本文共 1052 字,大约阅读时间需要 3 分钟。
import numpy as npimport mathimport matplotlib.pyplot as plt# Create random input and output datax = np.linspace(-math.pi, math.pi, 2000)y = np.sin(x)plt.scatter(x,y)plt.show()# Randomly initialize weightsa = np.random.randn()b = np.random.randn()c = np.random.randn()d = np.random.randn()learning_rate = 1e-6for t in range(4000): # Forward pass: compute predicted y # y = a + b x + c x^2 + d x^3 y_pred = a + b * x + c * x**2 + d * x**3 # Compute and print loss loss = np.square(y_pred - y).sum() if t % 100 == 99: print(t, loss) # Backprop to compute gradients of a, b, c, d with respect to loss grad_y_pred = 2.0 * (y_pred - y) grad_a = grad_y_pred.sum() grad_b = (grad_y_pred * x).sum() grad_c = (grad_y_pred * x**2).sum() grad_d = (grad_y_pred * x**3).sum() # Update weights a -= learning_rate * grad_a b -= learning_rate * grad_b c -= learning_rate * grad_c d -= learning_rate * grad_dprint(f'Result: y = {a} + {b} x + {c} x^2 + {d} x^3')
首先记住,权重w更新是 减去 损失函数L 对权重w的求导,即αL/αw
这里a,b,c,d都是权重
转载地址:http://sufgf.baihongyu.com/