關(guān)于pytorch中全連接神經(jīng)網(wǎng)絡(luò)搭建兩種模式詳解

發(fā)布時(shí)間：2020-09-15 16:46:59 來(lái)源：腳本之家閱讀：193 作者：geter_CS 欄目：開發(fā)技術(shù)

pytorch搭建神經(jīng)網(wǎng)絡(luò)是很簡(jiǎn)單明了的，這里介紹兩種自己常用的搭建模式：

import torch
import torch.nn as nn

first：

class NN(nn.Module):
 def __init__(self):
  super(NN,self).__init__()
  self.model=nn.Sequential(
   nn.Linear(30,40),
   nn.ReLU(),
   nn.Linear(40,60),
   nn.Tanh(),
   nn.Linear(60,10),
   nn.Softmax()
  )
  self.model[0].weight.data.uniform_(-3e-3, 3e-3)
  self.model[0].bias.data.uniform(-1,1)
 def forward(self,states):
  return self.model(states)

這一種是將整個(gè)網(wǎng)絡(luò)寫在一個(gè)Sequential中，網(wǎng)絡(luò)參數(shù)設(shè)置可以在網(wǎng)絡(luò)搭建好后單獨(dú)設(shè)置：self.model[0].weight.data.uniform_(-3e-3,3e-3)，這是設(shè)置第一個(gè)linear的權(quán)重是（-3e-3，3e-3）之間的均勻分布,bias是-1至1之間的均勻分布。

second：

class NN1(nn.Module):
 def __init__(self):
  super(NN1,self).__init__()
  self.Linear1=nn.Linear(30,40)
  self.Linear1.weight.data.fill_(-0.1)
  #self.Linear1.weight.data.uniform_(-3e-3,3e-3)
  self.Linear1.bias.data.fill_(-0.1)
  self.layer1=nn.Sequential(self.Linear1,nn.ReLU())

  self.Linear2=nn.Linear(40,60)
  self.layer2=nn.Sequential(self.Linear2,nn.Tanh())

  self.Linear3=nn.Linear(60,10)
  self.layer3=nn.Sequential(self.Linear3,nn.Softmax())


 def forward(self,states):
  return self.model(states)

網(wǎng)絡(luò)參數(shù)的設(shè)置可以在定義完線性層之后直接設(shè)置如這里對(duì)于第一個(gè)線性層是這樣設(shè)置：self.Linear1.weight.data.fill_(-0.1),self.Linear1.bias.data.fill_(-0.1)。

你可以看一下這樣定義完的參數(shù)的效果：

Net=NN()
print("0:",Net.model[0])
print("weight:",type(Net.model[0].weight))
print("weight:",type(Net.model[0].weight.data))
print("bias",Net.model[0].bias.data)
print('1:',Net.model[1])
#print("weight:",Net.model[1].weight.data)
print('2:',Net.model[2])
print('3:',Net.model[3])
#print(Net.model[-1])

Net1=NN1()

print(Net1.Linear1.weight.data)

輸出：

0: Linear (30 -> 40)
weight: <class 'torch.nn.parameter.Parameter'>
weight: <class 'torch.FloatTensor'>
bias 
-0.6287
-0.6573
-0.0452
 0.9594
-0.7477
 0.1363
-0.1594
-0.1586
 0.0360
 0.7375
 0.2501
-0.1371
 0.8359
-0.9684
-0.3886
 0.7200
-0.3906
 0.4911
 0.8081
-0.5449
 0.9872
 0.2004
 0.0969
-0.9712
 0.0873
 0.4562
-0.4857
-0.6013
 0.1651
 0.3315
-0.7033
-0.7440
 0.6487
 0.9802
-0.5977
 0.3245
 0.7563
 0.5596
 0.2303
-0.3836
[torch.FloatTensor of size 40]

1: ReLU ()
2: Linear (40 -> 60)
3: Tanh ()

-0.1000 -0.1000 -0.1000 ... -0.1000 -0.1000 -0.1000
-0.1000 -0.1000 -0.1000 ... -0.1000 -0.1000 -0.1000
-0.1000 -0.1000 -0.1000 ... -0.1000 -0.1000 -0.1000
   ...    ⋱    ...   
-0.1000 -0.1000 -0.1000 ... -0.1000 -0.1000 -0.1000
-0.1000 -0.1000 -0.1000 ... -0.1000 -0.1000 -0.1000
-0.1000 -0.1000 -0.1000 ... -0.1000 -0.1000 -0.1000
[torch.FloatTensor of size 40x30]


Process finished with exit code 0

這里要注意self.Linear1.weight的類型是網(wǎng)絡(luò)的parameter。而self.Linear1.weight.data是FloatTensor。

以上這篇關(guān)于pytorch中全連接神經(jīng)網(wǎng)絡(luò)搭建兩種模式詳解就是小編分享給大家的全部?jī)?nèi)容了，希望能給大家一個(gè)參考，也希望大家多多支持億速云。

向AI問(wèn)一下細(xì)節(jié)

關(guān)于pytorch中全連接神經(jīng)網(wǎng)絡(luò)搭建兩種模式詳解

猜你喜歡

最新資訊

相關(guān)推薦

相關(guān)標(biāo)簽