Skip to content

Latest commit

 

History

History
320 lines (272 loc) · 31.4 KB

training-a-classifier.md

File metadata and controls

320 lines (272 loc) · 31.4 KB

你已经了解了如何定义神经网络，计算损失值和网络里权重的更新。

现在你也许会想应该怎么处理数据？

通常来说，当你处理图像，文本，语音或者视频数据时，你可以使用标准 python 包将数据加载成 numpy 数组格式，然后将这个数组转换成 torch.*Tensor
  • 对于图像，可以用 Pillow，OpenCV
  • 对于语音，可以用 scipy，librosa
  • 对于文本，可以直接用 Python 或 Cython 基础数据加载模块，或者用 NLTK 和 SpaCy
特别是对于视觉，我们已经创建了一个叫做 totchvision 的包，该包含有支持加载类似Imagenet，CIFAR10，MNIST 等公共数据集的数据加载模块 torchvision.datasets 和支持加载图像数据数据转换模块 torch.utils.data.DataLoader。

这提供了极大的便利，并且避免了编写“样板代码”。

对于本教程，我们将使用CIFAR10数据集，它包含十个类别：‘airplane’, ‘automobile’, ‘bird’, ‘cat’, ‘deer’, ‘dog’, ‘frog’, ‘horse’, ‘ship’, ‘truck’。CIFAR-10 中的图像尺寸为33232，也就是RGB的3层颜色通道，每层通道内的尺寸为32*32。

训练一个图像分类器

我们将按次序的做如下几步：
  1. 使用torchvision加载并且归一化CIFAR10的训练和测试数据集
  2. 定义一个卷积神经网络
  3. 定义一个损失函数
  4. 在训练样本数据上训练网络
  5. 在测试样本数据上测试网络
加载并归一化 CIFAR10 使用 torchvision ,用它来加载 CIFAR10 数据非常简单。 
import torch
import torchvision
import torchvision.transforms as transforms
torchvision 数据集的输出是范围在[0,1]之间的 PILImage，我们将他们转换成归一化范围为[-1,1]之间的张量 Tensors。 
transform = transforms.Compose(
    [transforms.ToTensor(),
     transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])

trainset = torchvision.datasets.CIFAR10(root='./data', train=True,
download=True, transform=transform)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=4,
shuffle=True, num_workers=2)


testset = torchvision.datasets.CIFAR10(root='./data', train=False,
download=True, transform=transform)
testloader = torch.utils.data.DataLoader(testset, batch_size=4,
shuffle=False, num_workers=2)


classes = ('plane', 'car', 'bird', 'cat',
'deer', 'dog', 'frog', 'horse', 'ship', 'truck')
输出：

Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz
Files already downloaded and verified

让我们来展示其中的一些训练图片。

import matplotlib.pyplot as plt
import numpy as np

# functions to show an image


def imshow(img):
    img = img / 2 + 0.5     # unnormalize
    npimg = img.numpy()
    plt.imshow(np.transpose(npimg, (1, 2, 0)))
    plt.show()


# get some random training images
dataiter = iter(trainloader)
images, labels = dataiter.next()

# show images
imshow(torchvision.utils.make_grid(images))
# print labels
print(' '.join('%5s' % classes[labels[j]] for j in range(4)))

 

输出：

cat plane  ship  frog
定义一个卷积神经网络 在这之前先 从神经网络章节 复制神经网络，并修改它为3通道的图片(在此之前它被定义为1通道)
import torch.nn as nn
import torch.nn.functional as F

class Net(nn.Module):
def init(self):
super(Net, self).init()
self.conv1 = nn.Conv2d(3, 6, 5)
self.pool = nn.MaxPool2d(2, 2)
self.conv2 = nn.Conv2d(6, 16, 5)
self.fc1 = nn.Linear(16  5  5, 120)
self.fc2 = nn.Linear(120, 84)
self.fc3 = nn.Linear(84, 10)


<span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
    <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">pool</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)))</span>
    <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">pool</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv2</span><span class="p">(</span><span class="n">x</span><span class="p">)))</span>
    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">16</span> <span class="o">*</span> <span class="mi">5</span> <span class="o">*</span> <span class="mi">5</span><span class="p">)</span>
    <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">fc1</span><span class="p">(</span><span class="n">x</span><span class="p">))</span>
    <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">fc2</span><span class="p">(</span><span class="n">x</span><span class="p">))</span>
    <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
    <span class="k">return</span> <span class="n">x</span>



net = Net()

 

定义一个损失函数和优化器 让我们使用分类交叉熵Cross-Entropy 作损失函数，动量SGD做优化器。
import torch.optim as optim

criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)

训练网络 这里事情开始变得有趣，我们只需要在数据迭代器上循环传给网络和优化器 输入就可以。
for epoch in range(2):  # loop over the dataset multiple times

<span class="n">running_loss</span> <span class="o">=</span> <span class="mf">0.0</span>
<span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">data</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">trainloader</span><span class="p">,</span> <span class="mi">0</span><span class="p">):</span>
    <span class="c1"># get the inputs</span>
    <span class="n">inputs</span><span class="p">,</span> <span class="n">labels</span> <span class="o">=</span> <span class="n">data</span>

    <span class="c1"># zero the parameter gradients</span>
    <span class="n">optimizer</span><span class="o">.</span><span class="n">zero_grad</span><span class="p">()</span>

    <span class="c1"># forward + backward + optimize</span>
    <span class="n">outputs</span> <span class="o">=</span> <span class="n">net</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
    <span class="n">loss</span> <span class="o">=</span> <span class="n">criterion</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="n">labels</span><span class="p">)</span>
    <span class="n">loss</span><span class="o">.</span><span class="n">backward</span><span class="p">()</span>
    <span class="n">optimizer</span><span class="o">.</span><span class="n">step</span><span class="p">()</span>

    <span class="c1"># print statistics</span>
    <span class="n">running_loss</span> <span class="o">+=</span> <span class="n">loss</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
    <span class="k">if</span> <span class="n">i</span> <span class="o">%</span> <span class="mi">2000</span> <span class="o">==</span> <span class="mi">1999</span><span class="p">:</span>    <span class="c1"># print every 2000 mini-batches</span>
        <span class="k">print</span><span class="p">(</span><span class="s1">'[</span><span class="si">%d</span><span class="s1">, </span><span class="si">%5d</span><span class="s1">] loss: </span><span class="si">%.3f</span><span class="s1">'</span> <span class="o">%</span>
              <span class="p">(</span><span class="n">epoch</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">running_loss</span> <span class="o">/</span> <span class="mi">2000</span><span class="p">))</span>
        <span class="n">running_loss</span> <span class="o">=</span> <span class="mf">0.0</span>



print('Finished Training')

 输出：
[1,  2000] loss: 2.187
[1,  4000] loss: 1.852
[1,  6000] loss: 1.672
[1,  8000] loss: 1.566
[1, 10000] loss: 1.490
[1, 12000] loss: 1.461
[2,  2000] loss: 1.389
[2,  4000] loss: 1.364
[2,  6000] loss: 1.343
[2,  8000] loss: 1.318
[2, 10000] loss: 1.282
[2, 12000] loss: 1.286
Finished Training
在测试集上测试网络 我们已经通过训练数据集对网络进行了2次训练，但是我们需要检查网络是否已经学到了东西。

我们将用神经网络的输出作为预测的类标来检查网络的预测性能，用样本的真实类标来校对。如果预测是正确的，我们将样本添加到正确预测的列表里。

好的，第一步，让我们从测试集中显示一张图像来熟悉它。

输出：

GroundTruth:    cat  ship  ship plane
现在让我们看看 神经网络认为这些样本应该预测成什么： 
outputs = net(images)
输出是预测与十个类的近似程度，与某一个类的近似程度越高，网络就越认为图像是属于这一类别。所以让我们打印其中最相似类别类标： 
_, predicted = torch.max(outputs, 1)

print('Predicted: ', ' '.join('%5s' % classes[predicted[j]]
for j in range(4)))
输出：

Predicted:    cat  ship   car  ship
结果看起开非常好，让我们看看网络在整个数据集上的表现。 
correct = 0
total = 0
with torch.no_grad():
    for data in testloader:
        images, labels = data
        outputs = net(images)
        _, predicted = torch.max(outputs.data, 1)
        total += labels.size(0)
        correct += (predicted == labels).sum().item()

print('Accuracy of the network on the 10000 test images: %d %%' % (
100 * correct / total))
输出：

Accuracy of the network on the 10000 test images: 54 %
这看起来比随机预测要好，随机预测的准确率为10%（随机预测出为10类中的哪一类）。看来网络学到了东西。 
class_correct = list(0. for i in range(10))
class_total = list(0. for i in range(10))
with torch.no_grad():
    for data in testloader:
        images, labels = data
        outputs = net(images)
        _, predicted = torch.max(outputs, 1)
        c = (predicted == labels).squeeze()
        for i in range(4):
            label = labels[i]
            class_correct[label] += c[i].item()
            class_total[label] += 1

for i in range(10):
print('Accuracy of %5s : %2d %%' % (
classes[i], 100 * class_correct[i] / class_total[i]))
输出：

Accuracy of plane : 57 %
Accuracy of   car : 73 %
Accuracy of  bird : 49 %
Accuracy of   cat : 54 %
Accuracy of  deer : 18 %
Accuracy of   dog : 20 %
Accuracy of  frog : 58 %
Accuracy of horse : 74 %
Accuracy of  ship : 70 %
Accuracy of truck : 66 %

所以接下来呢？

我们怎么在GPU上跑这些神经网络？

在GPU上训练 就像你怎么把一个张量转移到GPU上一样，你要将神经网络转到GPU上。 如果CUDA可以用，让我们首先定义下我们的设备为第一个可见的cuda设备。

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")

# Assume that we are on a CUDA machine, then this should print a CUDA device:

print(device)

输出：

cuda:0
本节剩余部分都会假定设备就是台CUDA设备。

接着这些方法会递归地遍历所有模块，并将它们的参数和缓冲器转换为CUDA张量。

net.to(device)
记住你也必须在每一个步骤向GPU发送输入和目标： 
inputs, labels = inputs.to(device), labels.to(device)
为什么没有注意到与CPU相比巨大的加速？因为你的网络非常小。

 

练习：尝试增加你的网络宽度（首个 nn.Conv2d 参数设定为 2，第二个nn.Conv2d参数设定为1--它们需要有相同的个数），看看会得到怎么的速度提升。

目标：

  • 深度理解了PyTorch的张量和神经网络
  • 训练了一个小的神经网络来分类图像
在多个GPU上训练

如果你想要来看到大规模加速，使用你的所有GPU，请查看：数据并行性（https://pytorch.org/tutorials/beginner/blitz/data_parallel_tutorial.html）。PyTorch 60 分钟入门教程：数据并行处理

http://pytorchchina.com/2018/12/11/optional-data-parallelism/

下载 Python 源代码：

cifar10_tutorial.py

下载 Jupyter 源代码：

cifar10_tutorial.ipynb