软件工程第四次实验：CNN实战
- 猫狗大战挑战赛
1 环境准备
import numpy as np
import matplotlib.pyplot as plt
import os
import torch
import torch.nn as nn
import torchvision
from torchvision import models,transforms,datasets
import time
import json
# 判断是否存在GPU设备
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print('Using gpu: %s ' % torch.cuda.is_available())
1.1 下载图片
- 因为这个代码需要在colab上跑，速度会相对较慢。因此重新整理了数据，制作了新的数据集，训练集包含1800张图(猫的图片900张，狗的图片900张)，测试集包含2000张图。
1.2 数据处理
- 在使用CNN处理图像时，需要进行预处理。图片将被整理成 224x224x3的大小，同时还将进行归一化处理。
normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
vgg_format = transforms.Compose([
transforms.CenterCrop(224),
transforms.ToTensor(),
normalize,
])
data_dir = './dogscats'
dsets = {x: datasets.ImageFolder(os.path.join(data_dir, x), vgg_format)
for x in ['train', 'valid']}
dset_sizes = {x: len(dsets[x]) for x in ['train', 'valid']}
dset_classes = dsets['train'].classes
- 查看训练集的属性
print(dsets['train'].classes)
print(dsets['train'].class_to_idx)
print(dsets['train'].imgs[:5])
print('dset_sizes: ', dset_sizes)
- 可视化地查看加载后的图片，可以看到第一个batch的5张图片
loader_train = torch.utils.data.DataLoader(dsets['train'], batch_size=64, shuffle=True, num_workers=6)
loader_valid = torch.utils.data.DataLoader(dsets['valid'], batch_size=5, shuffle=False, num_workers=6)
out = torchvision.utils.make_grid(inputs_try)
imshow(out, title=[dset_classes[x] for x in labels_try])
2 创建VGG模型
- 在这部分代码中，对输入的5个图片利用VGG模型进行预测，同时，使用softmax对结果进行处理，随后展示了识别结果。
2.1 下载json文件
- 下载地址
- https://s3.amazonaws.com/deep-learning-models/image-models/imagenet_class_index.json
2.2 创建VGG模型
model_vgg = models.vgg16(pretrained=True)
with open('./imagenet_class_index.json') as f:
class_dict = json.load(f)
dic_imagenet = [class_dict[str(i)][1] for i in range(len(class_dict))]
inputs_try , labels_try = inputs_try.to(device), labels_try.to(device)
model_vgg = model_vgg.to(device)
outputs_try = model_vgg(inputs_try)
print(outputs_try)
print(outputs_try.shape)
'''
可以看到结果为5行，1000列的数据，每一列代表对每一种目标识别的结果。
但是我也可以观察到，结果非常奇葩，有负数，有正数，
为了将VGG网络输出的结果转化为对每一类的预测概率，我们把结果输入到 Softmax 函数
'''
m_softm = nn.Softmax(dim=1)
probs = m_softm(outputs_try)
vals_try,pred_try = torch.max(probs,dim=1)
print( 'prob sum: ', torch.sum(probs,1))
print( 'vals_try: ', vals_try)
print( 'pred_try: ', pred_try)
print([dic_imagenet[i] for i in pred_try.data])
imshow(torchvision.utils.make_grid(inputs_try.data.cpu()),
title=[dset_classes[x] for x in labels_try.data.cpu()])
- 输出如下
3 修改最后一层，冻结前面层的参数
- 为了在训练中冻结前面层的参数，需要设置 required_grad=False。这样，反向传播训练梯度时，前面层的权重就不会自动更新了。训练中，只会更新最后一层的参数。
print(model_vgg)
model_vgg_new = model_vgg;
for param in model_vgg_new.parameters():
param.requires_grad = False
model_vgg_new.classifier._modules['6'] = nn.Linear(4096, 2)
model_vgg_new.classifier._modules['7'] = torch.nn.LogSoftmax(dim = 1)
model_vgg_new = model_vgg_new.to(device)
print(model_vgg_new.classifier)
4 训练并测试全连接层
- 创建损失函数和优化器及模型训练
criterion = nn.NLLLoss()
# 学习率
lr = 0.001
optimizer_vgg = torch.optim.SGD(model_vgg_new.classifier[6].parameters(),lr = lr)
def train_model(model,dataloader,size,epochs=1,optimizer=None):
model.train()
for epoch in range(epochs):
running_loss = 0.0
running_corrects = 0
count = 0
for inputs,classes in dataloader:
inputs = inputs.to(device)
classes = classes.to(device)
outputs = model(inputs)
loss = criterion(outputs,classes)
optimizer = optimizer
optimizer.zero_grad()
loss.backward()
optimizer.step()
_,preds = torch.max(outputs.data,1)
# statistics
running_loss += loss.data.item()
running_corrects += torch.sum(preds == classes.data)
count += len(inputs)
print('Training: No. ', count, ' process ... total: ', size)
epoch_loss = running_loss / size
epoch_acc = running_corrects.data.item() / size
print('Loss: {:.4f} Acc: {:.4f}'.format(
epoch_loss, epoch_acc))
train_model(model_vgg_new,loader_train,size=dset_sizes['train'], epochs=1,
optimizer=optimizer_vgg)
- 测试模型
def test_model(model,dataloader,size):
model.eval()
predictions = np.zeros(size)
all_classes = np.zeros(size)
all_proba = np.zeros((size,2))
i = 0
running_loss = 0.0
running_corrects = 0
for inputs,classes in dataloader:
inputs = inputs.to(device)
classes = classes.to(device)
outputs = model(inputs)
loss = criterion(outputs,classes)
_,preds = torch.max(outputs.data,1)
# statistics
running_loss += loss.data.item()
running_corrects += torch.sum(preds == classes.data)
predictions[i:i+len(classes)] = preds.to('cpu').numpy()
all_classes[i:i+len(classes)] = classes.to('cpu').numpy()
all_proba[i:i+len(classes),:] = outputs.data.to('cpu').numpy()
i += len(classes)
print('Testing: No. ', i, ' process ... total: ', size)
epoch_loss = running_loss / size
epoch_acc = running_corrects.data.item() / size
print('Loss: {:.4f} Acc: {:.4f}'.format(
epoch_loss, epoch_acc))
return predictions, all_proba, all_classes
predictions, all_proba, all_classes = test_model(model_vgg_new,loader_valid,size=dset_sizes['valid'])
5 用训练好的模型提交csv到比赛页面
5.1 测试代码
dsets = datasets.ImageFolder('D:/dog_cat_data/dogscats/test', vgg_format)
final = {} # 结果数组
loader_test = torch.utils.data.DataLoader(dsets, batch_size=1, shuffle=False, num_workers=0)
model_vgg_new = torch.load("./model/md.pth")
def test(model, dataloader, size):
model.eval() # 参数固定
cnt = 0 # count
for inputs, _ in dataloader:
if cnt < size:
inputs = inputs.to(device)
outputs = model(inputs)
_, preds = torch.max(outputs.data, 1) # 预测值最大化
key = dsets.imgs[cnt][0].split("/")[-1].split('.')[0] # 对目录项进行分割
final[key] = preds[0]
cnt += 1
else:
break;
test(model_vgg_new, loader_test, size=2000)
with open("./cat_dog.csv", 'a+') as f:
for key in range(2000):
f.write("{},{}\n".format(key, final[str(key)]))
5.2 保存在csv中，提交文件
6 心得体会
通过本次实验，完整地体会到了从模型训练到预测数据的全过程，深刻了解cnn模型的构建和使用。卷积神经网络和多层感知机比较类似，不过他相对于多层感知机加入了一些层。这些层由卷积成和池化组成。所以，在CNN中神经网络的层次和结构都是不固定的可用调整的。
580
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
