其他分享
首页 > 其他分享> > 0902Softmax回归从零开始

0902Softmax回归从零开始

作者:互联网

点击查看代码
import pylab
import torch
from IPython import display
from d2l import torch as d2l

# softmax回归的从零开始实现
batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size)

# 初始化模型参数
# 将展平每个图像,把它们看作长度为784的向量
# 数据集有10个类别,所以网络输出维度为10
num_inputs = 784
num_outputs = 10

W = torch.normal(0, 0.01, size=(num_inputs, num_outputs), requires_grad=True)
b = torch.zeros(num_outputs, requires_grad=True)

# 定义softmax操作
# 给定一个矩阵X,可以对所有元素求和
X = torch.tensor([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
print(X.sum(0, keepdim=True))
print(X.sum(1, keepdim=True))

# 实现softmax
def softmax(X):
    X_exp = torch.exp(X)
    partition = X_exp.sum(1, keepdim=True)
    return X_exp / partition  # 广播机制
# 将每个元素变成一个非负数。 此外,依据概率原理,每行总和为1(partition为所有分子之和)
X = torch.normal(0, 1, (2, 5))
print(X)
X_prob = softmax(X)
print(X_prob)
print(X_prob.sum(1))

# 定义模型
# 实现softmax回归模型
def net(X):
    return softmax(torch.matmul(X.reshape((-1, W.shape[0])), W) + b)

# 定义损失函数
# 创建一个数据样本y_hat,其中包含2个样本在3个类别的预测概率, 以及它们对应的标签y
y = torch.tensor([0, 2, 1])
y_hat = torch.tensor([[0.1, 0.3, 0.6], [0.3, 0.2, 0.5], [0.3, 0.8, 0.5]])
print(y_hat[[0, 1, 2], y])

# 实现交叉熵损失函数
def cross_entropy(y_hat, y):
    return - torch.log(y_hat[range(len(y_hat)), y])

print('实现交叉熵损失函数', cross_entropy(y_hat, y))

"""
为了计算精度,我们执行以下操作。 
首先,如果y_hat是矩阵,那么假定第二个维度存储每个类的预测分数。 
我们使用argmax获得每行中最大元素的索引来获得预测类别。 
然后我们将预测类别与真实y元素进行比较。 
由于等式运算符“==”对数据类型很敏感, 
因此我们将y_hat的数据类型转换为与y的数据类型一致。 
结果是一个包含0(错)和1(对)的张量。 
最后,我们求和会得到正确预测的数量。
"""

def accuracy(y_hat, y):  #@save
    """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())

print('计算预测正确的数量',accuracy(y_hat, y))

# 评估在任意模型net的精度
def evaluate_accuracy(net, data_iter):  #@save
    """计算在指定数据集上模型的精度"""
    # isinstance 判断一个对象的变量类型
    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:
            metric.add(accuracy(net(X), y), y.numel())
    return metric[0] / metric[1]

class Accumulator:  #@save
    """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]

# 未进行训练模型的精度
# evaluate_accuracy(net, test_iter)

# 训练
def train_epoch_ch3(net, train_iter, loss, updater):
    """训练模型一个迭代周期(定义见第3章)"""
    # 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:
        # 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(X.shape[0])
        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]

class Animator:  #@save
    """在动画中绘制数据"""
    def __init__(self, xlabel=None, ylabel=None, legend=None, xlim=None,
                 ylim=None, xscale='linear', yscale='linear',
                 fmts=('-', 'm--', 'g-.', 'r:'), nrows=1, ncols=1,
                 figsize=(3.5, 2.5)):
        # 增量地绘制多条线
        if legend is None:
            legend = []
        d2l.use_svg_display()
        self.fig, self.axes = d2l.plt.subplots(nrows, ncols, figsize=figsize)
        if nrows * ncols == 1:
            self.axes = [self.axes, ]
        # 使用lambda函数捕获参数
        self.config_axes = lambda: d2l.set_axes(
            self.axes[0], xlabel, ylabel, xlim, ylim, xscale, yscale, legend)
        self.X, self.Y, self.fmts = None, None, fmts

    def add(self, x, y):
        # 向图表中添加多个数据点
        if not hasattr(y, "__len__"):
            y = [y]
        n = len(y)
        if not hasattr(x, "__len__"):
            x = [x] * n
        if not self.X:
            self.X = [[] for _ in range(n)]
        if not self.Y:
            self.Y = [[] for _ in range(n)]
        for i, (a, b) in enumerate(zip(x, y)):
            if a is not None and b is not None:
                self.X[i].append(a)
                self.Y[i].append(b)
        self.axes[0].cla()
        for x, y, fmt in zip(self.X, self.Y, self.fmts):
            self.axes[0].plot(x, y, fmt)
        self.config_axes()
        display.display(self.fig)
        # d2l.plt.draw()
        # d2l.plt.pause(0.001)
        display.clear_output(wait=True)

def train_ch3(net, train_iter, test_iter, loss, num_epochs, updater):  #@save
    """训练模型(定义见第3章)"""
    animator = Animator(xlabel='epoch', xlim=[1, num_epochs], ylim=[0.3, 0.9],
                        legend=['train loss', 'train acc', 'test acc'])
    for epoch in range(num_epochs):
        train_metrics = train_epoch_ch3(net, train_iter, loss, updater)
        test_acc = evaluate_accuracy(net, test_iter)
        # animator.add(epoch + 1, train_metrics + (test_acc,))
        print("loss:{},acc:{}".format(train_metrics[0], train_metrics[1]))
        print("acc:{}".format(test_acc))
    train_loss, train_acc = train_metrics

    # assert train_loss < 0.5, train_loss
    # assert train_acc <= 1 and train_acc > 0.7, train_acc
    # assert test_acc <= 1 and test_acc > 0.7, test_acc

# 小批量随机梯度下降来优化模型的损失函数
lr = 0.1

def updater(batch_size):
    return d2l.sgd([W, b], lr, batch_size)

# 训练模型10个迭代周期
num_epochs = 3
train_ch3(net, train_iter, test_iter, cross_entropy, num_epochs, updater)
# d2l.plt.show()

# 预测
def predict_ch3(net, test_iter, n=6):  #@save
    """预测标签(定义见第3章)"""
    for X, y in test_iter:
        break
    trues = d2l.get_fashion_mnist_labels(y)
    preds = d2l.get_fashion_mnist_labels(net(X).argmax(axis=1))
    print(trues[0:20])
    print(preds[0:20])
    # titles = [true +'\n' + pred for true, pred in zip(trues, preds)]
    # d2l.show_images(
    #     X[0:n].reshape((n, 28, 28)), 1, n, titles=titles[0:n])

predict_ch3(net, test_iter)

标签:从零开始,hat,回归,torch,iter,train,net,0902Softmax,self
来源: https://www.cnblogs.com/g932150283/p/16367971.html