pyotrch nn.Conv2d中groups参数的理解

在pytorch的Docs中有关于nn.Conv2d的具体描述：

 torch.nn.Conv2d(in_channels, out_channels, kernel_size, stride=1, padding=0, dilation=1, groups=1, bias=True)
 

https://zhuanlan.zhihu.com/p/35405071

  

   

    

     

    
    

     

      import torch
     
    
   

    

     

    
    

     

      import torch.nn as nn
     
    
   

    

     

    
    

     

      from torch.autograd import Variable
     
    
   

    

     

    
    

     
 
     
    
   

    

     

    
    

     

      input = torch.ones(1, 3, 224, 224)
     
    
   

    

     

    
    

     

      input = Variable(input)
     
    
   

    

     

    
    

     

      f = nn.Conv2d(in_channels=3, out_channels=9, kernel_size=5, groups=3)
     
    
   

    

     

    
    

     

      output = f(input)
     
    
   

    

     

    
    

     

      print(output.shape) # (1, 9, 220, 220)
     
    
  
 

我们通过实际的例子加以说明：

# pytorch 0.3.0
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Wed May  2 20:13:05 2018

@author: huijian
"""

# experiment about groups
import torch
import torch.nn as nn
from torch.autograd import Varia

文章来源: blog.csdn.net，作者：网奇，版权归原作者所有，如需转载，请联系作者。

原文链接：blog.csdn.net/jacke121/article/details/104008918