PyTorch深度学习实践 4.反向传播

两层计算图

反向传播

import numpy as np
import matplotlib.pyplot as plt
import torch

x_data = [1.0,2.0,3.0]
y_data = [2.0,4.0,6.0]

w = torch.Tensor([1.0])#初始权值
w.requires_grad = True

def forward(x):
    return x * w

def loss(x,y):
    y_pred=forward(x)
    return (y_pred-y)**2

print('Prediciton(before training)',4,forward(4).item())# 直接拿出来，变成一个标量

for epoch in range(100):
    for x,y in zip(x_data,y_data):
        l=loss(x,y) # forward
        l.backward() # 计算  .requires_grad=True 的梯度
        print('\tgrad:',x,y,w.grad.item())
        w.data=w.data-0.01*w.grad.data # utilize梯度来更新w，只是修改数值，用data

        w.grad.data.zero_() # 梯度需要清零，否则会累加

    print('progress:',epoch,'loss=',l.item())
print('Prediciton(before training)',4,forward(4).item())

exercise

import numpy as np
import matplotlib.pyplot as plt
import torch

x_data = [1.0,2.0,3.0]
y_data = [2.0,4.0,6.0]

w1 = torch.Tensor([1.0])#初始权值
w1.requires_grad = True

w2 = torch.Tensor([1.0])
w2.requires_grad = True

b = torch.Tensor([1.0])
b.requires_grad = True

def forward(x):
    return w1*x*x+w2*x+b

def loss(x,y):
    y_pred=forward(x)
    return (y_pred-y)**2

print('Prediciton(before training)',4,forward(4).item())# 直接拿出来，变成一个标量

for epoch in range(10000):
    for x,y in zip(x_data,y_data):
        l=loss(x,y) # forward
        l.backward() # 计算  .requires_grad=True 的梯度
        print('\tx:',x,'y:',y,'w1.grad:',w1.grad.item(),'w2.grad:',w2.grad.item(),'b.grad:',b.grad.item())
        w1.data=w1.data-0.01*w1.grad.data # utilize梯度来更新w，只是修改数值，用data
        w2.data = w2.data - 0.01 * w2.grad.data
        b.data = b.data - 0.01 * b.grad.data

        w1.grad.data.zero_() # 梯度需要清零，否则会累加
        w2.grad.data.zero_()
        b.grad.data.zero_()

    print('progress:',epoch,'loss=',l.item())
print('Prediciton(before training)',4,forward(4).item())

次数	Prediction
500	8.43
1000	8.33
5000	8.03
10000	8.00