I am using Ray's PPO algorithm, and see in PPOConfig that the training() function accepts a learning rate schedule as lr_schedule:
From the Docs:
Args:
lr_schedule: Learning rate schedule. In the format of
[[timestep, lr-value], [timestep, lr-value], ...]
Intermediary timesteps will be assigned to interpolated learning rate
values. A schedule should normally start from timestep 0.
Is there a readily accessible function I can use to generate the schedule? I am using Ray with PyToch, in case that helps.
Ray version: 2.9.3