Ericonaldo / ILSwiss

ILSwiss is an Easy-to-run Imitation Learning (IL, or Learning from Demonstration, LfD) and also Reinforcement Learning (RL) framework (template) in PyTorch.
MIT License
161 stars 12 forks source link

Where can I get the "Random" results shown in the readme? #18

Closed SunJiaJ closed 1 year ago

SunJiaJ commented 1 year ago
Ericonaldo commented 1 year ago
  1. Random就是不用训练策略的初始化网络直接跑出来的分数
  2. 我们在这个框架下发表了一些论文,例如COIL, DePO, EBIL(老版本的ILSwiss,很多不一致的地方)等。此外,正如README所说的,这份代码基于rlswiss和rlkit,所以基于这两套代码的论文的code风格也有很多兼容的部分。