PyTorch深度学习实践-P3梯度下降算法

• 又名贪心算法
• 上一讲的w用穷举法来找出，但如果w有两个，或者不是线性时，就很难找
• 
• 要找w 使得cost（w）最小
• 

 

 

•  求出了w的更新公式~
• 代码：

• import matplotlib.pyplot as plt
  
  # prepare the training set
  x_data = [1.0, 2.0, 3.0]
  y_data = [2.0, 4.0, 6.0]
  
  # initial guess of weight
  w = 1.0
  
  
  # define the model linear model y = w*x
  def forward(x):
      return x * w
  
  
  # define the cost function MSE
  def cost(xs, ys):
      cost = 0
      for x, y in zip(xs, ys):
          y_pred = forward(x)
          cost += (y_pred - y) ** 2
      return cost / len(xs)
  
  
  # define the gradient function  gd
  def gradient(xs, ys):
      grad = 0
      for x, y in zip(xs, ys):
          grad += 2 * x * (x * w - y)
      return grad / len(xs)
  
  
  epoch_list = []
  cost_list = []
  print('predict (before training)', 4, forward(4))
  for epoch in range(100):
      cost_val = cost(x_data, y_data)
      grad_val = gradient(x_data, y_data)
      w -= 0.01 * grad_val  # 0.01 learning rate
      print('epoch:', epoch, 'w=', w, 'loss=', cost_val)
      epoch_list.append(epoch)
      cost_list.append(cost_val)
  
  print('predict (after training)', 4, forward(4))
  plt.plot(epoch_list, cost_list)
  plt.ylabel('cost')
  plt.xlabel('epoch')
  plt.show()

  

   Stochastic Gradient Descent:随机梯度下降

• 不同之处：cost 变成了loss
• 

 代码：

import matplotlib.pyplot as plt

# prepare the training set
x_data = [1.0, 2.0, 3.0]
y_data = [2.0, 4.0, 6.0]

# initial guess of weight
w = 1.0


# define the model linear model y = w*x
def forward(x):
    return x * w


# define the cost function MSE
def loss(x, y):
    y_pred = forward(x)
    return (y_pred - y) ** 2


# define the gradient function  gd
def gradient(x, y):
    return x*x*(x*w-y)

epoch_list = []
cost_list = []
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 - 0.01 * grad
    print("\tgrad: ", x, y, grad)
    l = loss(x, y)

    print("progress:", epoch, "w=", w, "loss=", l)
    epoch_list.append(epoch)
    cost_list.append(l)

print('predict (after training)', 4, forward(4))
plt.plot(epoch_list, cost_list)
plt.ylabel('cost')
plt.xlabel('epoch')
plt.show()

 

 

 progress: 97 w= 1.9999999999999596 loss= 1.4590771302967834e-26
    grad:  3.0 6.0 -1.0871303857129533e-12
progress: 98 w= 1.9999999999999705 loss= 7.888609052210118e-27
    grad:  3.0 6.0 -7.993605777301127e-13
progress: 99 w= 1.9999999999999785 loss= 4.203839763922772e-27
predict (after training) 4 7.999999999999914