Ziwen Zhuang

Something confused me. During training, it seems the VAEWrapperedEnv object's `__goal_sampling_mode` has never been to 'custom_goal_sampler', why it suddenly changed into 'custom_goal_sampler' during loading? BTW, I'm running using sawyer_door.py. Edit:...

So, there are some 'software engineering' problems. If I'm going to implement an "oracle goal sampler", what is the required output in order to match the repository interface?

Considering HER is based on goal-conditioned RL, I believe it is better to implement some plugins as multi-modal RL first. And both "achieved_goal" and "desired_goal" can be a field in...

I have implemented HER by adding some multi-modal interface in the repo, please check out my implementation at https://github.com/ZiwenZhuang/rlpyt/tree/HER/rlpyt/projects/goalRL There is a Readme waiting for you. 😄

This should be the early termination of the subprocess which is running the function `get_example_outputs` during sampler initialization. I'm not sure why the subprocess must be True during the initialization....

老哥，你的wsl有配置图形化界面嘛？我记得GL是跟图形化界面有关的库

It is automatically updated by isaacgym while doing simulation, which is the same as `self.all_root_states`.

By checking the built-in config parsing implementation, the "contact_collection" is already set to `2` by default. I'm afraid that you didn't change anything. > Actually, I was able to get...

Hi, due to isaacgym overrides the cuda backend of pytorch, it probabily skips some validity check when you operating tenors. Could you locate which line of code that triggers the...

Hmmm, line 330 in `legged_robot_field.py` is the process of translating `obs_scale` in to tensor if needed. This cannot be the code that triggers the error. For the out-of-memory error, maybe...