基于Pytorch的简单小案例

　　神经网络的理论知识不是本文讨论的重点，假设读者们都是已经了解RNN的基本概念，并希望能用一些框架做一些简单的实现。这里推荐神经网络必读书目：邱锡鹏《神经网络与深度学习》。本文基于Pytorch简单实现CIFAR-10、MNIST手写体识别，读者可以基于此两个简单案例进行拓展，实现自己的深度学习入门。

环境说明

　　python 3.6.7

　　Pytorch的CUP版本

　　Pycharm编辑器

　　部分可能报错：参见pytorch安装错误及解决

基于Pytorch的CIFAR-10图片分类

代码实现

# coding = utf-8

import torch
import torch.nn
import numpy as np
from torchvision.datasets import CIFAR10
from torchvision import transforms
from torch.utils.data import DataLoader
from torch.utils.data.sampler import SubsetRandomSampler
import torch.nn.functional as F
import torch.optim as optimizer

'''
The compose function allows for multiple transforms.
transform.ToTensor() converts our PILImage to a tensor of
shape (C x H x W) in the range [0, 1]
transform.Normalize(mean, std) normalizes a tensor to a (mean, std)
for (R, G, B)
'''
_task = transforms.Compose([
    transforms.ToTensor(),
    transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])
])

# 注意：此处数据集在本地，因此download=False;若需要下载的改为True
# 同样的，第一个参数为数据存放路径
data_path = '../CIFAR_10_zhuanzhi/cifar10'
cifar = CIFAR10(data_path, train=True, download=False, transform=_task)

# 这里只是为了构造取样的角标，可根据自己的思路进行拓展
# 此处使用了前百分之八十作为训练集，百分之八十到九十的作为验证集，后百分之十为测试集
samples_count = len(cifar)
split_train = int(0.8 * samples_count)
split_valid = int(0.9 * samples_count)

index_list = list(range(samples_count))
train_idx, valid_idx, test_idx = index_list[:split_train], index_list[split_train:split_valid], index_list[split_valid:]

# 定义采样器
# create training and validation, test sampler
train_sampler = SubsetRandomSampler(train_idx)
valid_sampler = SubsetRandomSampler(valid_idx)
test_samlper  = SubsetRandomSampler(test_idx )

# create iterator for train and valid, test dataset
trainloader = DataLoader(cifar, batch_size=256, sampler=train_sampler)
validloader = DataLoader(cifar, batch_size=256, sampler=valid_sampler)
testloader  = DataLoader(cifar, batch_size=256, sampler=test_samlper )

# 网络设计
class Net(torch.nn.Module):
    """
    网络设计了三个卷积层，一个池化层，一个全连接层
    """
    def __init__(self):
        super(Net, self).__init__()

        self.conv1 = torch.nn.Conv2d(3, 16, 3, padding=1)
        self.conv2 = torch.nn.Conv2d(16, 32, 3, padding=1)
        self.conv3 = torch.nn.Conv2d(32, 64, 3, padding=1)
        self.pool = torch.nn.MaxPool2d(2, 2)
        self.linear1 = torch.nn.Linear(1024, 512)
        self.linear2 = torch.nn.Linear(512, 10)

    # 前向传播
    def forward(self, x):
        x = self.pool(F.relu(self.conv1(x)))
        x = self.pool(F.relu(self.conv2(x)))
        x = self.pool(F.relu(self.conv3(x)))
        x = x.view(-1, 1024)
        x = F.relu(self.linear1(x))
        x = F.relu(self.linear2(x))

        return x

if __name__ == "__main__":

    net = Net()  # 实例化网络
    loss_function = torch.nn.CrossEntropyLoss()  # 定义交叉熵损失

    # 定义优化算法
    optimizer = optimizer.SGD(net.parameters(), lr=0.01, weight_decay=1e-6, momentum=0.9, nesterov=True)

    # 迭代次数
    for epoch in range(1, 31):
        train_loss, valid_loss = [], []

        net.train()  # 训练开始
        for data, target in trainloader:
            optimizer.zero_grad()  # 梯度置0
            output = net(data)
            loss = loss_function(output, target)  # 计算损失
            loss.backward()   # 反向传播
            optimizer.step()  # 更新参数
            train_loss.append(loss.item())

        net.eval()  # 验证开始
        for data, target in validloader:
            output = net(data)
            loss = loss_function(output, target)
            valid_loss.append(loss.item())

        print("Epoch:{}, Training Loss:{}, Valid Loss:{}".format(epoch, np.mean(train_loss), np.mean(valid_loss)))
    print("======= Training Finished ! =========")

    print("Testing Begining ... ")  # 模型测试
    total = 0
    correct = 0
    for i, data_tuple in enumerate(testloader, 0):

        data, labels = data_tuple
        output = net(data)
        _, preds_tensor = torch.max(output, 1)

        total += labels.size(0)
        correct += np.squeeze((preds_tensor == labels).sum().numpy())
    print("Accuracy : {} %".format(correct/total))

实验结果

经验总结

1.激活函数的选择。

• 激活函数可选择sigmoid函数或者Relu函数，亲测使用Relu函数后，分类的正确率会高使用sigmoid函数很多；
• Relu函数的导入有两种：import torch.nn.functional as F, 然后F.relu()，还有一种是torch.nn.Relu() 两种方式实验结果没区别，但是推荐使用后者；因为前者是以函数的形式导入的，在模型保存时，F中相关参数会被释放，无法保存下去，而后者会保留参数。

2.预测结果的处理。

　　Pytorch预测的结果，返回的是一个Tensor,需要处理成数值才能进行准确率计算，.numpy()方法能将Tensor转化为数组，然后使用squeeze能够将数组转化为数值。

3. 数据加载。Pytorch是采用批量加载数据的，因此使用for循环迭代从采样器中加载数据，batch_size参数指定每次加载数据量的大小

4.注意维度。

• 网络设计中的维度。网络层次设计中，要谨记前一层的输出是后一层的输入，维度要对应的上。
• 全连接中的维度。全连接中要从特征图中选取特征，这些特征不是一维的，而全连接输出的结果是一维的，因此从特征图中选取特征作为全连接层输入前，需要将特征展开，例如：x = x.view(-1, 28*28)

基于Pytorch的MNIST手写体识别

代码实现

# coding = utf-8
import numpy as np
import torch
from torchvision import transforms

_task = transforms.Compose([
    transforms.ToTensor(),
    transforms.Normalize(
        [0.5], [0.5]
    )
])

from torchvision.datasets import MNIST

# 数据集加载
mnist = MNIST('./data', download=False, train=True, transform=_task)

# 训练集和验证集划分
from torch.utils.data import DataLoader
from torch.utils.data.sampler import SubsetRandomSampler

# create training and validation split
index_list = list(range(len(mnist)))

split_train = int(0.8*len(mnist))
split_valid = int(0.9*len(mnist))

train_idx, valid_idx, test_idx = index_list[:split_train], index_list[split_train:split_valid], index_list[split_valid:]

# create sampler objects using SubsetRandomSampler
train_sampler = SubsetRandomSampler(train_idx)
valid_sampler = SubsetRandomSampler(valid_idx)
test_sampler = SubsetRandomSampler(test_idx)

# create iterator objects for train and valid dataset
trainloader = DataLoader(mnist, batch_size=256, sampler=train_sampler)
validloader = DataLoader(mnist, batch_size=256, sampler=valid_sampler)
test_loader = DataLoader(mnist, batch_size=256, sampler=test_sampler )

# design for net
import torch.nn.functional as F
class NetModel(torch.nn.Module):
    def __init__(self):
        super(NetModel, self).__init__()
        self.hidden = torch.nn.Linear(28*28, 300)
        self.output = torch.nn.Linear(300, 10)

    def forward(self, x):
        x = x.view(-1, 28*28)
        x = self.hidden(x)
        x = F.relu(x)
        x = self.output(x)
        return x

if __name__ == "__main__":
    net = NetModel()

    from torch import optim
    loss_function = torch.nn.CrossEntropyLoss()
    optimizer = optim.SGD(net.parameters(), lr=0.01, weight_decay=1e-6, momentum=0.9, nesterov=True)

    for epoch in range(1, 12):
        train_loss, valid_loss = [], []
        # net.train()
        for data, target in trainloader:
            optimizer.zero_grad()
            # forward propagation
            output = net(data)
            loss = loss_function(output, target)
            loss.backward()
            optimizer.step()
            train_loss.append(loss.item())
        # net.eval()
        for data, target in validloader:
            output = net(data)
            loss = loss_function(output, target)
            valid_loss.append(loss.item())
        print("Epoch:", epoch, "Training Loss:", np.mean(train_loss), "Valid Loss:", np.mean(valid_loss))

    print("testing ... ")
    total = 0
    correct = 0
    for i, test_data in enumerate(test_loader, 0):
        data, label = test_data
        output = net(data)
        _, predict = torch.max(output.data, 1)

        total += label.size(0)
        correct += np.squeeze((predict == label).sum().numpy())
    print("Accuracy:", (correct/total)*100, "%")

实验结果

经验总结

　　1.网络设计的使用只用了一个隐层，单隐层神经网络经过10词迭代，对手写体识别准确率高达97%！！简直变态啊！

　　2.loss.item()和loss.data[0]。好像新版本的pytorch放弃了loss.data[0]的表达方式。

　　3.手写体识别的图片是单通道图片，因此在transforms.Compose()中做标准化的时候，只需要指定一个值即可；而cifar中的图片是三通道的，因此需要指定三个参数。