Prabhasak's workspace
Runs
56
State
Notes
User
Tags
Created
Runtime
Sweep
algo
batch_size
buffer_size
cg_damping
cg_iters
check_callback
cliprange
device
ent_coef
entcoeff
env
eval_callback
exp_id
gamma
gradient_steps
lam
learning_rate
learning_starts
max_kl
n_steps
n_timesteps
nminibatches
noptepochs
policy
save_best_model
seed
tensorboard
timesteps_IL
timesteps_RL
timesteps_per_batch
train_IL
train_RL
train_freq
traj_use
verbose
vf_iters
vf_stepsize
wandb_log
num_trajs
BC_max_iter
checkpoint_dir
env_id
env_kwargs.expert
env_kwargs.name
Finished
-
prabhasak
7d 7h 50m 30s
-
["sac","trpo"]
256
1000000
0.1
10
false
-
gpu
auto
0
AirSim-v0
true
1.45455
0.99
1
0.98
lin_3e-4
10000
0.01
-
-
-
-
["CustomSACPolicy","MlpPolicy"]
true
42
true
[0,"1e6"]
0
1024
[false,true]
[false,true]
1
-
0
5
0.001
true
9.54545
["1e6","5e5"]
models
AirSim-v0
["bochan","prabhasa"]
["Drone2","Drone3"]
Finished
-
prabhasak
3d 18h 26m 41s
-
["sac","trpo"]
256
1000000
0.1
15
false
-
gpu
auto
0
HalfCheetah-v2
true
1.66667
0.99
1
0.95
0.0003
10000
0.01
-
-
-
-
["CustomSACPolicy","MlpPolicy"]
true
42
true
0
0
2048
[false,true]
[false,true]
1
-
0
5
0.001
true
8.33333
5e5
models
HalfCheetah-v2
-
-
Finished
-
prabhasak
7d 2h 53m 56s
-
["sac","td3"]
208
-
0.1
10
false
-
gpu
-
0
LunarLanderContinuous-v2
true
1.1875
0.9925
-
0.98
-
1000
0.01
-
-
-
-
MlpPolicy
true
42
true
[0,"2e6"]
0
1024
[false,true]
[false,true]
-
-
0
5
0.001
true
11.5625
[100000,"1e5","1e6","3e5"]
models
LunarLanderContinuous-v2
-
-
Finished
-
prabhasak
6d 21h 16m 51s
-
["sac","td3"]
128
-
0.0000235
-
false
-
gpu
-
0.01118
Pendulum-v0
true
1.27273
0.99
-
0.9
-
1000
0.000193
-
-
-
-
MlpPolicy
true
42
true
0
[0,"1e5"]
1024
[false,true]
[false,true]
-
-
0
10
0.00428
true
11.25
[100000,"1e5"]
models
Pendulum-v0
-
-
Finished
-
prabhasak
1d 22h 51m 21s
-
["sac","trpo"]
256
1000000
0.1
15
false
-
gpu
0.01
0.01
Hopper-v2
true
1
0.99
1
0.95
lin_3e-4
1000
0.005
-
-
-
-
["CustomSACPolicy","MlpPolicy"]
true
42
true
0
0
2048
[false,true]
[false,true]
1
-
0
5
0.001
true
11.25
5e5
models
Hopper-v2
-
-
Finished
-
prabhasak
1d 5h 5m 19s
-
["dqn","ppo2"]
-
50000
0.001
10
false
lin_0.2
gpu
0
0
CartPole-v1
true
1
0.98667
-
0.93333
[0.001,"lin_0.001"]
-
0.001
32
-
1
20
["CustomDQNPolicy","MlpPolicy"]
true
42
true
[0,"5e5"]
0
512
[false,true]
[false,true]
-
-
0
3
0.0001
true
7.5
-
-
-
-
-
1-6
of 6