import torch
import torchvision
from torchvision import datasets,transforms, models
import os
import numpy as np
import matplotlib.pyplot as plt
from torch.autograd import Variable
import time
%matplotlib inline
transform = transforms.Compose([transforms.CenterCrop(224), #将给定的PIL.Image进行中心切割，得到给定的size将给定的PIL.Image进行中心切割，得到给定的size
transforms.ToTensor(),
transforms.Normalize([0.5,0.5,0.5], [0.5,0.5,0.5])])
path = "D:/Jupyter/PyTorch/CNN/Cats_Vs_Dogs/data"
data_image = {x:datasets.ImageFolder(root = os.path.join(path,x), #数据加载成字典，并做了上面定义的预处理
transform = transform)
for x in ["train", "val"]}
data_loader_image = {x:torch.utils.data.DataLoader(dataset=data_image[x], #数据加载器；用于迭代训练
batch_size = 4,
shuffle = True)
for x in ["train", "val"]}
use_gpu = torch.cuda.is_available()
print(use_gpu)
classes = data_image["train"].classes
classes_index = data_image["train"].class_to_idx
print(classes)
print(classes_index)
['cat', 'dog']
{'cat': 0, 'dog': 1}
print("train data set:", len(data_image["train"]))
print("val data set:", len(data_image["val"]))
train data set: 976
val data set: 200
# 装载完成后，我们可以选取其中一个批次的数据进行预览。进行数据预览的代码如下：
X_train,y_train = next(iter(data_loader_image["train"]))
mean = [0.5, 0.5, 0.5]
std = [0.5, 0.5, 0.5]
print(X_train.size())
img = torchvision.utils.make_grid(X_train) # 这里的图像数据的size是（4,channel,height,weight）, 将一小batch图片变为一张图。
print(img.size()) #而我们显示图 像的size顺序为（height,weight,channel）
img = img.numpy().transpose((1,2,0)) # 转为（height,weight,channel）并转化为numpy形式
print('numpy.shape',img.shape)
img = img*std + mean # 将上面Normalize后的图像还原
print([classes[i] for i in y_train])
plt.imshow(img)
torch.Size([4, 3, 224, 224])
torch.Size([3, 228, 906])
numpy.shape (228, 906, 3)
['cat', 'dog', 'cat', 'dog']
Out[44]:
<matplotlib.image.AxesImage at 0x17584132ac8>
model = models.vgg16(pretrained = True)
print(model)
VGG(
(features): Sequential(
(0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): ReLU(inplace)
(2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(3): ReLU(inplace)
(4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(6): ReLU(inplace)
(7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(8): ReLU(inplace)
(9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(11): ReLU(inplace)
(12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(13): ReLU(inplace)
(14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(15): ReLU(inplace)
(16): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(17): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(18): ReLU(inplace)
(19): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(20): ReLU(inplace)
(21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(22): ReLU(inplace)
(23): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(24): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(25): ReLU(inplace)
(26): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(27): ReLU(inplace)
(28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(29): ReLU(inplace)
(30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(classifier): Sequential(
(0): Linear(in_features=25088, out_features=4096, bias=True)
(1): ReLU(inplace)
(2): Dropout(p=0.5)
(3): Linear(in_features=4096, out_features=4096, bias=True)
(4): ReLU(inplace)
(5): Dropout(p=0.5)
(6): Linear(in_features=4096, out_features=1000, bias=True)
)
)
for parma in model.parameters():
parma.requires_grad = False
model.classifier = torch.nn.Sequential(torch.nn.Linear(25088, 4096), #修改全连接层参数
torch.nn.ReLU(),
torch.nn.Dropout(p=0.5),
torch.nn.Linear(4096, 4096),
torch.nn.ReLU(),
torch.nn.Dropout(p=0.5),
torch.nn.Linear(4096, 2)) #因为是二分类 ，所以输出层改为2
for index, parma in enumerate(model.classifier.parameters()):
if index == 6:
parma.requires_grad = True #parma.requires_grid = False是冻结参数，即使发生新的训练也不会进行参数的更新
if use_gpu:
model = model.cuda()
cost = torch.nn.CrossEntropyLoss()
optimizer = torch.optim.Adam(model.classifier.parameters()) #只对全连接层参数进行更新优化
print(model)
VGG(
(features): Sequential(
(0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): ReLU(inplace)
(2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(3): ReLU(inplace)
(4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(6): ReLU(inplace)
(7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(8): ReLU(inplace)
(9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(11): ReLU(inplace)
(12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(13): ReLU(inplace)
(14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(15): ReLU(inplace)
(16): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(17): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(18): ReLU(inplace)
(19): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(20): ReLU(inplace)
(21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(22): ReLU(inplace)
(23): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(24): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(25): ReLU(inplace)
(26): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(27): ReLU(inplace)
(28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(29): ReLU(inplace)
(30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(classifier): Sequential(
(0): Linear(in_features=25088, out_features=4096, bias=True)
(1): ReLU()
(2): Dropout(p=0.5)
(3): Linear(in_features=4096, out_features=4096, bias=True)
(4): ReLU()
(5): Dropout(p=0.5)
(6): Linear(in_features=4096, out_features=2, bias=True)
)
n_epochs = 1
for epoch in range(n_epochs):
since = time.time()
print("Epoch{}/{}".format(epoch, n_epochs))
print("-"*10)
for param in ["train", "val"]:
if param == "train":
model.train = True
else:
model.train = False
running_loss = 0.0
running_correct = 0
batch = 0
for data in data_loader_image[param]:
batch += 1
X,y = data
if use_gpu:
X,y - Variable(X.cuda()), Variable(y.cuda())
else:
X,y = Variable(X), Variable(y)
optimizer.zero_grad()
y_pred = model(X)
_,pred = torch.max(y_pred.data, 1)
loss = cost(y_pred,y)
if param == "train":
loss.backward()
optimizer.step()
running_loss += loss.data[0]
running_correct += torch.sum(pred == y.data)
if batch%5 == 0 and param == "train":
print("Batch {}, Train Loss:{:.4f}, Train ACC:{:.4f}".format(
batch, running_loss/(4*batch), 100*running_correct/(4*batch)))
epoch_loss = running_loss/len(data_image[param])
epoch_correct = 100*running_correct/len(data_image[param])
print("{} Loss:{:.4f}, Correct:{:.4f}".format(param, epoch_loss, epoch_correct))
now_time = time.time() - since
print("Training time is:{:.0f}m {:.0f}s".format(now_time//60, now_time%60))
Batch 5, Train Loss:0.8224, Train ACC:75.0000
Batch 10, Train Loss:0.9506, Train ACC:75.0000
Batch 15, Train Loss:0.8541, Train ACC:78.0000
Batch 20, Train Loss:0.9347, Train ACC:81.0000
Batch 25, Train Loss:1.3462, Train ACC:80.0000
......
Batch 225, Train Loss:1.8574, Train ACC:86.0000
Batch 230, Train Loss:1.8220, Train ACC:86.0000
Batch 235, Train Loss:1.8113, Train ACC:86.0000
Batch 240, Train Loss:1.8529, Train ACC:86.0000
train Loss:1.8225, Correct:86.0000
val Loss:2.0843, Correct:83.0000
Training time is:17m 41s
)
data_test_img = datasets.ImageFolder(root="D:/Jupyter/PyTorch/CNN/Cats_Vs_Dogs/data/val/", transform = transform)
data_loader_test_img = torch.utils.data.DataLoader(dataset=data_test_img,
batch_size = 16)
image, label = next(iter(data_loader_test_img))
images = Variable(image)
y_pred = model(images)
_,pred = torch.max(y_pred.data, 1)
print(pred)
tensor([1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1])
img = torchvision.utils.make_grid(image)
img = img.numpy().transpose(1,2,0)
mean = [0.5, 0.5, 0.5]
std = [0.5, 0.5, 0.5]
img = img * std + mean
print("Pred Label:", [classes[i] for i in pred])
plt.imshow(img)
Pred Label: ['dog', 'cat', 'cat', 'cat', 'cat', 'cat', 'cat', 'cat', 'dog', 'cat', 'cat', 'cat', 'cat', 'cat', 'cat', 'dog']
Out[47]:
<matplotlib.image.AxesImage at 0x17584304dd8>
