一起深度学习

从零实现SoftMax回归

图片分类数据集

import matplotlib.pyplot as plt
import torch
import torchvision
from torch.utils import data
from torchvision import transforms
from d2l import torch as d2l
import os
d2l.use_svg_display()
trans = torchvision.transforms.ToTensor()
mnist_train = torchvision.datasets.FashionMNIST(root="/data",train=True,transform=trans,download=True)
mnist_test = torchvision.datasets.FashionMNIST(root="/data",train=False,transform=trans,download=True)
# print(len(mnist_train))
# print(len(mnist_test))
# print(mnist_train[0][0].shape) #拿到数据集中的第一个样本


# 定义两个函数来画数据集
def get_fashion_mnist_labels(labels):  # 返回FashionMNIST数据集的文本标签
    text_labels = ['t-shirt', 'trouser', 'pullover', 'dress', 'coat', 'sandal', 'shirt', 'sneaber', 'bag', 'ankle boot']
    return [text_labels[int(i)] for i in labels]


def show_images(imgs, num_rows, num_cols, titles=None, scale=1.5):  # 画图
    figsize = (num_cols * scale, num_rows * scale)  # 图片尺寸
    _, axes = d2l.plt.subplots(num_rows, num_cols, figsize=figsize)  # subplots函数绘制多个子图
    axes = axes.flatten()  # flatten函数返回一个一维数组
    for i, (ax, img) in enumerate(zip(axes, imgs)):  # emuerate函数：列举，for循环中得到计数
        ax.set_title(titles[i])  # 显示图片对应名称
        if torch.is_tensor(img):  # 图片张量
            ax.imshow(img.numpy())  # imshow( )函数功能就是把你刚才载入的图片显示出来。
        else:
            ax.imshow(img)  # PIL图片

#仅为测试所用
# X,Y =next(iter(data.DataLoader(mnist_train,batch_size=18)))
# show_images(X.reshape(18,28,28),2,9,titles=get_fashion_mnist_labels(Y))
# d2l.plt.show()

batch_size = 256
def get_dataloader_workers():
    return 4
train_iter = data.DataLoader(mnist_train,batch_size,shuffle=True,num_workers=get_dataloader_workers())
timer = d2l.Timer()
for X,y in train_iter:
    continue
print(f'{timer.stop():.2f}sec')

def load_data_fashion_mnist(batch_size,resize=None):
    trans = [transforms.ToTensor()]
    if resize:
        trans.insert(0,transforms.Resize(resize))
    trans = transforms.Compose(trans)
    mnist_train = torchvision.datasets.FashionMNIST(root="/data", train=True, transform=trans, download=True)
    mnist_test = torchvision.datasets.FashionMNIST(root="/data", train=False, transform=trans, download=True)
    return (data.DataLoader(mnist_train,batch_size,shuffle=True,num_workers=get_dataloader_workers()),
            data.DataLoader(mnist_test,batch_size,shuffle=True,num_workers=get_dataloader_workers()))

从零实现：

import torch
import torchvision
import matplotlib.pyplot as plt
from IPython import display
from d2l import torch as d2l
from torch.utils.data import DataLoader
import load
batch_size = 16
# train_iter,test_iter = d2l.load_data_fashion_mnist(batch_size) # 每次随机读取256张图片
train_iter,test_iter = load.load_data_fashoin_mnist(batch_size) # 每次随机读取256张图片
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函数
def softMax(X):
    X_exp = torch.exp(X)
    partition = X_exp.sum(dim=1,keepdim=True)
    return X_exp/partition
# #验证softmax函数
# X = torch.normal(0,1,(2,5)) #均值为0，标准差为1的正太分布，且为2行5列
# X_prob = softMax(X)
# print(X_prob)
# print(X_prob.sum(1))  #按照列进行相加，若为1，则正确

#实现softmax回归模型
def net(X):
    return softMax(torch.matmul(X.reshape((-1,w.shape[0])),w)+b)
# #两个真实的样本对应的类别
y = torch.tensor([0,2])
#两个预测的样本，以及每个样本中三个类别对应的预测值
y_hat = torch.tensor([[0.1,0.3,0.6],[0.2,0.3,0.5]])
#取出y_hat[0,0],也就是取出预测中的样本的第一个且对应的第一个类别的预测值,再取出y_hat[1,2]
# print(y_hat[[0,1],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(cross_entropy(y_hat,y))  #

def accuracy(y_hat,y):
    #判断是否为二维张量，若是二维则代表呢有多个样本，则取出其中最大的预测值，
    # 取出全部样本对应类别的预测值之后形成一个一维的张量
    if len(y_hat.shape)>1 and y_hat.shape[1] >1:
        y_hat = y_hat.argmax(axis=1) #取出行中，预测值最大的
    #将y_hat的类型转成Y的类型
    cmp = y_hat.type(y.dtype) ==y
    return float(cmp.type(y.dtype).sum()) #转成与y相同的类型并进行求和
# print(accuracy(y_hat,y)/len(y))


# Accumulater实例中创建了2个变量，用于分别储存正确预测数量和预测的总数量

class Accumulater:
    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 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()  # 将模型设为评估模式
    metric = Accumulater(2)  # 分类正确的累加器，这里是添加两个
    with torch.no_grad():  #torch.no_grad()能够关闭梯度计算，避免梯度计算
        for X, y in data_iter:
            #对准确率和样本数量进行累加
            metric.add(accuracy(net(X), y), y.numel())
    return metric[0] / metric[1]  #总预测率滁样本总量，得预测率的平均值


def train_epoch_ch3(net, train_iter, loss, updater):  # 训练模型一个迭代周期
    if isinstance(net, torch.nn.Module):
        net.train()  # 将模型设为训练模式
    # 训练损失总和、训练准确度总和、样本数
    metric = Accumulater(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:
            # 使用定制的损失器和优化函数
            l.sum().backward()
            updater(X.shape[0])
        metric.add(float(l.sum()), accuracy(y_hat, y), y.numel())
    return metric[0] / metric[2], metric[1] / metric[2]  # 返回训练损失和损失精度


# 在展⽰训练函数的实现之前，定义⼀个在动画中绘制数据的实⽤程序类Animator，它能够简化本书其 余部分的代码
class Animator:
    def __init__(self, xlabel=None, ylabel=None, legend=None, xlim=None, ylim=None, xscale='linear',
                 yscale='linear', fmts=('-', 'm--', 'g-.', 'r:'), nrows=1, nclos=1, figsize=(3.5, 2.5)):
        # 增量地绘制多条曲线
        if legend is None:
            legend = []
        d2l.use_svg_display()
        self.fig, self.axes = d2l.plt.subplots(nrows, nclos, figsize=figsize)
        if nrows * nclos == 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()
        display.clear_output(wait=True)


# 训练
def train_ch3(net, train_iter, test_iter, loss, num_epochs, updater):
    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,))
    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, train_acc
    d2l.plt.show()

lr = 0.1


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


num_epochs = 10
train_ch3(net, train_iter, test_iter, cross_entropy, num_epochs, updater)
plt.show()


# 预测
def predict_ch3(net, test_iter, n=6):
    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))
    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])
    plt.show()



predict_ch3(net, test_iter)