tp5网站开发模板,游戏推广员一个月能赚多少,网络营销渠道策略包括,网站建设有哪些内容一、训练任务概述
动机#xff1a;由于后续的课题中会用到类似图像去噪的算法#xff0c;考虑先用U-Net#xff0c;这里做一个前置的尝试。
训练任务#xff1a;分割出图像中的细胞。
数据集#xff1a;可私
数据集结构#xff1a; 二、具体实现
U-Net的网络实现是现…一、训练任务概述
动机由于后续的课题中会用到类似图像去噪的算法考虑先用U-Net这里做一个前置的尝试。
训练任务分割出图像中的细胞。
数据集可私
数据集结构 二、具体实现
U-Net的网络实现是现成的只需要在网上找一个比较漂亮的实现一般都是模块化写的很漂亮copy就可以了需要特别注意的是最后整合的模型
2.1 基础模型模块实现
双卷积模块
class DoubleConv(nn.Module):(convolution [BN] ReLU) * 2def __init__(self, in_channels, out_channels, mid_channelsNone):super().__init__()if not mid_channels:mid_channels out_channelsself.double_conv nn.Sequential(nn.Conv2d(in_channels, mid_channels, kernel_size3, padding1, biasFalse),nn.BatchNorm2d(mid_channels),nn.ReLU(inplaceTrue),nn.Conv2d(mid_channels, out_channels, kernel_size3, padding1, biasFalse),nn.BatchNorm2d(out_channels),nn.ReLU(inplaceTrue))def forward(self, x):return self.double_conv(x)
上采样模块
class Up(nn.Module):Upscaling then double convdef __init__(self, in_channels, out_channels, bilinearTrue):super().__init__()# if bilinear, use the normal convolutions to reduce the number of channelsif bilinear:self.up nn.Upsample(scale_factor2, modebilinear, align_cornersTrue)self.conv DoubleConv(in_channels, out_channels, in_channels // 2)else:self.up nn.ConvTranspose2d(in_channels, in_channels // 2, kernel_size2, stride2)self.conv DoubleConv(in_channels, out_channels)def forward(self, x1, x2):x1 self.up(x1)# input is CHWdiffY x2.size()[2] - x1.size()[2]diffX x2.size()[3] - x1.size()[3]x1 torch.nn.functional.pad(x1, [diffX // 2, diffX - diffX // 2,diffY // 2, diffY - diffY // 2])x torch.cat([x2, x1], dim1)return self.conv(x)
下采样模块
class Down(nn.Module):Downscaling with maxpool then double convdef __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)
输出层
class OutConv(nn.Module):def __init__(self, in_channels, out_channels):super(OutConv, self).__init__()self.conv nn.Conv2d(in_channels, out_channels, kernel_size1)def forward(self, x):return self.conv(x)
2.2 整合模块-模型
class UNet(L.LightningModule):def __init__(self, n_channels, n_classes, bilinearFalse):super(UNet, self).__init__()self.n_channels n_channelsself.n_classes n_classesself.bilinear bilinearself.inc (DoubleConv(n_channels, 64))self.down1 (Down(64, 128))self.down2 (Down(128, 256))self.down3 (Down(256, 512))factor 2 if bilinear else 1self.down4 (Down(512, 1024 // factor))self.up1 (Up(1024, 512 // factor, bilinear))self.up2 (Up(512, 256 // factor, bilinear))self.up3 (Up(256, 128 // factor, 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# 对应的层设置检查点节省显存m可用可不用def use_checkpointing(self):self.inc torch.utils.checkpoint(self.inc)self.down1 torch.utils.checkpoint(self.down1)self.down2 torch.utils.checkpoint(self.down2)self.down3 torch.utils.checkpoint(self.down3)self.down4 torch.utils.checkpoint(self.down4)self.up1 torch.utils.checkpoint(self.up1)self.up2 torch.utils.checkpoint(self.up2)self.up3 torch.utils.checkpoint(self.up3)self.up4 torch.utils.checkpoint(self.up4)self.outc torch.utils.checkpoint(self.outc)# 定义优化器def configure_optimizers(self):optimizer torch.optim.Adam(self.parameters(),lr0.001)return optimizer# 定义train的单步流程def training_step(self,train_batch,batch_index):image,label train_batchimage_hat self.forward(image)# U-Net的lossloss nn.functional.mse_loss(image_hat,label)return loss# 定义val的单步流程def validation_step(self, val_batch,batch_index):image,label val_batchimage_hat self.forward(image)# U-Net的lossloss nn.functional.mse_loss(image_hat,label)self.log(val_loss,loss)return loss
注意模块可以不需要继承自L.LightningModule只要最后整合的时候继承自L.LightningModule就可以了。
2.3 数据划分
重定义Dataset类供数据集划分函数调用二者要相互配合
class UDataset(Dataset):def __init__(self,image_dir,mask_dir,transformNone):self.image_dir image_dirself.mask_dir mask_dirif transform is not None:self.transform transformelse:self.transform Nonedef __getitem__(self, index):image Image.open(self.image_dir[index]).convert(RGB)label Image.open(self.mask_dir[index]).convert(RGB)if self.transform is not None:image self.transform(image)label self.transform(label)return image,labeldef __len__(self):return len(self.image_dir) 定义数据集划分函数包括找出文件列表、定义数据预处理方式、“定义批量大小”
train_image_dir ./data/train/image/*.png
train_label_dir ./data/train/label/*.png
val_image_dir ./data/val/image/*.png
val_label_dir ./data/val/label/*.png def data_process(train_image_dir,train_label_dir,val_image_dir,val_label_dir):# 查找路径下的所有文件返回文件路径列表train_image_list glob.glob(train_image_dir)train_label_list glob.glob(train_label_dir)val_image_list glob.glob(val_image_dir)val_label_list glob.glob(val_label_dir)# 数据处理train_data_transform transforms.Compose([transforms.Resize((256,256)),transforms.ToTensor()])val_data_transform transforms.Compose([ transforms.Resize((256,256)),transforms.ToTensor()])train_dataloader data.DataLoader(UDataset(train_image_list,train_label_list,train_data_transform),batch_size5,shuffleTrue)val_dataloader data.DataLoader(UDataset(val_image_list,val_label_list,val_data_transform),batch_size5,shuffleFalse)return train_dataloader,val_dataloader
2.4 模型验证
在训练之前要看一下模型的结构有没有错误用summary打印出网络的结构 # 模型验证device torch.device(cuda if torch.cuda.is_available() else cpu)model UNet(n_channels3,n_classes1).to(device)print(summary(model,(3,512,512)))
也可以用其他的方法查看网络结构
2.5 模型训练
加入TensorBoardLogger是为了可视化训练Loss
训练的流程遵循前文的基本流程 # 创建 TensorBoardLoggerlogger TensorBoardLogger(tb_logs, nameunet)# 创建 Trainertrainer L.Trainer(max_epochs20, loggerlogger)# 划分数据集train_dataloader,val_dataloader data_process(train_image_dir,train_label_dir,val_image_dir,val_label_dir)# 创建模型model UNet(n_channels3,n_classes1)# 启动模型训练过程trainer.fit(model,train_dataloader,val_dataloader)# 保存模型权重torch.save(model.state_dict(),./model.pth)
