AndreaRossetto
AndreaRossetto
I'm facing the same problem with skrl, there is a way to solve this?
My values for the video_length and video_interval are 2000 and 1000 (I use the default ones) ` parser.add_argument("--video_length", type=int, default=2000, help="Length of the recorded video (in steps).") parser.add_argument("--video_interval", type=int, default=1000,...
My system has 32GB of ram and I 137GB of swap memory
I've reduced the video_interval and video_length values to (50 and 500) but over time the memory keep be allocated, slower but the usage keep increase.
I've upgrade my system to 64GB of RAM but the problem still there.... The memory usage whatever I change keep increasing.
Update: For some reason the values went to NaN as shown in Tensorboard:  
https://github.com/user-attachments/assets/2067cc65-f720-40c6-bd66-7aa96f4e2be2
I have done a step forward with the issue. At a certain time during the training the PPO action values are Nan, Why?
Thank you @zoctipus for your suggestion, my dt is 0.005 ( I can try a smaller one). If I decrese the learning rate of the ppo the problems is much...
@zoctipus It seems my network is generating the nan, the rewards seems to be not so high (I took the values from the cassie task): ` lin_vel_z_weight = -2.0 ang_vel_xy_weight...