多层感知机的从零开始实现与pytorch简洁实现（pycharm版）

# -*- coding: UTF-8 -*- """ @Author ：Douglas @Project ：deepLearning_try @File ：mlp_from_zero.py @IDE ：PyCharm @Date ：2024/8/5 """ import torch import torchvision from torch.utils import data from torchvision import transforms # from torch.utils.tensorboard import SummaryWriter from torch import nn def try_gpu(i=0): # 有 GPU 则使用，否则使用 CPU """如果存在，则返回gpu(i)，否则返回cpu()""" if torch.cuda.device_count() >= i + 1: # return torch.device(f'cuda:{i}') return torch.device('cpu') return torch.device('cpu') def load_data_fashion_mnist(batch_size, loader_num, resize=None): # 数据集准备 """下载Fashion-MNIST数据集，然后将其加载到内存中""" trans = [transforms.ToTensor()] # 数据变换操作列表 if resize: # 如果需要改变形状的话，把resize的操作加到变换列表里 trans.insert(0, transforms.Resize(resize)) trans = transforms.Compose(trans) # 将变换操作集合起来 # 下载训练集和数据集 mnist_train = torchvision.datasets.FashionMNIST( root="../dataset", train=True, transform=trans, download=True) mnist_test = torchvision.datasets.FashionMNIST( root="../dataset", train=False, transform=trans, download=True) return (data.DataLoader(mnist_train, batch_size, shuffle=True, num_workers=loader_num), data.DataLoader(mnist_test, batch_size, shuffle=False, num_workers=loader_num)) def relu(X): # 激活函数 a = torch.zeros_like(X) return torch.max(X, a) def net(X): # 模型 X = X.reshape((-1, num_inputs)) # 展平 H = relu(X@W1 + b1) # 这里“@”代表矩阵乘法 return (H@W2 + b2) def accuracy(y_hat, y): """计算预测正确的数量""" if len(y_hat.shape) > 1 and y_hat.shape[1] > 1: y_hat = y_hat.argmax(axis=1) cmp = y_hat.type(y.dtype) == y return float(cmp.type(y.dtype).sum()) def evaluate_accuracy(net, data_iter): """计算在指定数据集上模型的精度""" if isinstance(net, torch.nn.Module): net.eval() # 将模型设置为评估模式 metric = Accumulator(2) # 正确预测数、预测总数 with torch.no_grad(): for X, y in data_iter: X = X.to(try_gpu()) y = y.to(try_gpu()) metric.add(accuracy(net(X), y), y.numel()) return metric[0] / metric[1] class Accumulator: """在n个变量上累加""" def __init__(self, n): self.data = [0.0] * n def add(self, *args): self.data = [a + float(b) for a, b in zip(self.data, args)] def reset(self): self.data = [0.0] * len(self.data) def __getitem__(self, idx): return self.data[idx] def train_epoch(net, train_iter, loss, updater): #@save """训练模型一个迭代周期""" # 将模型设置为训练模式 if isinstance(net, torch.nn.Module): net.train() # 训练损失总和、训练准确度总和、样本数 metric = Accumulator(3) for X, y in train_iter: X = X.to(try_gpu()) y = y.to(try_gpu()) # 计算梯度并更新参数 y_hat = net(X) l = loss(y_hat, y) if isinstance(updater, torch.optim.Optimizer): # 使用PyTorch内置的优化器和损失函数 updater.zero_grad() l.mean().backward() updater.step() else: # 使用定制的优化器和损失函数 l.sum().backward() updater(params, lr, batch_size) metric.add(float(l.sum()), accuracy(y_hat, y), y.numel()) # 返回训练损失和训练精度 return metric[0] / metric[2], metric[1] / metric[2] def train(net, train_iter, test_iter, loss, num_epochs, updater): """训练模型""" for epoch in range(num_epochs): train_loss, train_acc = train_epoch(net, train_iter, loss, updater) test_acc = evaluate_accuracy(net, test_iter) # writer.add_scalar('train_loss', train_loss, epoch) # writer.add_scalar('train_acc', train_acc, epoch) # writer.add_scalar('test_acc', test_acc, epoch) print(f'第{epoch + 1}轮的训练损失为{train_loss}') print(f'第{epoch + 1}轮的训练精度为{train_acc}') print(f'第{epoch + 1}轮的测试集精度为{test_acc}') return train_loss, train_acc, test_acc if __name__ == '__main__': batch_size = 256 train_iter, test_iter = load_data_fashion_mnist(batch_size, 4) # writer = SummaryWriter('./runs/mlp_from_zero') # 设置tb文件保存位置 num_inputs, num_outputs, num_hiddens = 784, 10, 256 W1 = nn.Parameter(torch.randn( num_inputs, num_hiddens, requires_grad=True) * 0.01) b1 = nn.Parameter(torch.zeros(num_hiddens, requires_grad=True)) W2 = nn.Parameter(torch.randn( num_hiddens, num_outputs, requires_grad=True) * 0.01) b2 = nn.Parameter(torch.zeros(num_outputs, requires_grad=True)) params = [W1, b1, W2, b2] loss = nn.CrossEntropyLoss(reduction='none') # 损失函数 num_epochs, lr = 10, 0.1 optimizer = torch.optim.SGD(params, lr=lr) train(net, train_iter, test_iter, loss, num_epochs, optimizer) # writer.close()