DI-engine
latest

User Guide

  • Installation
  • Quick Start
  • Key Concept
  • Introduction to RL
  • Hands on RL
  • Best Practice
  • API Doc
    • Config
    • Env
    • Policy
    • Model
    • Reward Model
    • League
    • Learner
    • Collector
    • Buffer
    • Coordinator
    • RL Utils
      • rl_tuils.td
      • rl_utils.gae
      • rl_utils.ppo
      • rl_utils.adder
      • rl_utils.exploration
      • rl_utils.a2c
      • rl_utils.isw
      • rl_tuils.vtrace
      • rl_tuils.value_rescale
      • rl_utils.coma
      • rl_tuils.upgo
    • Torch Utils
    • Utils
    • Interaction
  • FAQ
  • Feature

Developer Guide

  • Developer Guide
  • Tutorial-Developer
  • Architecture Design
DI-engine
  • »
  • API Doc »
  • RL Utils
  • Edit on GitHub

RL UtilsΒΆ

  • rl_tuils.td
    • Temporal Differnece
      • dist_nstep_td_error
      • q_nstep_td_error
      • q_nstep_td_error_with_rescale
      • td_lambda_error
      • generalized_lambda_returns
      • multistep_forward_view
  • rl_utils.gae
    • gae
      • gae
  • rl_utils.ppo
    • ppo
      • ppo_error
  • rl_utils.adder
    • adder
      • Adder
  • rl_utils.exploration
    • exploration
      • get_epsilon_greedy_fn
      • BaseNoise
      • GaussianNoise
      • OUNoise
      • create_noise_generator
  • rl_utils.a2c
    • a2c
      • a2c_error
  • rl_utils.isw
    • isw
      • compute_importance_weights
  • rl_tuils.vtrace
    • vtrace
      • vtrace_error
  • rl_tuils.value_rescale
    • value_rescale
      • value_transform
      • value_inv_transform
  • rl_utils.coma
    • coma
      • coma_error
  • rl_tuils.upgo
    • UPGO
      • upgo_returns
      • upgo_loss
Next Previous

© Copyright 2021, OpenDILab Contributors. Revision ce42f771.

Built with Sphinx using a theme provided by Read the Docs.