Pytorch实现CNN的通用框架

1.导入库函数

import torch as t
from torch.utils.data import Dataset,DataLoader
from torch import optim
from torchvision import transforms
from torchvision.datasets import ImageFolder

2.数据处理

device = t.device("cuda:0" if t.cuda.is_available() else "cpu")
 
transform = transforms.Compose([transforms.Resize((224,224)),transforms.ToTensor(), # 转为Tensor
        transforms.Normalize((0.5,), (0.5,)), ])# 归一化
 
train = ImageFolder('D:/py_files/dog_cat/data/dogcat',transform)
valid = ImageFolder('D:/py_files/dog_cat/data/dogcat',transform)
 
train_dataloader = t.utils.data.DataLoader(train,batch_size=32,num_workers=0,shuffle=True)
test_dataloader = t.utils.data.DataLoader(valid,batch_size=32,num_workers=0,shuffle=False)

3.定义模型

自定义了VGG模型

import torch.nn as nn
import