刘二大人《Pytorch深度学习与实践》03梯度下降算法

1、梯度下降与随机梯度下降，其中梯度下降被隐藏起来了，只要把"""删除掉就可以运行

import matplotlib.pyplot as plt

x_data = [1.0, 2.0, 3.0]
y_data = [2.0, 4.0, 6.0]
w = 1.0
epoch_list = []
cost_list = []


def forward(x):
    return x * w


"""
def cost(x1, y1):
    cost = 0
    for x, y in zip(x1, y1):
        y_pred = forward(x)
        cost += (y_pred - y) ** 2
    cost_list.append(cost)
    # 除以样本数量求均值
    return cost / len(x1)


def gradient(x1, y1):
    grad = 0
    for x, y in zip(x1, y1):
        # 损失函数对w的求导，cost=(x*w-y)**2
        grad += 2 * x * (x * w - y)
    # 返回整个验证集中所产生的所有梯度迭代值
    return grad / len(x1)


print('predict(before training)', 4, forward(4))

# 开始迭代
for epoch in range(100):
    # 损失验证值和梯度验证值，其中损失验证值是为了画图
    cost_val = cost(x_data, y_data)
    grad_val = gradient(x_data, y_data)
    # 更新权重值，设定学习率为0.01
    w -= 0.05 * grad_val
    print('epoch:', epoch, 'w:', w, 'loss:', cost_val)
    epoch_list.append(epoch)
    """
"""随机梯度下降"""


def loss(x, y):
    y_pred = forward(x)
    loss1 = 0
    loss1 += (y_pred - y) ** 2
    cost_list.append(loss1)
    return (y_pred - y) ** 2


def gradient(x, y):
    return 2 * x * (x * w - y)


print('predict(before training)', 4, forward(4))
for epoch in range(100):
    for x, y in zip(x_data, y_data):
        grad = gradient(x, y)
        w = w - 0.08 * grad
        print('\tgrad:', x, y, grad)
        l = loss(x, y)
        epoch_list.append(epoch)
    print('progress；', epoch, 'w=', w, 'loss:', l)
print('predict(after training)', 4, forward(4))
plt.plot(epoch_list, cost_list)
plt.ylabel('cost')
plt.xlabel('epoch')
plt.show()

2、梯度求导

import torch

A = torch.arange(4., requires_grad=True)  # 将A的梯度存储到grad这里,[0,1,2,3]，将A设置为可导
B = 2 * torch.dot(A, A)  # 将A与A内积，再乘以2；即2乘以x的平方
B.backward()  # 利用反向传播函数进行梯度求导，应为4乘以x
print(A)
print(B)
print(A.grad)  # 对A求导
B = A.sum()
A.grad.zero_()  # 为什么这个不添加的话，A.grad的输出是([1,5,9,13])
B.backward()
print(B)
print(A.grad)  # 查看对A求导得结果

x = torch.tensor([1., 2., 3.], requires_grad=True)
y = torch.pow(x, 2)

gradient = torch.tensor([1.0, 1.0, 0.004])# 与上述X变量的格式一一对应，将[1,2,3]的权重分别设置为1,1,0.004
y.backward(gradient)
print(x.grad)