Pytorch梯度下降——up主：刘二大人《PyTorch深度学习实践》

教程： https://www.bilibili.com/video/BV1Y7411d7Ys?p=2&vd_source=715b347a0d6cb8aa3822e5a102f366fe
数据集：
$$\begin{gathered} x_{data}=[1.0,2.0,3.0] \\ {y}_{data}=[2.0,4.0,6.0] \end{gathered}$$
参数：
$$\begin{gathered} w_{list}=[0.0,4.0,0.1] \\ {b}_{list}=[-2.0,2.1,0.1] \end{gathered}$$
模型：
$$y=w\ast x$$

import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
x_data = [1.0, 2.0, 3.0]
y_data = [2.0, 4.0, 6.0]
epoch_log = []
cost_log = []
#随机化参数
w = 1.0

def forward(x):
  return x * w

#定义损失函数
def loss(xs, ys):
  cost = 0
  for x,y in zip(xs, ys):
    y_pred = forward(x)
    cost += (y_pred - y) **2
  return cost / len(xs)

def gradient(xs, ys):
  grad = 0
  for x, y in zip(xs, ys):
    grad += 2 * x * (x * w - y)
  return grad / len(xs)

for epoch in range(100):
  cost_val = loss(x_data, y_data)
  cost_log.append(cost_val)
  grad_val = gradient(x_data, y_data)
  epoch_log.append(epoch)
  w -= 0.01 * grad_val
  print('Epoch:', epoch, 'w =', w, 'loss =', cost_val)

plt.figure()
plt.plot(epoch_log, cost_log, c='b')
plt.xlabel('Epoch')
plt.ylabel('Cost')
plt.show()