我正在嘗試編寫自定義學(xué)習(xí)率調(diào)度程序：帶預(yù)熱的余弦退火。但我既不能在 Keras 中使用它，也不能在 Tensorflow 中使用它。下面是代碼：import tensorflow as tfimport numpy as npdef make_linear_lr(min_lr, max_lr, number_of_steps):    def gen_lr(step):        return (max_lr - min_lr) / number_of_steps * step + min_lr    return gen_lrdef make_cosine_anneal_lr(learning_rate, alpha, decay_steps):    def gen_lr(global_step):        global_step = min(global_step, decay_steps)        cosine_decay = 0.5 * (1 + np.cos(np.pi * global_step / decay_steps))        decayed = (1 - alpha) * cosine_decay + alpha        decayed_learning_rate = learning_rate * decayed        return decayed_learning_rate    return gen_lrdef make_cosine_annealing_with_warmup(min_lr, max_lr, number_of_steps, alpha, decay_steps):    gen_lr_1 = make_linear_lr(min_lr, max_lr, number_of_steps)    gen_lr_2 = make_cosine_anneal_lr(max_lr, alpha, decay_steps)    def gen_lr(global_step):        if global_step < number_of_steps:            return gen_lr_1(global_step)        else:            return gen_lr_2(global_step - number_of_steps)            return gen_lrclass CosineAnnealingWithWarmUP(tf.keras.optimizers.schedules.LearningRateSchedule):  def __init__(self, min_lr, max_lr, number_of_steps, alpha, decay_steps):    super(CosineAnnealingWithWarmUP, self).__init__()    self.gen_lr_ca =  make_cosine_annealing_with_warmup(min_lr, max_lr, number_of_steps, alpha, decay_steps)  def __call__(self, step):    return tf.cast(self.gen_lr_ca(step), tf.float32)learning_rate_fn = CosineAnnealingWithWarmUP(.0000001, 0.01, 10_000, 0, 150_000)optimizer=tf.keras.optimizers.SGD(    learning_rate=learning_rate_fn, 當(dāng)我嘗試將它與 TensorFlow 一起使用時(shí)，在 get_model_train_step_function 中傳遞優(yōu)化器 - 如果我刪除 @tf.function 裝飾器，它就會(huì)起作用。但它不適用于@tf.function，錯(cuò)誤顯示：OperatorNotAllowedInGraphError: using a tf.Tensoras a Python boolis not allowed: AutoGraph did conversion this function。這可能表明您正在嘗試使用不受支持的功能。我應(yīng)該如何編寫自定義學(xué)習(xí)率調(diào)度程序？另外，我想將此時(shí)間表與 Keras 一起使用。但它在那里根本不起作用。