1. 程式人生 > 程式設計 >Pytorch.nn.conv2d 過程驗證方式(單,多通道卷積過程)

Pytorch.nn.conv2d 過程驗證方式(單,多通道卷積過程)

今天在看文件的時候,發現pytorch 的conv操作不是很明白,於是有了一下記錄

首先提出兩個問題:

1.輸入圖片是單通道情況下的filters是如何操作的? 即一通道卷積核卷積過程

2.輸入圖片是多通道情況下的filters是如何操作的? 即多通道多個卷積核卷積過程

這裡首先貼出官方文件:

classtorch.nn.Conv2d(in_channels,out_channels,kernel_size,stride=1,padding=0,dilation=1,groups=1,bias=True)[source]

Parameters:

in_channels (int) – Number of channels in the input image

out_channels (int) – Number of channels produced by the convolution
kernel_size (intortuple) – Size of the convolving kernel
stride (intortuple,optional) – Stride of the convolution. Default: 1
padding (intortuple,optional) – Zero-padding added to both sides of the input. Default: 0
dilation (intortuple,optional) – Spacing between kernel elements. Default: 1
groups (int,optional) – Number of blocked connections from input channels to output channels. Default: 1
bias (bool,optional) – If True,adds a learnable bias to the output. Default: True

這個文件中的公式對我來說,並不能看的清楚

一通道卷積核卷積過程:

比如32個卷積核,可以學習32種特徵。在有多個卷積核時,如下圖所示:輸出就為32個feature map

也就是, 當conv2d( in_channels = 1,out_channels = N)

有N個filter對輸入進行濾波。同時輸出N個結果即feature map,每個filter濾波輸出一個結果.

import torch
from torch.autograd import Variable
##單位矩陣來模擬輸入
input=torch.ones(1,1,5,5)
input=Variable(input)
x=torch.nn.Conv2d(in_channels=1,out_channels=3,kernel_size=3,groups=1)
out=x(input)
print(out)
print(list(x.parameters()))

輸出out的結果和conv2d 的引數如下,可以看到,conv2d是有3個filter加一個bias

# out的結果
Variable containing:
(0,.,.) = 
 -0.3065 -0.3065 -0.3065
 -0.3065 -0.3065 -0.3065
 -0.3065 -0.3065 -0.3065

(0,.) = 
 -0.3046 -0.3046 -0.3046
 -0.3046 -0.3046 -0.3046
 -0.3046 -0.3046 -0.3046

(0,2,.) = 
 0.0710 0.0710 0.0710
 0.0710 0.0710 0.0710
 0.0710 0.0710 0.0710
[torch.FloatTensor of size 1x3x3x3]

# conv2d的引數
[Parameter containing:
(0,.) = 
 -0.0789 -0.1932 -0.0990
 0.1571 -0.1784 -0.2334
 0.0311 -0.2595 0.2222

(1,.) = 
 -0.0703 -0.3159 -0.3295
 0.0723 0.3019 0.2649
 -0.2217 0.0680 -0.0699

(2,.) = 
 -0.0736 -0.1608 0.1905
 0.2738 0.2758 -0.2776
 -0.0246 -0.1781 -0.0279
[torch.FloatTensor of size 3x1x3x3],Parameter containing:
 0.3255
-0.0044
 0.0733
[torch.FloatTensor of size 3]
]

驗證如下,因為是單位矩陣,所以直接對引數用sum()來模擬卷積過程:

f_p=list(x.parameters())[0]
f_p=f_p.data.numpy()
print("the result of first channel in image:",f_p[0].sum()+(0.3255))

可以看到結果是和(0,.) = -0.3065 ....一樣的. 說明操作是通過卷積求和的.

the result of first channel in image: -0.306573044777

多通道卷積核卷積過程:

下圖展示了在四個通道上的卷積操作,有兩個卷積核,生成兩個通道。其中需要注意的是,四個通道上每個通道對應一個卷積核,先將w2忽略,只看w1,那麼在w1的某位置(i,j)處的值,是由四個通道上(i,j)處的卷積結果相加得到的。 所以最後得到兩個feature map, 即輸出層的卷積核核個數為 feature map 的個數。

在pytorch 中的展示為

conv2d( in_channels = X(x>1),out_channels = N)

有N乘X個filter(N組filters,每組X 個)對輸入進行濾波。即每次有一組裡X個filter對原X個channels分別進行濾波最後相加輸出一個結果,最後輸出N個結果即feature map。

驗證如下:

##單位矩陣來模擬輸入
input=torch.ones(1,3,5)
input=Variable(input)
x=torch.nn.Conv2d(in_channels=3,out_channels=4,groups=1)
out=x(input)
print(list(x.parameters()))

可以看到共有4*3=12個filter 和一個1×4的bias 作用在這個(3,5)的單位矩陣上

## out輸出的結果
Variable containing:
(0,.) = 
 -0.6390 -0.6390 -0.6390
 -0.6390 -0.6390 -0.6390
 -0.6390 -0.6390 -0.6390

(0,.) = 
 -0.1467 -0.1467 -0.1467
 -0.1467 -0.1467 -0.1467
 -0.1467 -0.1467 -0.1467

(0,.) = 
 0.4138 0.4138 0.4138
 0.4138 0.4138 0.4138
 0.4138 0.4138 0.4138

(0,.) = 
 -0.3981 -0.3981 -0.3981
 -0.3981 -0.3981 -0.3981
 -0.3981 -0.3981 -0.3981
[torch.FloatTensor of size 1x4x3x3]

## x的引數設定
[Parameter containing:
(0,.) = 
 -0.0803 0.1473 -0.0762
 0.0284 -0.0050 -0.0246
 0.1438 0.0955 -0.0500

(0,.) = 
 0.0716 0.0062 -0.1472
 0.1793 0.0543 -0.1764
 -0.1548 0.1379 0.1143

(0,.) = 
 -0.1741 -0.1790 -0.0053
 -0.0612 -0.1856 -0.0858
 -0.0553 0.1621 -0.1822

(1,.) = 
 -0.0773 -0.1385 0.1356
 0.1794 -0.0534 -0.1110
 -0.0137 -0.1744 -0.0188

(1,.) = 
 -0.0396 0.0149 0.1537
 0.0846 -0.1123 -0.0556
 -0.1047 -0.1783 -0.0630

(1,.) = 
 0.1850 0.0325 0.0332
 -0.0487 0.0018 0.1668
 0.0569 0.0267 0.0124

(2,.) = 
 0.1880 -0.0152 -0.1088
 -0.0105 0.1805 -0.0343
 -0.1676 0.1249 0.1872

(2,.) = 
 0.0299 0.0449 0.1179
 0.1280 -0.1545 0.0593
 -0.1489 0.1378 -0.1495

(2,.) = 
 -0.0922 0.1873 -0.1163
 0.0970 -0.0682 -0.1110
 0.0614 -0.1877 0.1918

(3,.) = 
 -0.1257 -0.0814 -0.1923
 0.0048 -0.0789 -0.0048
 0.0780 -0.0290 0.1287

(3,.) = 
 -0.0649 0.0773 -0.0584
 0.0092 -0.1168 -0.0923
 0.0614 0.1159 0.0134

(3,.) = 
 0.0426 -0.1055 0.1022
 -0.0810 0.0540 -0.1011
 0.0698 -0.0799 -0.0786
[torch.FloatTensor of size 4x3x3x3],Parameter containing:
-0.1367
-0.0410
 0.0424
 0.1353
[torch.FloatTensor of size 4]
]

因為是單位矩陣,所以直接對引數用sum()來模擬卷積過程,結果-0.639065589142 與之前的out結果的(0,.) = -0.6390 相同, 即conv2d 是通過利用4組filters,每組filter對每個通道分別卷積相加得到結果。

f_p=list(x.parameters())[0]
f_p=f_p.data.numpy()
print(f_p[0].sum()+(-0.1367))

-0.639065589142

再更新

import torch
from torch.autograd import Variable
input=torch.ones(1,groups=1)
out=x(input)

f_p=list(x.parameters())[0]
f_p=f_p.data.numpy()
f_b=list(x.parameters())[1]
f_b=f_b.data.numpy()

print("output result is:",out[0][0])
print("the result of first channel in image:",f_p[0].sum()+f_b[0])

output result is: Variable containing:
0.6577 0.6577 0.6577
0.6577 0.6577 0.6577
0.6577 0.6577 0.6577
[torch.FloatTensor of size 3x3]

the result of first channel in image: 0.657724

input=torch.ones(1,5)
input=Variable(input)
print(input.size())
x=torch.nn.Conv2d(in_channels=3,groups=1)
out=x(input)

f_p=list(x.parameters())[0]
f_b=list(x.parameters())[1]
f_p=f_p.data.numpy()
f_b=f_b.data.numpy()
# print(f_p[...,0])
# print(f_p[...,0].shape)
# print(f_p[...,0].sum()+f_b[0])
print("output result :",out[0][0])
print("simlatuate the result:",f_p[0].sum()+f_b[0])

torch.Size([1,5])
output result : Variable containing:
-0.2087 -0.2087 -0.2087
-0.2087 -0.2087 -0.2087
-0.2087 -0.2087 -0.2087
[torch.FloatTensor of size 3x3]

simlatuate the result: -0.208715

以上這篇Pytorch.nn.conv2d 過程驗證方式(單,多通道卷積過程)就是小編分享給大家的全部內容了,希望能給大家一個參考,也希望大家多多支援我們。