buzinguyen's picture
Add the trained policies (checkpoints/, archive, manifest)
c923466 verified
Raw History Blame Contribute Delete
700 Bytes
adaptive_lr: false
adversary: true
desired_kl: 0.01
device: cuda:0
dstb_pretrain: 20
end_criterion: null
ent_coef: 0.001
force_floor: 0.3
force_init: 0.5
force_max: 25.0
force_ramp_frac: 0.55
fwd_force: 15.0
gamma: 0.99
gamma_anneal: false
l_anneal_steps: 0
l_hold_steps: 20000000
leg_death: false
load: checkpoints/getup_stage1/final/final_model.zip
load_tensornorm: null
lr: 0.0001
max_std: 0.4
net: 128,128,128
no_wandb: true
norm_freeze_steps: 5000000
num_envs: 1024
out: runs/getup
reset_log_std: null
reset_value: false
seed: 0
std_ceil: null
std_floor: null
steps: 50000000
target_kl: 0.01
task: go2_getup
terminal_type: all
vf_coef: 0.5
video_interval: 5000000
wandb_project: odd-conditioned