tf.train.exponential_decay 이거


인자들이 각각


▪ learning_rate: A scalar float32 or float64 Tensor or a Python number. The initial learning rate.

▪ global_step : A scalar int32 or int64 Tensor or a Python number. Global step to use for the decay computation. Must not be negative.

▪ decay_steps : A scalar int32 or int64 Tensor or a Python number. Must be positive. See the decay computation above.

▪ decay_rate : A scalar float32 or float64 Tensor or a Python number. The decay rate.

▪ staircase : Boolean. If True decay the learning rate at discrete intervals

▪ name : String. Optional name of the operation. Defaults to 'ExponentialDecay'.



이런데...


어느정도 학습률을 변화시킨단건 알겠는데 결국 언제 변한다는거임?


학습률이 언제 변함?