Pytorch深度学习实战教程(三):UNet模型训练
1
本文属于 Pytorch 深度学习语义分割系列教程。
该系列文章的内容有:
Pytorch 的基本使用
语义分割算法讲解
由于微信不允许外部链接,你需要点击页尾左下角的“阅读原文”,才能访问文中的链接,文中的所有外部链接都已使用蓝色字体标记。
PS:文中出现的所有代码,均可在我的github上下载,欢迎Follow、Star:点击查看
2
项目背景
深度学习算法,无非就是我们解决一个问题的方法。选择什么样的网络去训练,进行什么样的预处理,采用什么Loss和优化方法,都是根据具体的任务而定的。
所以,让我们先看一下今天的任务。
没错,就是 UNet 论文中的经典任务:医学图像分割。
选择它作为今天的任务,就是因为简单,好上手。
简单描述一个这个任务:如动图所示,给一张细胞结构图,我们要把每个细胞互相分割开来。
这个训练数据只有30张,分辨率为512x512,这些图片是果蝇的电镜图。
好了,任务介绍完毕,开始准备训练模型。
3
数据加载:数据怎么加载,标签怎么定义,用什么数据增强方法,都是这一步进行。
模型选择:模型我们已经准备好了,就是该系列上篇文章讲到的 UNet 网络。
算法选择:算法选择也就是我们选什么 loss ,用什么优化算法。
# ================================================================== #
# Input pipeline for custom dataset #
# ================================================================== #
# You should build your custom dataset as below.
class CustomDataset(torch.utils.data.Dataset):
def __init__(self):
# TODO
# 1. Initialize file paths or a list of file names.
pass
def __getitem__(self, index):
# TODO
# 1. Read one data from file (e.g. using numpy.fromfile, PIL.Image.open).
# 2. Preprocess the data (e.g. torchvision.Transform).
# 3. Return a data pair (e.g. image and label).
pass
def __len__(self):
# You should change 0 to the total size of your dataset.
return 0
# You can then use the prebuilt data loader.
custom_dataset = CustomDataset()
train_loader = torch.utils.data.DataLoader(dataset=custom_dataset,
batch_size=64,
shuffle=True)
import torch
import cv2
import os
import glob
from torch.utils.data import Dataset
import random
class ISBI_Loader(Dataset):
def __init__(self, data_path):
# 初始化函数,读取所有data_path下的图片
self.data_path = data_path
self.imgs_path = glob.glob(os.path.join(data_path, 'image/*.png'))
def augment(self, image, flipCode):
# 使用cv2.flip进行数据增强,filpCode为1水平翻转,0垂直翻转,-1水平+垂直翻转
flip = cv2.flip(image, flipCode)
return flip
def __getitem__(self, index):
# 根据index读取图片
image_path = self.imgs_path[index]
# 根据image_path生成label_path
label_path = image_path.replace('image', 'label')
# 读取训练图片和标签图片
image = cv2.imread(image_path)
label = cv2.imread(label_path)
# 将数据转为单通道的图片
image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
label = cv2.cvtColor(label, cv2.COLOR_BGR2GRAY)
image = image.reshape(1, image.shape[0], image.shape[1])
label = label.reshape(1, label.shape[0], label.shape[1])
# 处理标签,将像素值为255的改为1
if label.max() > 1:
label = label / 255
# 随机进行数据增强,为2时不做处理
flipCode = random.choice([-1, 0, 1, 2])
if flipCode != 2:
image = self.augment(image, flipCode)
label = self.augment(label, flipCode)
return image, label
def __len__(self):
# 返回训练集大小
return len(self.imgs_path)
if __name__ == "__main__":
isbi_dataset = ISBI_Loader("data/train/")
print("数据个数:", len(isbi_dataset))
train_loader = torch.utils.data.DataLoader(dataset=isbi_dataset,
batch_size=2,
shuffle=True)
for image, label in train_loader:
print(image.shape)
""" Parts of the U-Net model """
"""https://github.com/milesial/Pytorch-UNet/blob/master/unet/unet_parts.py"""
import torch
import torch.nn as nn
import torch.nn.functional as F
class DoubleConv(nn.Module):
"""(convolution => [BN] => ReLU) * 2"""
def __init__(self, in_channels, out_channels):
super().__init__()
self.double_conv = nn.Sequential(
nn.Conv2d(in_channels, out_channels, kernel_size=3, padding=1),
nn.BatchNorm2d(out_channels),
nn.ReLU(inplace=True),
nn.Conv2d(out_channels, out_channels, kernel_size=3, padding=1),
nn.BatchNorm2d(out_channels),
nn.ReLU(inplace=True)
)
def forward(self, x):
return self.double_conv(x)
class Down(nn.Module):
"""Downscaling with maxpool then double conv"""
def __init__(self, in_channels, out_channels):
super().__init__()
self.maxpool_conv = nn.Sequential(
nn.MaxPool2d(2),
DoubleConv(in_channels, out_channels)
)
def forward(self, x):
return self.maxpool_conv(x)
class Up(nn.Module):
"""Upscaling then double conv"""
def __init__(self, in_channels, out_channels, bilinear=True):
super().__init__()
# if bilinear, use the normal convolutions to reduce the number of channels
if bilinear:
self.up = nn.Upsample(scale_factor=2, mode='bilinear', align_corners=True)
else:
self.up = nn.ConvTranspose2d(in_channels // 2, in_channels // 2, kernel_size=2, stride=2)
self.conv = DoubleConv(in_channels, out_channels)
def forward(self, x1, x2):
x1 = self.up(x1)
# input is CHW
diffY = torch.tensor([x2.size()[2] - x1.size()[2]])
diffX = torch.tensor([x2.size()[3] - x1.size()[3]])
x1 = F.pad(x1, [diffX // 2, diffX - diffX // 2,
diffY // 2, diffY - diffY // 2])
x = torch.cat([x2, x1], dim=1)
return self.conv(x)
class OutConv(nn.Module):
def __init__(self, in_channels, out_channels):
super(OutConv, self).__init__()
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size=1)
def forward(self, x):
return self.conv(x)
""" Full assembly of the parts to form the complete network """
"""Refer https://github.com/milesial/Pytorch-UNet/blob/master/unet/unet_model.py"""
import torch.nn.functional as F
from .unet_parts import *
class UNet(nn.Module):
def __init__(self, n_channels, n_classes, bilinear=True):
super(UNet, self).__init__()
self.n_channels = n_channels
self.n_classes = n_classes
self.bilinear = bilinear
self.inc = DoubleConv(n_channels, 64)
self.down1 = Down(64, 128)
self.down2 = Down(128, 256)
self.down3 = Down(256, 512)
self.down4 = Down(512, 512)
self.up1 = Up(1024, 256, bilinear)
self.up2 = Up(512, 128, bilinear)
self.up3 = Up(256, 64, bilinear)
self.up4 = Up(128, 64, bilinear)
self.outc = OutConv(64, n_classes)
def forward(self, x):
x1 = self.inc(x)
x2 = self.down1(x1)
x3 = self.down2(x2)
x4 = self.down3(x3)
x5 = self.down4(x4)
x = self.up1(x5, x4)
x = self.up2(x, x3)
x = self.up3(x, x2)
x = self.up4(x, x1)
logits = self.outc(x)
return logits
if __name__ == '__main__':
net = UNet(n_channels=3, n_classes=1)
print(net)
from model.unet_model import UNet
from utils.dataset import ISBI_Loader
from torch import optim
import torch.nn as nn
import torch
def train_net(net, device, data_path, epochs=40, batch_size=1, lr=0.00001):
# 加载训练集
isbi_dataset = ISBI_Loader(data_path)
train_loader = torch.utils.data.DataLoader(dataset=isbi_dataset,
batch_size=batch_size,
shuffle=True)
# 定义RMSprop算法
optimizer = optim.RMSprop(net.parameters(), lr=lr, weight_decay=1e-8, momentum=0.9)
# 定义Loss算法
criterion = nn.BCEWithLogitsLoss()
# best_loss统计,初始化为正无穷
best_loss = float('inf')
# 训练epochs次
for epoch in range(epochs):
# 训练模式
net.train()
# 按照batch_size开始训练
for image, label in train_loader:
optimizer.zero_grad()
# 将数据拷贝到device中
image = image.to(device=device, dtype=torch.float32)
label = label.to(device=device, dtype=torch.float32)
# 使用网络参数,输出预测结果
pred = net(image)
# 计算loss
loss = criterion(pred, label)
print('Loss/train', loss.item())
# 保存loss值最小的网络参数
if loss < best_loss:
best_loss = loss
torch.save(net.state_dict(), 'best_model.pth')
# 更新参数
loss.backward()
optimizer.step()
if __name__ == "__main__":
# 选择设备,有cuda用cuda,没有就用cpu
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
# 加载网络,图片单通道1,分类为1。
net = UNet(n_channels=1, n_classes=1)
# 将网络拷贝到deivce中
net.to(device=device)
# 指定训练集地址,开始训练
data_path = "data/train/"
train_net(net, device, data_path)
4
预测
模型训练好了,我们可以用它在测试集上看下效果。
在工程根目录创建 predict.py 文件,编写如下代码:
import glob
import numpy as np
import torch
import os
import cv2
from model.unet_model import UNet
if __name__ == "__main__":
# 选择设备,有cuda用cuda,没有就用cpu
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
# 加载网络,图片单通道,分类为1。
net = UNet(n_channels=1, n_classes=1)
# 将网络拷贝到deivce中
net.to(device=device)
# 加载模型参数
net.load_state_dict(torch.load('best_model.pth', map_location=device))
# 测试模式
net.eval()
# 读取所有图片路径
tests_path = glob.glob('data/test/*.png')
# 遍历所有图片
for test_path in tests_path:
# 保存结果地址
save_res_path = test_path.split('.')[0] + '_res.png'
# 读取图片
img = cv2.imread(test_path)
# 转为灰度图
img = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
# 转为batch为1,通道为1,大小为512*512的数组
img = img.reshape(1, 1, img.shape[0], img.shape[1])
# 转为tensor
img_tensor = torch.from_numpy(img)
# 将tensor拷贝到device中,只用cpu就是拷贝到cpu中,用cuda就是拷贝到cuda中。
img_tensor = img_tensor.to(device=device, dtype=torch.float32)
# 预测
pred = net(img_tensor)
# 提取结果
pred = np.array(pred.data.cpu()[0])[0]
# 处理结果
pred[pred >= 0.5] = 255
pred[pred < 0.5] = 0
# 保存图片
cv2.imwrite(save_res_path, pred)
运行完后,你可以在data/test目录下,看到预测结果:
大功告成!
5
本文主要讲解了训练模型的三个步骤:数据加载、模型选择、算法选择。
这是一个简单的例子,训练正常的视觉任务,要复杂很多。比如:在训练模型的时候,需要根据模型在验证集上的准确率选择保存哪个模型;需要支持tensorboard方便我们观察loss收敛情况等等。
PS:如果觉得本篇本章对您有所帮助,欢迎关注、评论、赞!