2.3.2 训练模式和预测模式 有 关键性错误

[xinetzone]

with tf.GradientTape(persistent=True) as g:
    g.watch(x)
    y = x * x
    z = y * y
    dz_dx = g.gradient(z, x)  # 108.0 (4*x^3 at x = 3)
    dy_dx = g.gradient(y, x)  # 6.0
dz_dx,dy_dx

需要改为：

with tf.GradientTape(persistent=True) as g:
    g.watch(x)
    y = x * x
    z = y * y
dz_dx = g.gradient(z, x)  # 108.0 (4*x^3 at x = 3)
dy_dx = g.gradient(y, x)  # 6.0
dz_dx,dy_dx