一、数据准备与模型调用(Main.py)
# --- 1、GPU设置 ---
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
# --- 2、导入数据 ---
# 获取训练集
train_ds = torchvision.datasets.MNIST('data', train=True, transform=torchvision.transforms.ToTensor(), download=False)
print(len(train_ds))
# 获取测试集
test_ds = torchvision.datasets.MNIST('data', train=False, transform=torchvision.transforms.ToTensor(), download=False)
batch_size = 32
train_dl = torch.utils.data.DataLoader(train_ds,
batch_size=batch_size,
shuffle=True)
test_dl = torch.utils.data.DataLoader(test_ds, batch_size=batch_size)
imgs, labels = next(iter(train_dl))
print(imgs.shape)
# --- 3、数据可视化 ---
plt.figure(figsize=(20, 5))
for i, imgs in enumerate(imgs[:20]):
npimg = np.squeeze(imgs.numpy())
plt.subplot(2, 10, i + 1)
plt.imshow(npimg, cmap=plt.cm.binary)
plt.axis('off')
plt.show()
# --- 模型参数设置
model = Model().to(device)
# summary(model)
loss_fn = nn.CrossEntropyLoss() # 创建损失函数
learn_rate = 1e-2 # 学习率
opt = torch.optim.SGD(model.parameters(), lr=learn_rate)
# --- 4、模型训练 ---
epochs = 1
train_loss = []
train_acc = []
test_loss = []
test_acc = []
for epoch in range(epochs):
model.train()
epoch_train_acc, epoch_train_loss = model.train1(train_dl, model, loss_fn, opt)
# model.eval()
# epoch_test_acc, epoch_test_loss = model.test1(test_dl, model, loss_fn)
#
# train_acc.append(epoch_train_acc)
# train_loss.append(epoch_train_loss)
# test_acc.append(epoch_test_acc)
# test_loss.append(epoch_test_loss)
#
# template = 'Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%，Test_loss:{:.3f}'
#
# print(template.format(epoch + 1, epoch_train_acc * 100, epoch_train_loss, epoch_test_acc * 100, epoch_test_loss))
# print('Done')
# --- 5、结果可视化 ---
import matplotlib.pyplot as plt
# 隐藏警告
import warnings
# warnings.filterwarnings("ignore") # 忽略警告信息
# plt.rcParams['font.sans-serif'] = ['SimHei'] # 用来正常显示中文标签
# plt.rcParams['axes.unicode_minus'] = False # 用来正常显示负号
# plt.rcParams['figure.dpi'] = 100 # 分辨率
#
# epochs_range = range(epochs)
#
# plt.figure(figsize=(12, 3))
# plt.subplot(1, 2, 1) # 1行2列 后面1表示图的编号
#
# plt.plot(epochs_range, train_acc, label='Training Accuracy')
# plt.plot(epochs_range, test_acc, label='Test Accuracy')
# plt.legend(loc='lower right')
# plt.title('Training and Validation Accuracy')
#
# plt.subplot(1, 2, 2)
# plt.plot(epochs_range, train_loss, label='Training Loss')
# plt.plot(epochs_range, test_loss, label='Test Loss')
# plt.legend(loc='upper right')
# plt.title('Training and Validation Loss')
# plt.show()
# --- 6、单个图片预测 ---
# 从测试集中找一个图片来测试
for X, y in test_dl:
print(y.shape)
print(y)
plt.figure(figsize=(2, 2))
npimg = np.squeeze(X[1:2, :].numpy())
plt.subplot(1, 1, 1)
plt.imshow(npimg, cmap=plt.cm.binary)
plt.axis('off')
plt.show()
Z = X[1:2, :]
print(Z.shape)
result = model.predict(Z, model)
item1 = result.item()
item2 = y[1].item()
print('real:' + str(item2))
print('predict:' + str(item1))
print('结束一轮')
二、模型构建、训练与预测（Model.py）
import torch
import torch.nn.functional as F
import torch.nn as nn
num_classes = 10 # 图片的类别
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
class Model(nn.Module):
def __init__(self):
super().__init__()
# 特征提取网络构建
self.conv1 = nn.Conv2d(1, 32, kernel_size=3) # 第一层卷积，卷积核大小为3*3
self.pool1 = nn.MaxPool2d(2) # 设置池化层，池化核大小为2*2
self.conv2 = nn.Conv2d(32, 64, kernel_size=3) # 第二层卷积，卷积核大小为3*3
self.pool2 = nn.MaxPool2d(2)
# 分类网络
self.fc1 = nn.Linear(1600, 64)
self.fc2 = nn.Linear(64, num_classes)
# 前向传播方法
def forward(self, x):
x = self.pool1(F.relu(self.conv1(x)))
x = self.pool2(F.relu(self.conv2(x)))
x = torch.flatten(x, start_dim=1)
x = F.relu(self.fc1(x))
x = self.fc2(x)
return x
# 模型训练
def train1(self, dataloader, model, loss_fn, optimizer):
size = len(dataloader.dataset) # 训练集大小，一共6000张图片
num_batches = len(dataloader) # 批次数目， 6000/32 = 1875
train_loss, train_acc = 0, 0 # 初始化训练损失和正确值
i = 0
for X, y in dataloader: # 获取图片及其标签
X, y = X.to(device), y.to(device)
# 计算预测误差
pred = model.forward(X) # 网络输出
loss = loss_fn(pred, y) # 计算网络输出和真实值之间的差距，targets真实值，计算二者差值即为损失 (pred为minibatch的预测类别的概率， y为所属类别的索引)
# 反向传播
optimizer.zero_grad() # grad属性归零
loss.backward() # 反向传播
optimizer.step() # 每一步自动更新
# 记录acc argmax(1)求每行 最大值的索引值 添加batch里面正确预测的个数。 argmax(0)表示求列最大值的索引
train_acc += (pred.argmax(1) == y).type(torch.float).sum().item()
train_loss += loss.item()
# 跑完一个epoch，计算所有图片的平均预测准确率
train_acc /= size
# 跑完一个epoch,计算所有batch的平均loss
train_loss /= num_batches
return train_acc, train_loss
# 模型测试
def test1(self, dataloader, model, loss_fn):
size = len(dataloader.dataset) # 测试集的大小，一共10000张图片
num_batches = len(dataloader) # 批次数目，313（10000/32=312.5，向上取整）
test_loss, test_acc = 0, 0 # 初始化训练损失和正确值
# 当不尽兴训练时，停止梯度更新，节省计算内存消耗
with torch.no_grad():
for imgs, target in dataloader:
imgs, target = imgs.to(device), target.to(device)
# 计算loss
target_pred = model(imgs)
loss = loss_fn(target_pred, target)
test_loss += loss.item()
test_acc += (target_pred.argmax(1) == target).type(torch.float).sum().item()
test_acc /= size
test_loss /= num_batches
return test_acc, test_loss
def predict(self, img, model):
pt = model.forward(img)
result = pt.argmax(1)
return result
