Skip to content

Latest commit

 

History

History
5 lines (5 loc) · 770 Bytes

developping-algorithms.md

File metadata and controls

5 lines (5 loc) · 770 Bytes

Developping Algorithms

Although users may do whatever they like to design and try their algorithms. We recommend wrapping a new algorithm as an Agent class the example agent. To be compatible with the basic interfaces, the agent should have the following functions and attribute:

  • step: Given the current state, predict the next action.
  • eval_step: Similar to step, but for evaluation purpose. Reinforcement learning algorithms will usually add some noise for better exploration in training. In evaluation, no noise will be added to make predictions.
  • use_raw: A boolean attribute. True if the agent uses raw states to do reasoning; False if the agent uses numerical values to play (such as neural networks).