Caomingjun's workspace
Runs
490
Name
0 visualized
task: pen-human-v1
task: pen-human-v1
5
task: pen-cloned-v1
task: pen-cloned-v1
5
State
Notes
User
Tags
Created
Runtime
Sweep
algo.actor_hidden_dims
algo.actor_lr
algo.beta
algo.conditional_logstd
algo.critic_ensemble_size
algo.critic_hidden_dims
algo.critic_lr
algo.deterministic_actor
algo.discount
algo.expectile
algo.layer_norm
algo.lr_decay_steps
algo.max_action
algo.min_action
algo.name
algo.opt_decay_schedule
algo.policy_logstd_min
algo.tau
algo.value_hidden_dims
algo.value_lr
data.batch_size
data.clip_eps
data.dataset
data.norm_reward
data.scan
device
eval.interval
eval.num_episodes
eval.num_samples
eval.stats_interval
eval.temperature
log.dir
log.entity
log.interval
log.project
log.save_ckpt
log.save_video
log.tag
mode
norm_obs
pretrain_steps
seed
task
train_steps
Finished
-
gaochenxiao
53s
-
-
-
-
-
-
-
-
-
0.99
0.9
-
${train_steps}
1
-1
dtql
-
-
-
-
-
256
0.00001
d4rl
none
false
7
50000
10
1024
2000
-
logs
lamda-rl
500
flow-rl
false
false
default
-
-
50000
2
pen-human-v1
300000
Finished
-
gaochenxiao
9s
-
-
-
-
-
-
-
-
-
0.99
0.9
-
${train_steps}
1
-1
dtql
-
-
-
-
-
256
0.00001
d4rl
none
false
7
50000
10
1024
2000
-
logs
lamda-rl
500
flow-rl
false
false
default
-
-
50000
3
pen-human-v1
300000
Finished
-
gaochenxiao
9s
-
-
-
-
-
-
-
-
-
0.99
0.9
-
${train_steps}
1
-1
dtql
-
-
-
-
-
256
0.00001
d4rl
none
false
7
50000
10
1024
2000
-
logs
lamda-rl
500
flow-rl
false
false
default
-
-
50000
2
pen-human-v1
300000
Finished
-
gaochenxiao
7s
-
-
-
-
-
-
-
-
-
0.99
0.9
-
${train_steps}
1
-1
dtql
-
-
-
-
-
256
0.00001
d4rl
none
false
7
50000
10
1024
2000
-
logs
lamda-rl
500
flow-rl
false
false
default
-
-
50000
0
pen-human-v1
300000
Finished
-
gaochenxiao
7s
-
-
-
-
-
-
-
-
-
0.99
0.9
-
${train_steps}
1
-1
dtql
-
-
-
-
-
256
0.00001
d4rl
none
false
7
50000
10
1024
2000
-
logs
lamda-rl
500
flow-rl
false
false
default
-
-
50000
4
pen-human-v1
300000
Finished
-
gaochenxiao
9s
-
-
-
-
-
-
-
-
-
0.99
0.9
-
${train_steps}
1
-1
dtql
-
-
-
-
-
256
0.00001
d4rl
none
false
7
50000
10
1024
2000
-
logs
lamda-rl
500
flow-rl
false
false
default
-
-
50000
1
pen-human-v1
300000
Finished
-
gaochenxiao
53s
-
-
-
-
-
-
-
-
-
0.99
0.7
-
${train_steps}
1
-1
dtql
-
-
-
-
-
256
0.00001
d4rl
none
false
["3","7"]
50000
10
1024
2000
-
logs
lamda-rl
500
flow-rl
false
false
default
-
-
50000
2
pen-cloned-v1
200000
1-2
of 2