AttributeError: module 'tensorflow._api.v2.train' has no attribute 'exponential_decay'

This error occurs when you try to use the `exponential_decay` function from the `train` module in TensorFlow 2.x, but it is not available in that module anymore.

Instead, you can use the `tf.keras.optimizers.schedules.ExponentialDecay` class to achieve similar functionality. Here's an example:

initial_learning_rate = 0.1
decay_steps = 10000
decay_rate = 0.96

lr_schedule = tf.keras.optimizers.schedules.ExponentialDecay(
    initial_learning_rate,
    decay_steps=decay_steps,
    decay_rate=decay_rate)

optimizer = tf.keras.optimizers.SGD(learning_rate=lr_schedule)

This creates an exponentially decaying learning rate schedule that starts at 0.1 and decays by a factor of 0.96 every 10,000 steps. You can then pass this schedule to your optimizer.