Hello, I'm a newbie in the usage of the IsaacGymEnvs package. I can get an optimal policy by training the policy in the train mode. However, when I use test mode, I cannot change maximum iterations even when I try to change max_iterations in the cfg/config.yaml or add max_iterations=300000 in the command line.
I want to ask, how to change the maximum iterations in the "test" mode after obtaining the optimal policy?
this is my command line
python3 train.py task=AnymalTerrainLF test=True checkpoint=./best_1004.pth num_envs=100 headless=True max_iterations=300000
when I try to change the value of max_iterations, the simulation always stops after 5-10 min. I tried to print the step in post_physics_step(), but it always stopped in the 1700 steps.
Hello, I'm a newbie in the usage of the IsaacGymEnvs package. I can get an optimal policy by training the policy in the
train
mode. However, when I usetest
mode, I cannot change maximum iterations even when I try to changemax_iterations
in thecfg/config.yaml
or addmax_iterations=300000
in the command line.I want to ask, how to change the maximum iterations in the "test" mode after obtaining the optimal policy?
this is my command line
python3 train.py task=AnymalTerrainLF test=True checkpoint=./best_1004.pth num_envs=100 headless=True max_iterations=300000
when I try to change the value of
max_iterations
, the simulation always stops after 5-10 min. I tried to print the step inpost_physics_step()
, but it always stopped in the 1700 steps.thank you