?卷積神經(jīng)網(wǎng)絡(luò)目前被廣泛地用在圖片識(shí)別上, 已經(jīng)有層出不窮的應(yīng)用.

更多可以查看官網(wǎng) :
* PyTorch 官網(wǎng)

MNIST手寫數(shù)據(jù)

import torch
import torch.nn as nn
from torch.autograd import Variable
import torch.utils.data as Data
import torchvision      # 數(shù)據(jù)庫模塊
import matplotlib.pyplot as plt

torch.manual_seed(1)    # reproducible

# Hyper Parameters
EPOCH = 1           # 訓(xùn)練整批數(shù)據(jù)多少次, 為了節(jié)約時(shí)間, 只訓(xùn)練一次
BATCH_SIZE = 50
LR = 0.001          # 學(xué)習(xí)率
DOWNLOAD_MNIST = True  # 如果你已經(jīng)下載好了mnist數(shù)據(jù)就寫上 Fasle


# Mnist 手寫數(shù)字
train_data = torchvision.datasets.MNIST(
    root='./mnist/',    # 保存或者提取位置
    train=True,  # this is training data
    transform=torchvision.transforms.ToTensor(),    # 轉(zhuǎn)換 PIL.Image or numpy.ndarray 成
                                                    # torch.FloatTensor (C x H x W), 訓(xùn)練的時(shí)候 normalize 成 [0.0, 1.0] 區(qū)間
    download=DOWNLOAD_MNIST,          # 沒下載就下載, 下載了就不用再下了
)

還是來自MNIST數(shù)據(jù)集的示例圖像

每個(gè)圖中黑色的地方的值都是0, 白色的地方值大于0.

同樣, 我們除了訓(xùn)練數(shù)據(jù), 還給一些測(cè)試數(shù)據(jù), 測(cè)試看看它有沒有訓(xùn)練好.

test_data = torchvision.datasets.MNIST(root='./mnist/', train=False)

# 批訓(xùn)練 50samples, 1 channel, 28x28 (50, 1, 28, 28)
train_loader = Data.DataLoader(dataset=train_data, batch_size=BATCH_SIZE, shuffle=True)

# 為了節(jié)約時(shí)間, 我們測(cè)試時(shí)只測(cè)試前2000個(gè)
test_x = Variable(torch.unsqueeze(test_data.test_data, dim=1), volatile=True).type(torch.FloatTensor)[:2000]/255.   # shape from (2000, 28, 28) to (2000, 1, 28, 28), value in range(0,1)
test_y = test_data.test_labels[:2000]

CNN模型

用一個(gè) class 來建立 CNN 模型. 這個(gè) CNN 整體流程是 卷積(Conv2d) ->
激勵(lì)函數(shù)(ReLU) -> 池化, 向下采樣 (MaxPooling) -> 再來一遍 -> 展平多維的卷積成的特征圖 ->
接入全連接層 (Linear) -> 輸出

class CNN(nn.Module):
    def __init__(self):
        super(CNN, self).__init__()
        self.conv1 = nn.Sequential(  # input shape (1, 28, 28)
            nn.Conv2d(
                in_channels=1,      # input height
                out_channels=16,    # n_filters
                kernel_size=5,      # filter size
                stride=1,           # filter movement/step
                padding=2,      # 如果想要 con2d 出來的圖片長寬沒有變化, padding=(kernel_size-1)/2 當(dāng) stride=1
            ),      # output shape (16, 28, 28)
            nn.ReLU(),    # activation
            nn.MaxPool2d(kernel_size=2),    # 在 2x2 空間里向下采樣, output shape (16, 14, 14)
        )
        self.conv2 = nn.Sequential(  # input shape (1, 28, 28)
            nn.Conv2d(16, 32, 5, 1, 2),  # output shape (32, 14, 14)
            nn.ReLU(),  # activation
            nn.MaxPool2d(2),  # output shape (32, 7, 7)
        )
        self.out = nn.Linear(32 * 7 * 7, 10)   # fully connected layer, output 10 classes

    def forward(self, x):
        x = self.conv1(x)
        x = self.conv2(x)
        x = x.view(x.size(0), -1)   # 展平多維的卷積圖成 (batch_size, 32 * 7 * 7)
        output = self.out(x)
        return output

cnn = CNN()
print(cnn)  # net architecture
"""
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)
)
"""

訓(xùn)練

開始訓(xùn)練, 將 x y 都用 Variable 包起來, 然后放入 cnn 中計(jì)算 output, 最后再計(jì)算誤差.

# 代碼省略了計(jì)算精確度 `accuracy` 的部分

optimizer = torch.optim.Adam(cnn.parameters(), lr=LR)   # optimize all cnn parameters
loss_func = nn.CrossEntropyLoss()   # the target label is not one-hotted

# training and testing
for epoch in range(EPOCH):
    for step, (x, y) in enumerate(train_loader):   # 分配 batch data, normalize x when iterate train_loader
        b_x = Variable(x)   # batch x
        b_y = Variable(y)   # batch y

        output = cnn(b_x)               # cnn output
        loss = loss_func(output, b_y)   # cross entropy loss
        optimizer.zero_grad()           # clear gradients for this training step
        loss.backward()                 # backpropagation, compute gradients
        optimizer.step()                # apply gradients

"""
...
Epoch:  0 | train loss: 0.0306 | test accuracy: 0.97
Epoch:  0 | train loss: 0.0147 | test accuracy: 0.98
Epoch:  0 | train loss: 0.0427 | test accuracy: 0.98
Epoch:  0 | train loss: 0.0078 | test accuracy: 0.98
"""

最后取10個(gè)數(shù)據(jù), 看看預(yù)測(cè)的值到底對(duì)不對(duì):

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')

"""
[7 2 1 0 4 1 4 9 5 9] prediction number
[7 2 1 0 4 1 4 9 5 9] real number
"""