorch 人工智能

本文链接：https://blog.youkuaiyun.com/Magic_o/article/details/122237617

from model import dfsfne
import torch
from ptflops import get_model_complexity_info
from option import args

with torch.cuda.device(0):
  net  = dfsfne.DFSFN(args)
  flops, params = get_model_complexity_info(net, (3, 360, 240), as_strings=True, print_per_layer_stat=True)
  print('{:<30}  {:<8}'.format('Computational complexity: ', flops))
  print('{:<30}  {:<8}'.format('Number of parameters: ', params))