Task1
1.什么是PyTorch,为什么选择PyTroch?
什么是PyTorch,我们先来回答“PyTorch是什么”,简单的说,PyTorch就是一个深度学习框架,而反过来,深度学习的框架可不仅仅是Pytorch,好了,废话完了。来说说什么是PyTorch吧,网上已经有很多资料对Pytorch进行的详细的描述,自身能力有限,不能够详尽的描述出其定义,我主要通过网上开源的资料进入,结合自己的一些看法,选择性的进行定义的描述,有不对的地方希望大家能够补充。 要了解一个框架就应该八卦一下它,这样才能让人记忆深刻吧。 这个文章里就扒了TensorFlow和PyTorch的区别,举例来说,PyTorch更加容易学习和调试,而且Facebook也将之前开发的Torch和Caffe等框架往上边迁移,而且近期发布的1.1版本还原生支持了Tensorboard。那么我通过添加对逆命题添加两个限定条件来给个不学术的定义,易于学习的发展趋势迅猛的深度学习框架是PyTorch。
至于为什么要用PyTorch,主要是因为越来越读论文使用PyTorch实现并开源代码,选择PyTorch,这样我们就能愉快地搬砖了。
2.PyTorch的安装
Pytroch的安装,官网提供了4种安装方式,目前仅使用过conda和pip安装 。推荐使用conda安装,它会将需要的依赖都一并安装上,同时应该注意安装之前先用conda构建虚拟环境,这样你就有多个python2或者python3,Pyroch0.4.1或者PyTorch1.0的版本了。
3.配置Python环境
这里我使用Anaconda配置python环境,这里面集成了很多工具,同时也方便版本管理。首先下载安装包,然后在对应文件夹目录打开命令行工具通过下面的命令进行安装。
bash Anaconda***.sh
4.准备Python管理器
上面我已经安装了Anaconda就可以使用里面的conda工具进行虚拟环境以及依赖包的安装。
conda create --name py3.6 python=3.6
conda activate py3.6
conda list
conda install ×××
5.通过命令行安装PyTorch
同样接着上边,进入官网,选择机子环境后获取conda安装命令。
conda install pytorch torchvision cudatoolkit=10.0 -c pytorch
6.PyTorch基础概念
根据这篇文章的介绍,主要将PyTorch分为三个部分,数据处理、模型和损失及优化器。数据处理,在cv中,PyTorch有很好的vision库,transform很方便的对图像进行翻转旋转以及归一化等操作。模型,nn模块里集成了很多SOTA的模型,简单几行代码就能够使用已经训练过的模型。损失函数以及优化器,对不同的任务会使用不同的损失进行调优,而优化的算法自己也只是照搬的Adam优化,对这一部分了解不是很深入,等后续将文章里的资料详细了解之后再回来填坑。
7.通过代码实现流程(实现一个深度学习的代码流程)
自己没有造轮子的技能,就先按照官网的教程实现了一遍Mnist的分类
import torch
import torchvision
import torchvision.transforms as transforms
import matplotlib.pyplot as plt
import numpy as np
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
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')
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)))
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)
def forward(self, x):
x = self.pool(F.relu(self.conv1(x)))
x = self.pool(F.relu(self.conv2(x)))
x = x.view(-1, 16 * 5 * 5)
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = self.fc3(x)
return x
net = Net()
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)
for epoch in range(2): # loop over the dataset multiple times
running_loss = 0.0
for i, data in enumerate(trainloader, 0):
# get the inputs
inputs, labels = data
# zero the parameter gradients
optimizer.zero_grad()
# forward + backward + optimize
outputs = net(inputs)
loss = criterion(outputs, labels)
loss.backward()
optimizer.step()
# print statistics
running_loss += loss.item()
if i % 2000 == 1999: # print every 2000 mini-batches
print('[%d, %5d] loss: %.3f' %
(epoch + 1, i + 1, running_loss / 2000))
running_loss = 0.0
print('Finished Training')
dataiter = iter(testloader)
images, labels = dataiter.next()
# print images
imshow(torchvision.utils.make_grid(images))
print('GroundTruth: ', ' '.join('%5s' % classes[labels[j]] for j in range(4)))
outputs = net(images)
_, predicted = torch.max(outputs, 1)
print('Predicted: ', ' '.join('%5s' % classes[predicted[j]]
for j in range(4)))
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))
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]))
CPU 上训练
[1, 2000] loss: 2.217
[1, 4000] loss: 1.889
[1, 6000] loss: 1.667
[1, 8000] loss: 1.562
[1, 10000] loss: 1.517
[1, 12000] loss: 1.472
[2, 2000] loss: 1.397
[2, 4000] loss: 1.373
[2, 6000] loss: 1.326
[2, 8000] loss: 1.303
[2, 10000] loss: 1.294
[2, 12000] loss: 1.273
Finished Training
Accuracy of plane : 59 %
Accuracy of car : 77 %
Accuracy of bird : 43 %
Accuracy of cat : 30 %
Accuracy of deer : 51 %
Accuracy of dog : 52 %
Accuracy of frog : 77 %
Accuracy of horse : 59 %
Accuracy of ship : 63 %
Accuracy of truck : 48 %
GPU 上训练
[1, 2000] loss: 2.194
[1, 4000] loss: 1.851
[1, 6000] loss: 1.691
[1, 8000] loss: 1.581
[1, 10000] loss: 1.535
[1, 12000] loss: 1.492
[2, 2000] loss: 1.397
[2, 4000] loss: 1.379
[2, 6000] loss: 1.347
[2, 8000] loss: 1.314
[2, 10000] loss: 1.302
[2, 12000] loss: 1.278
Finished Training
Accuracy of plane : 69 %
Accuracy of car : 65 %
Accuracy of bird : 43 %
Accuracy of cat : 18 %
Accuracy of deer : 50 %
Accuracy of dog : 53 %
Accuracy of frog : 55 %
Accuracy of horse : 66 %
Accuracy of ship : 75 %
Accuracy of truck : 63 %