You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.

rein.rst 855 B

4 years ago
4 years ago
4 years ago
123456789101112131415161718192021222324252627282930313233343536
  1. API - Reinforcement Learning
  2. ==============================
  3. We provide two reinforcement learning libraries:
  4. - `RL-tutorial <https://github.com/tensorlayer/tensorlayer/tree/master/examples/reinforcement_learning>`__ for professional users with low-level APIs.
  5. - `RLzoo <https://rlzoo.readthedocs.io/en/latest/>`__ for simple usage with high-level APIs.
  6. .. automodule:: tensorlayer.rein
  7. .. autosummary::
  8. discount_episode_rewards
  9. cross_entropy_reward_loss
  10. log_weight
  11. choice_action_by_probs
  12. Reward functions
  13. ---------------------
  14. .. autofunction:: discount_episode_rewards
  15. Cost functions
  16. ---------------------
  17. Weighted Cross Entropy
  18. ^^^^^^^^^^^^^^^^^^^^^^^^
  19. .. autofunction:: cross_entropy_reward_loss
  20. Log weight
  21. ^^^^^^^^^^^^^^
  22. .. autofunction:: log_weight
  23. Sampling functions
  24. ---------------------
  25. .. autofunction:: choice_action_by_probs

TensorLayer3.0 是一款兼容多种深度学习框架为计算后端的深度学习库。计划兼容TensorFlow, Pytorch, MindSpore, Paddle.