This can't be done now since algorithm does not own sampler.
贡献者指南
技术栈
python
领域
machine learning
议题类型
feature
难度面向新贡献者的预计实现难度,1 表示很小改动,5 表示专家级工作。
3
预计时间有经验贡献者完成调查、实现、测试并准备 pull request 的粗略时间范围。
1-2 days
活动状态议题当前的可参与程度:新鲜、活跃、陈旧、阻塞或等待维护者输入。
stale
清晰度议题是否清楚说明期望改动、验收标准和下一步。
mostly clear
前置要求
Understanding of RL2 algorithmKnowledge of gym environment wrappingFamiliarity with garage's sampler module
新手友好度1-100 的估计分数,表示该议题对首次贡献者的友好程度。
30
研究方向
Investigate the current relationship between the RL2 algorithm and the sampler in the garage codebase. Look at files such as garage/sampler/ and garage/algos/rl2.py to understand how env wrapping is handled. Determine if ownership of the sampler can be transferred to the algorithm or if a wrapper can be applied at a different level.