快捷導(dǎo)航

PyTorch上實(shí)現(xiàn)卷積神經(jīng)網(wǎng)絡(luò)CNN的方法

更新時(shí)間：2018年04月28日 09:10:06 作者：https://blog.csdn.net/marsjhao/article/details/72179517

本篇文章主要介紹了PyTorch上實(shí)現(xiàn)卷積神經(jīng)網(wǎng)絡(luò)CNN的方法，小編覺(jué)得挺不錯(cuò)的，現(xiàn)在分享給大家，也給大家做個(gè)參考。一起跟隨小編過(guò)來(lái)看看吧

一、卷積神經(jīng)網(wǎng)絡(luò)

卷積神經(jīng)網(wǎng)絡(luò)（ConvolutionalNeuralNetwork，CNN）最初是為解決圖像識(shí)別等問(wèn)題設(shè)計(jì)的，CNN現(xiàn)在的應(yīng)用已經(jīng)不限于圖像和視頻，也可用于時(shí)間序列信號(hào)，比如音頻信號(hào)和文本數(shù)據(jù)等。CNN作為一個(gè)深度學(xué)習(xí)架構(gòu)被提出的最初訴求是降低對(duì)圖像數(shù)據(jù)預(yù)處理的要求，避免復(fù)雜的特征工程。在卷積神經(jīng)網(wǎng)絡(luò)中，第一個(gè)卷積層會(huì)直接接受圖像像素級(jí)的輸入，每一層卷積（濾波器）都會(huì)提取數(shù)據(jù)中最有效的特征，這種方法可以提取到圖像中最基礎(chǔ)的特征，而后再進(jìn)行組合和抽象形成更高階的特征，因此CNN在理論上具有對(duì)圖像縮放、平移和旋轉(zhuǎn)的不變性。

卷積神經(jīng)網(wǎng)絡(luò)CNN的要點(diǎn)就是局部連接（LocalConnection）、權(quán)值共享（WeightsSharing）和池化層（Pooling）中的降采樣（Down-Sampling）。其中，局部連接和權(quán)值共享降低了參數(shù)量，使訓(xùn)練復(fù)雜度大大下降并減輕了過(guò)擬合。同時(shí)權(quán)值共享還賦予了卷積網(wǎng)絡(luò)對(duì)平移的容忍性，池化層降采樣則進(jìn)一步降低了輸出參數(shù)量并賦予模型對(duì)輕度形變的容忍性，提高了模型的泛化能力?？梢园丫矸e層卷積操作理解為用少量參數(shù)在圖像的多個(gè)位置上提取相似特征的過(guò)程。

二、代碼實(shí)現(xiàn)

import torch 
import torch.nn as nn 
from torch.autograd import Variable 
import torch.utils.data as Data 
import torchvision 
import matplotlib.pyplot as plt 
 
torch.manual_seed(1) 
 
EPOCH = 1 
BATCH_SIZE = 50 
LR = 0.001 
DOWNLOAD_MNIST = True 
 
# 獲取訓(xùn)練集dataset 
training_data = torchvision.datasets.MNIST( 
       root='./mnist/', # dataset存儲(chǔ)路徑 
       train=True, # True表示是train訓(xùn)練集，F(xiàn)alse表示test測(cè)試集 
       transform=torchvision.transforms.ToTensor(), # 將原數(shù)據(jù)規(guī)范化到（0,1）區(qū)間 
       download=DOWNLOAD_MNIST, 
       ) 
 
# 打印MNIST數(shù)據(jù)集的訓(xùn)練集及測(cè)試集的尺寸 
print(training_data.train_data.size()) 
print(training_data.train_labels.size()) 
# torch.Size([60000, 28, 28]) 
# torch.Size([60000]) 
 
plt.imshow(training_data.train_data[0].numpy(), cmap='gray') 
plt.title('%i' % training_data.train_labels[0]) 
plt.show() 
 
# 通過(guò)torchvision.datasets獲取的dataset格式可直接可置于DataLoader 
train_loader = Data.DataLoader(dataset=training_data, batch_size=BATCH_SIZE, 
                shuffle=True) 
 
# 獲取測(cè)試集dataset 
test_data = torchvision.datasets.MNIST(root='./mnist/', train=False) 
# 取前2000個(gè)測(cè)試集樣本 
test_x = Variable(torch.unsqueeze(test_data.test_data, dim=1), 
         volatile=True).type(torch.FloatTensor)[:2000]/255 
# (2000, 28, 28) to (2000, 1, 28, 28), in range(0,1) 
test_y = test_data.test_labels[:2000] 
 
class CNN(nn.Module): 
  def __init__(self): 
    super(CNN, self).__init__() 
    self.conv1 = nn.Sequential( # (1,28,28) 
           nn.Conv2d(in_channels=1, out_channels=16, kernel_size=5, 
                stride=1, padding=2), # (16,28,28) 
    # 想要con2d卷積出來(lái)的圖片尺寸沒(méi)有變化, padding=(kernel_size-1)/2 
           nn.ReLU(), 
           nn.MaxPool2d(kernel_size=2) # (16,14,14) 
           ) 
    self.conv2 = nn.Sequential( # (16,14,14) 
           nn.Conv2d(16, 32, 5, 1, 2), # (32,14,14) 
           nn.ReLU(), 
           nn.MaxPool2d(2) # (32,7,7) 
           ) 
    self.out = nn.Linear(32*7*7, 10) 
 
  def forward(self, x): 
    x = self.conv1(x) 
    x = self.conv2(x) 
    x = x.view(x.size(0), -1) # 將（batch，32,7,7）展平為（batch，32*7*7） 
    output = self.out(x) 
    return output 
 
cnn = CNN() 
print(cnn) 
''''' 
CNN ( 
 (conv1): Sequential ( 
  (0): Conv2d(1, 16, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2)) 
  (1): ReLU () 
  (2): MaxPool2d (size=(2, 2), stride=(2, 2), dilation=(1, 1)) 
 ) 
 (conv2): Sequential ( 
  (0): Conv2d(16, 32, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2)) 
  (1): ReLU () 
  (2): MaxPool2d (size=(2, 2), stride=(2, 2), dilation=(1, 1)) 
 ) 
 (out): Linear (1568 -> 10) 
) 
''' 
optimizer = torch.optim.Adam(cnn.parameters(), lr=LR) 
loss_function = nn.CrossEntropyLoss() 
 
for epoch in range(EPOCH): 
  for step, (x, y) in enumerate(train_loader): 
    b_x = Variable(x) 
    b_y = Variable(y) 
 
    output = cnn(b_x) 
    loss = loss_function(output, b_y) 
    optimizer.zero_grad() 
    loss.backward() 
    optimizer.step() 
 
    if step % 100 == 0: 
      test_output = cnn(test_x) 
      pred_y = torch.max(test_output, 1)[1].data.squeeze() 
      accuracy = sum(pred_y == test_y) / test_y.size(0) 
      print('Epoch:', epoch, '|Step:', step, 
         '|train loss:%.4f'%loss.data[0], '|test accuracy:%.4f'%accuracy) 
 
test_output = cnn(test_x[:10]) 
pred_y = torch.max(test_output, 1)[1].data.numpy().squeeze() 
print(pred_y, 'prediction number') 
print(test_y[:10].numpy(), 'real number') 
''''' 
Epoch: 0 |Step: 0 |train loss:2.3145 |test accuracy:0.1040 
Epoch: 0 |Step: 100 |train loss:0.5857 |test accuracy:0.8865 
Epoch: 0 |Step: 200 |train loss:0.0600 |test accuracy:0.9380 
Epoch: 0 |Step: 300 |train loss:0.0996 |test accuracy:0.9345 
Epoch: 0 |Step: 400 |train loss:0.0381 |test accuracy:0.9645 
Epoch: 0 |Step: 500 |train loss:0.0266 |test accuracy:0.9620 
Epoch: 0 |Step: 600 |train loss:0.0973 |test accuracy:0.9685 
Epoch: 0 |Step: 700 |train loss:0.0421 |test accuracy:0.9725 
Epoch: 0 |Step: 800 |train loss:0.0654 |test accuracy:0.9710 
Epoch: 0 |Step: 900 |train loss:0.1333 |test accuracy:0.9740 
Epoch: 0 |Step: 1000 |train loss:0.0289 |test accuracy:0.9720 
Epoch: 0 |Step: 1100 |train loss:0.0429 |test accuracy:0.9770 
[7 2 1 0 4 1 4 9 5 9] prediction number 
[7 2 1 0 4 1 4 9 5 9] real number 
'''

三、分析解讀

通過(guò)利用torchvision.datasets可以快速獲取可以直接置于DataLoader中的dataset格式的數(shù)據(jù)，通過(guò)train參數(shù)控制是獲取訓(xùn)練數(shù)據(jù)集還是測(cè)試數(shù)據(jù)集，也可以在獲取的時(shí)候便直接轉(zhuǎn)換成訓(xùn)練所需的數(shù)據(jù)格式。

卷積神經(jīng)網(wǎng)絡(luò)的搭建通過(guò)定義一個(gè)CNN類來(lái)實(shí)現(xiàn)，卷積層conv1，conv2及out層以類屬性的形式定義，各層之間的銜接信息在forward中定義，定義的時(shí)候要留意各層的神經(jīng)元數(shù)量。

CNN的網(wǎng)絡(luò)結(jié)構(gòu)如下：

CNN (

 (conv1): Sequential (

  (0): Conv2d(1, 16,kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))

  (1): ReLU ()

  (2): MaxPool2d (size=(2,2), stride=(2, 2), dilation=(1, 1))

 )

 (conv2): Sequential (

  (0): Conv2d(16, 32,kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))

  (1): ReLU ()

  (2): MaxPool2d (size=(2,2), stride=(2, 2), dilation=(1, 1))

 )

 (out): Linear (1568 ->10)

)

經(jīng)過(guò)實(shí)驗(yàn)可見(jiàn)，在EPOCH=1的訓(xùn)練結(jié)果中，測(cè)試集準(zhǔn)確率可達(dá)到97.7%。

以上就是本文的全部?jī)?nèi)容，希望對(duì)大家的學(xué)習(xí)有所幫助，也希望大家多多支持腳本之家。

您可能感興趣的文章: