pytorch-03.梯度下降算法(随机梯度下降)

cost_list，grad_list，w_list每个列表中有300个值，因为数据有3个，epoch=100，故计算了300次梯度。
       

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]

w = 1

def forward(x,w):
    return x*w

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


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

if __name__ == '__main__':
    print("Predict (before training)",4,forward(4,w))
    cost_list = []
    w_list = []
    grad_list = []
    for epoch in range(100):
        # cost_val = cost(x_data,y_data,w)
        # grad_val = gradient(x_data,y_data,w)
        # w -= 0.01*grad_val
        for x,y in zip(x_data,y_data):
            grad = gradient(x,y,w)
            w -= 0.01*grad
            print("\tgrad:",grad,"\tx:",x,"\ty:",y)
            loss = cost(x,y,w)
            cost_list.append(loss)
            grad_list.append(grad)
            w_list.append(w)
        print('Epoch:',epoch+1," ",'w=',w," ",'loss=',loss)

    print("Predict (after training)",4,forward(4,w))
    plt.plot(w_list,cost_list)
    plt.xlabel("w")
    plt.ylabel("loss")
    plt.show()

    plt.plot(w_list,grad_list)
    plt.xlabel("w")
    plt.ylabel("grad")
    plt.show()