memorax.utils.generalized_advantage_estimatation

memorax.utils.generalized_advantage_estimatation#

memorax.utils.generalized_advantage_estimatation(gamma, gae_lambda, final_value, transitions)[source]#

Compute Generalized Advantage Estimates (GAE) for a trajectory.

Parameters:
  • gamma (float)

  • gae_lambda (float)

  • final_value (Array)