《动手学深度学习》Pytorch版学习笔记（一）：Task02

课程背景

在疫情的影响下，不少学校已经做出了延迟开学的决定，：伯禹教育、Datawhale、和鲸科技牵头与多家AI企业合作，让在家的同学也能免费学习优质的付费课程，同时为学习者创建好的学习环境，提供就业绿色通道。

主修课程

《动手学深度学习》 代码讲解Pytorch版：该书是2019年国内最受欢迎的人工智能学习教材之一，是一本面向中文读者的能运行、可讨论的深度学习教科书，书籍作者之一亚马逊首席科学家李沐，毕业于上海交大。伯禹教育携手上海交通大学团队，基于此书籍，将其中的代码框架由MXNET迁移至PyTorch，并对这些代码制作了讲解视频。帮助大家边动手写代码边巩固理论知识，从原理到实践，上手深度学习。

1 任务

【第一次打卡】内容（2月11日-14日）
Task01：线性回归；Softmax与分类模型、多层感知机（1天）
Task02：过拟合、欠拟合及其解决方案；梯度消失、梯度爆炸、梯度偏移；卷积神经网络基础（1天）
Task03：LeNet；卷积神经网络进阶；批量归一化和残差网络（1天）
打卡时间：【2020-02-11 08:00 -- 2020-02-14 22:00】
打卡链接：学习开始放出

2 过拟合、欠拟合及其解决方案

2.1 过拟合、欠拟合的概念

这里就不对概念进行展开介绍了，具体概念可见：伯禹学习平台

#多项式函数拟合实验
#导入包
%matplotlib inline
import torch
import numpy as np
import sys
sys.path.append("/home/kesci/input")
import d2lzh1981 as d2l

#初始化模型参数
n_train, n_test, true_w, true_b = 100, 100, [1.2, -3.4, 5.6], 5
features = torch.randn((n_train + n_test, 1))
poly_features = torch.cat((features, torch.pow(features, 2), torch.pow(features, 3)), 1) 
labels = (true_w[0] * poly_features[:, 0] + true_w[1] * poly_features[:, 1]
          + true_w[2] * poly_features[:, 2] + true_b)
labels += torch.tensor(np.random.normal(0, 0.01, size=labels.size()), dtype=torch.float)

#定义、训练和测试模型
def semilogy(x_vals, y_vals, x_label, y_label, x2_vals=None, y2_vals=None,
             legend=None, figsize=(3.5, 2.5)):
    # d2l.set_figsize(figsize)
    d2l.plt.xlabel(x_label)
    d2l.plt.ylabel(y_label)
    d2l.plt.semilogy(x_vals, y_vals)
    if x2_vals and y2_vals:
        d2l.plt.semilogy(x2_vals, y2_vals, linestyle=':')
        d2l.plt.legend(legend)

num_epochs, loss = 100, torch.nn.MSELoss()

def fit_and_plot(train_features, test_features, train_labels, test_labels):
    # 初始化网络模型
    net = torch.nn.Linear(train_features.shape[-1], 1)
    # 通过Linear文档可知，pytorch已经将参数初始化了，所以我们这里就不手动初始化了
    
    # 设置批量大小
    batch_size = min(10, train_labels.shape[0])    
    dataset = torch.utils.data.TensorDataset(train_features, train_labels)      # 设置数据集
    train_iter = torch.utils.data.DataLoader(dataset, batch_size, shuffle=True) # 设置获取数据方式
    
    optimizer = torch.optim.SGD(net.parameters(), lr=0.01)                      # 设置优化函数，使用的是随机梯度下降优化
    train_ls, test_ls = [], []
    for _ in range(num_epochs):
        for X, y in train_iter:                                                 # 取一个批量的数据
            l = loss(net(X), y.view(-1, 1))                                     # 输入到网络中计算输出，并和标签比较求得损失函数
            optimizer.zero_grad()                                               # 梯度清零，防止梯度累加干扰优化
            l.backward()                                                        # 求梯度
            optimizer.step()                                                    # 迭代优化函数，进行参数优化
        train_labels = train_labels.view(-1, 1)
        test_labels = test_labels.view(-1, 1)
        train_ls.append(loss(net(train_features), train_labels).item())         # 将训练损失保存到train_ls中
        test_ls.append(loss(net(test_features), test_labels).item())            # 将测试损失保存到test_ls中
    print('final epoch: train loss', train_ls[-1], 'test loss', test_ls[-1])    
    semilogy(range(1, num_epochs + 1), train_ls, 'epochs', 'loss',
             range(1, num_epochs + 1), test_ls, ['train', 'test'])
    print('weight:', net.weight.data,
          '\nbias:', net.bias.data)

#三阶多项式函数拟合（正常）
fit_and_plot(poly_features[:n_train, :], poly_features[n_train:, :], labels[:n_train], labels[n_train:])

#线性函数拟合（欠拟合）
fit_and_plot(features[:n_train, :], features[n_train:, :], labels[:n_train], labels[n_train:])

#训练样本不足（过拟合）
fit_and_plot(poly_features[0:2, :], poly_features[n_train:, :], labels[0:2], labels[n_train:])

2.2 权重衰减

权重衰减等价于 L2 范数正则化（regularization）。正则化通过为模型损失函数添加惩罚项使学出的模型参数值较小，是应对过拟合的常用手段。

#高维线性回归实验从零开始的实现
#导入包
%matplotlib inline
import torch
import torch.nn as nn
import numpy as np
import sys
sys.path.append("/home/kesci/input")
import d2lzh1981 as d2l

#初始化模型参数
n_train, n_test, num_inputs = 20, 100, 200
true_w, true_b = torch.ones(num_inputs, 1) * 0.01, 0.05

features = torch.randn((n_train + n_test, num_inputs))
labels = torch.matmul(features, true_w) + true_b
labels += torch.tensor(np.random.normal(0, 0.01, size=labels.size()), dtype=torch.float)
train_features, test_features = features[:n_train, :], features[n_train:, :]
train_labels, test_labels = labels[:n_train], labels[n_train:]

# 定义参数初始化函数，初始化模型参数并且附上梯度
def init_params():
    w = torch.randn((num_inputs, 1), requires_grad=True)
    b = torch.zeros(1, requires_grad=True)
    return [w, b]

#定义L2范数惩罚项
def l2_penalty(w):
    return (w**2).sum() / 2

#定义训练和测试
batch_size, num_epochs, lr = 1, 100, 0.003
net, loss = d2l.linreg, d2l.squared_loss

dataset = torch.utils.data.TensorDataset(train_features, train_labels)
train_iter = torch.utils.data.DataLoader(dataset, batch_size, shuffle=True)

def fit_and_plot(lambd):
    w, b = init_params()
    train_ls, test_ls = [], []
    for _ in range(num_epochs):
        for X, y in train_iter:
            # 添加了L2范数惩罚项
            l = loss(net(X, w, b), y) + lambd * l2_penalty(w)
            l = l.sum()
            
            if w.grad is not None:
                w.grad.data.zero_()
                b.grad.data.zero_()
            l.backward()
            d2l.sgd([w, b], lr, batch_size)
        train_ls.append(loss(net(train_features, w, b), train_labels).mean().item())
        test_ls.append(loss(net(test_features, w, b), test_labels).mean().item())
    d2l.semilogy(range(1, num_epochs + 1), train_ls, 'epochs', 'loss',
                 range(1, num_epochs + 1), test_ls, ['train', 'test'])
    print('L2 norm of w:', w.norm().item())

#观察过拟合
fit_and_plot(lambd=0)

#使用权重衰减
fit_and_plot(lambd=3)

#简洁实现
def fit_and_plot_pytorch(wd):
    # 对权重参数衰减。权重名称一般是以weight结尾
    net = nn.Linear(num_inputs, 1)
    nn.init.normal_(net.weight, mean=0, std=1)
    nn.init.normal_(net.bias, mean=0, std=1)
    optimizer_w = torch.optim.SGD(params=[net.weight], lr=lr, weight_decay=wd) # 对权重参数衰减
    optimizer_b = torch.optim.SGD(params=[net.bias], lr=lr)  # 不对偏差参数衰减
    
    train_ls, test_ls = [], []
    for _ in range(num_epochs):
        for X, y in train_iter:
            l = loss(net(X), y).mean()
            optimizer_w.zero_grad()
            optimizer_b.zero_grad()
            
            l.backward()
            
            # 对两个optimizer实例分别调用step函数，从而分别更新权重和偏差
            optimizer_w.step()
            optimizer_b.step()
        train_ls.append(loss(net(train_features), train_labels).mean().item())
        test_ls.append(loss(net(test_features), test_labels).mean().item())
    d2l.semilogy(range(1, num_epochs + 1), train_ls, 'epochs', 'loss',
                 range(1, num_epochs + 1), test_ls, ['train', 'test'])
    print('L2 norm of w:', net.weight.data.norm().item())

2.3 丢弃法

丢弃法从零开始的实现

#丢弃法从零开始的实现
%matplotlib inline
import torch
import torch.nn as nn
import numpy as np
import sys
sys.path.append("/home/kesci/input")
import d2lzh1981 as d2l

def dropout(X, drop_prob):
    X = X.float()
    assert 0 <= drop_prob <= 1
    keep_prob = 1 - drop_prob
    # 这种情况下把全部元素都丢弃
    if keep_prob == 0:
        return torch.zeros_like(X)
    mask = (torch.rand(X.shape) < keep_prob).float()
    
    return mask * X / keep_prob

# 参数的初始化
num_inputs, num_outputs, num_hiddens1, num_hiddens2 = 784, 10, 256, 256

W1 = torch.tensor(np.random.normal(0, 0.01, size=(num_inputs, num_hiddens1)), dtype=torch.float, requires_grad=True)
b1 = torch.zeros(num_hiddens1, requires_grad=True)
W2 = torch.tensor(np.random.normal(0, 0.01, size=(num_hiddens1, num_hiddens2)), dtype=torch.float, requires_grad=True)
b2 = torch.zeros(num_hiddens2, requires_grad=True)
W3 = torch.tensor(np.random.normal(0, 0.01, size=(num_hiddens2, num_outputs)), dtype=torch.float, requires_grad=True)
b3 = torch.zeros(num_outputs, requires_grad=True)

params = [W1, b1, W2, b2, W3, b3]

drop_prob1, drop_prob2 = 0.2, 0.5

def net(X, is_training=True):
    X = X.view(-1, num_inputs)
    H1 = (torch.matmul(X, W1) + b1).relu()
    if is_training:  # 只在训练模型时使用丢弃法
        H1 = dropout(H1, drop_prob1)  # 在第一层全连接后添加丢弃层
    H2 = (torch.matmul(H1, W2) + b2).relu()
    if is_training:
        H2 = dropout(H2, drop_prob2)  # 在第二层全连接后添加丢弃层
    return torch.matmul(H2, W3) + b3

def evaluate_accuracy(data_iter, net):
    acc_sum, n = 0.0, 0
    for X, y in data_iter:
        if isinstance(net, torch.nn.Module):
            net.eval() # 评估模式, 这会关闭dropout
            acc_sum += (net(X).argmax(dim=1) == y).float().sum().item()
            net.train() # 改回训练模式
        else: # 自定义的模型
            if('is_training' in net.__code__.co_varnames): # 如果有is_training这个参数
                # 将is_training设置成False
                acc_sum += (net(X, is_training=False).argmax(dim=1) == y).float().sum().item() 
            else:
                acc_sum += (net(X).argmax(dim=1) == y).float().sum().item() 
        n += y.shape[0]
    return acc_sum / n

num_epochs, lr, batch_size = 5, 100.0, 256  # 这里的学习率设置的很大，原因与之前相同。
loss = torch.nn.CrossEntropyLoss()
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, root='/home/kesci/input/FashionMNIST2065')
d2l.train_ch3(
    net,
    train_iter,
    test_iter,
    loss,
    num_epochs,
    batch_size,
    params,
    lr)

简洁实现

#简洁实现
net = nn.Sequential(
        d2l.FlattenLayer(),
        nn.Linear(num_inputs, num_hiddens1),
        nn.ReLU(),
        nn.Dropout(drop_prob1),
        nn.Linear(num_hiddens1, num_hiddens2), 
        nn.ReLU(),
        nn.Dropout(drop_prob2),
        nn.Linear(num_hiddens2, 10)
        )

for param in net.parameters():
    nn.init.normal_(param, mean=0, std=0.01)

optimizer = torch.optim.SGD(net.parameters(), lr=0.5)
d2l.train_ch3(net, train_iter, test_iter, loss, num_epochs, batch_size, None, None, optimizer)

2.4 总结

1. 欠拟合现象：模型无法达到一个较低的误差；
2. 过拟合现象：训练误差较低但是泛化误差依然较高，二者相差较大。

3 梯度消失、梯度爆炸以及Kaggle房价预测

3.1 梯度消失和梯度爆炸

深度模型有关数值稳定性的典型问题是消失（vanishing） 和 爆炸（explosion）。

当神经网络的层数较多时，模型的数值稳定性容易变差。

3.2 考虑到环境因素的其他问题

1. 协变量偏移
2. 标签偏移
3. 概念偏移

3.3 Kaggle房价预测

#导入包
%matplotlib inline
import torch
import torch.nn as nn
import numpy as np
import pandas as pd
import sys
sys.path.append("/home/kesci/input")
import d2lzh1981 as d2l
print(torch.__version__)
torch.set_default_tensor_type(torch.FloatTensor)

#获取和读取数据集
test_data = pd.read_csv("/home/kesci/input/houseprices2807/house-prices-advanced-regression-techniques/test.csv")
train_data = pd.read_csv("/home/kesci/input/houseprices2807/house-prices-advanced-regression-techniques/train.csv")

#预处理数据
numeric_features = all_features.dtypes[all_features.dtypes != 'object'].index
all_features[numeric_features] = all_features[numeric_features].apply(
    lambda x: (x - x.mean()) / (x.std()))
# 标准化后，每个数值特征的均值变为0，所以可以直接用0来替换缺失值
all_features[numeric_features] = all_features[numeric_features].fillna(0)

# dummy_na=True将缺失值也当作合法的特征值并为其创建指示特征
all_features = pd.get_dummies(all_features, dummy_na=True)
all_features.shape
n_train = train_data.shape[0]
train_features = torch.tensor(all_features[:n_train].values, dtype=torch.float)
test_features = torch.tensor(all_features[n_train:].values, dtype=torch.float)
train_labels = torch.tensor(train_data.SalePrice.values, dtype=torch.float).view(-1, 1)

#训练模型
loss = torch.nn.MSELoss()

def get_net(feature_num):
    net = nn.Linear(feature_num, 1)
    for param in net.parameters():
        nn.init.normal_(param, mean=0, std=0.01)
    return net

def log_rmse(net, features, labels):
    with torch.no_grad():
        # 将小于1的值设成1，使得取对数时数值更稳定
        clipped_preds = torch.max(net(features), torch.tensor(1.0))
        rmse = torch.sqrt(2 * loss(clipped_preds.log(), labels.log()).mean())
    return rmse.item()

def train(net, train_features, train_labels, test_features, test_labels,
          num_epochs, learning_rate, weight_decay, batch_size):
    train_ls, test_ls = [], []
    dataset = torch.utils.data.TensorDataset(train_features, train_labels)
    train_iter = torch.utils.data.DataLoader(dataset, batch_size, shuffle=True)
    # 这里使用了Adam优化算法
    optimizer = torch.optim.Adam(params=net.parameters(), lr=learning_rate, weight_decay=weight_decay) 
    net = net.float()
    for epoch in range(num_epochs):
        for X, y in train_iter:
            l = loss(net(X.float()), y.float())
            optimizer.zero_grad()
            l.backward()
            optimizer.step()
        train_ls.append(log_rmse(net, train_features, train_labels))
        if test_labels is not None:
            test_ls.append(log_rmse(net, test_features, test_labels))
    return train_ls, test_ls

#K折交叉验证
def get_k_fold_data(k, i, X, y):
    # 返回第i折交叉验证时所需要的训练和验证数据
    assert k > 1
    fold_size = X.shape[0] // k
    X_train, y_train = None, None
    for j in range(k):
        idx = slice(j * fold_size, (j + 1) * fold_size)
        X_part, y_part = X[idx, :], y[idx]
        if j == i:
            X_valid, y_valid = X_part, y_part
        elif X_train is None:
            X_train, y_train = X_part, y_part
        else:
            X_train = torch.cat((X_train, X_part), dim=0)
            y_train = torch.cat((y_train, y_part), dim=0)
    return X_train, y_train, X_valid, y_valid

def k_fold(k, X_train, y_train, num_epochs,
           learning_rate, weight_decay, batch_size):
    train_l_sum, valid_l_sum = 0, 0
    for i in range(k):
        data = get_k_fold_data(k, i, X_train, y_train)
        net = get_net(X_train.shape[1])
        train_ls, valid_ls = train(net, *data, num_epochs, learning_rate,
                                   weight_decay, batch_size)
        train_l_sum += train_ls[-1]
        valid_l_sum += valid_ls[-1]
        if i == 0:
            d2l.semilogy(range(1, num_epochs + 1), train_ls, 'epochs', 'rmse',
                         range(1, num_epochs + 1), valid_ls,
                         ['train', 'valid'])
        print('fold %d, train rmse %f, valid rmse %f' % (i, train_ls[-1], valid_ls[-1]))
    return train_l_sum / k, valid_l_sum / k

#模型选择
k, num_epochs, lr, weight_decay, batch_size = 5, 100, 5, 0, 64
train_l, valid_l = k_fold(k, train_features, train_labels, num_epochs, lr, weight_decay, batch_size)
print('%d-fold validation: avg train rmse %f, avg valid rmse %f' % (k, train_l, valid_l))

#预测并在Kaggle中提交结果
def train_and_pred(train_features, test_features, train_labels, test_data,
                   num_epochs, lr, weight_decay, batch_size):
    net = get_net(train_features.shape[1])
    train_ls, _ = train(net, train_features, train_labels, None, None,
                        num_epochs, lr, weight_decay, batch_size)
    d2l.semilogy(range(1, num_epochs + 1), train_ls, 'epochs', 'rmse')
    print('train rmse %f' % train_ls[-1])
    preds = net(test_features).detach().numpy()
    test_data['SalePrice'] = pd.Series(preds.reshape(1, -1)[0])
    submission = pd.concat([test_data['Id'], test_data['SalePrice']], axis=1)
    submission.to_csv('./submission.csv', index=False)
    # sample_submission_data = pd.read_csv("../input/house-prices-advanced-regression-techniques/sample_submission.csv")