Could you please add the way to deal with rewards with same steps in a multi processes training?

Question

Could you please add the way to deal with rewards with same steps in a multi processes training?

Opened this issue a year ago · 2 comments

In my training process, I use a multi-processes PPO.
When I want to draw the reward curve with rl-plotter, I found that:

Just like the image, there are rewards with the same steps. But it seems that it only shows one point in the curve?

Answer 1 · 2024-03-23T11:55:49.000Z

Hi, I also meet the same challenge, have you solved it?

Answer 2 · 2024-03-23T12:53:54.000Z

Hi, I also meet the same challenge, have you solved it?

No... I didn't manage to solve the problem because it seems that the curve looks no problem..