刘二大人《PyTorch深度学习实践》第2讲

指路 ☞ 《PyTorch深度学习实践》完结合集_哔哩哔哩_bilibili

知识补充：

1、zip(x_data, y_data) 表示将将对象中对应的元素打包成一个个元组，然后返回由这些元组组成的列表。即x_data和y_data一一对应，形成9个坐标

2、append（）表示在列表最后再添加一个元素

a=[1,2,3]

a.append(6)

运行结果为[1,2,3,6]

3、Python enumerate() 函数 | 菜鸟教程

**********************************************************************************************************

import numpy as np
import matplotlib.pyplot as plt

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

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

w_list = []
mse_list = []
for w in np.arange(0.0, 4.1, 0.1):
    print('w', w)
    l_sum = 0
    for x_val, y_val in zip(x_data, y_data):
        y_p_val = forward(x_val)
        loss_val = loss(x_val, y_val)
        l_sum += loss_val
        print('\t', x_val, y_val, y_p_val, l_sum)
    print('MSE =', l_sum / 3)
    w_list.append(w)
    mse_list.append(l_sum / 3)
plt.plot(w_list, mse_list)
plt.xlabel('w')
plt.ylabel('Loss')
plt.show()

部分运行结果如下：

 *****************************************************************************************

作业：实现y=2.5x+1，并输出loss的3D图像

import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D

#这里设函数为y=3x+2
x_data = [1.0,2.0,3.0]
y_data = [5.0,8.0,11.0]


#前向传播
def forward(x):
    return x * w + b

#损失函数
def loss(x,y):
    y_pred = forward(x)
    return (y_pred-y)*(y_pred-y)

l_sum =0
w_list =[]
b_list =[]
mse_list = np.zeros((40, 40), dtype=float)
for i, w in enumerate(np.arange(0.0, 4.0, 0.1)):
  print("w=", w)
  for j, b in enumerate(np.arange(-2.0, 2.0, 0.1)):
     print("b=", b)
     l_sum =0.0
     for x_val, y_val in zip(x_data,y_data):
            y_pred_val = forward(x_val)
            loss_val = loss(x_val, y_val)
            l_sum = l_sum+loss_val
            #print('\t', x_val, y_val, y_pred_val, loss_val)
    # print("MSE=", l_sum/3)
     mse_list[i][j] = l_sum/3
     if w == 0:
        b_list.append(b)
        print('b_list=', b_list)
  w_list.append(w)
  #print('b_list=', b_list)
  print('w_list=', w_list)
x, y = np.meshgrid(w_list, b_list)
z = np.array(mse_list)
fig = plt.figure()
ax = Axes3D(fig)
ax.plot_surface(x, y, z)
plt.show()