Gaochenxiao's workspace
Runs
60
Name
5 visualized
task: walker2d-random-v2
task: walker2d-random-v2
1
5
task: walker2d-medium-expert-v2
task: walker2d-medium-expert-v2
1
5
task: hopper-medium-replay-v2
task: hopper-medium-replay-v2
1
5
task: hopper-medium-expert-v2
task: hopper-medium-expert-v2
1
5
task: hopper-random-v2
task: hopper-random-v2
1
5
task: halfcheetah-random-v2
task: halfcheetah-random-v2
1
5
task: hopper-medium-v2
task: hopper-medium-v2
1
5
task: walker2d-medium-replay-v2
task: walker2d-medium-replay-v2
1
5
task: walker2d-medium-v2
task: walker2d-medium-v2
1
5
task: halfcheetah-medium-replay-v2
task: halfcheetah-medium-replay-v2
1
5
task: halfcheetah-medium-v2
task: halfcheetah-medium-v2
1
5
task: halfcheetah-medium-expert-v2
task: halfcheetah-medium-expert-v2
1
5
State
Notes
User
Tags
Created
Runtime
Sweep
UtilsRL.numpy_fp
UtilsRL.precision
UtilsRL.torch_fp
actor_lr
actor_update_interval
alpha
batch_size
critic_lr
debug
device
discount
eval_episode
eval_interval
hidden_dims
log_interval
max_action
max_epoch
name
noise_clip
normalize_obs
normalize_reward
policy_noise
save_interval
seed
step_per_epoch
task
tau
wandb.entity
wandb.project
Eval/length_mean
Eval/length_std
Eval/normalized_score_mean
Eval/normalized_score_std
loss/actor_bc_loss
loss/actor_q_loss
loss/actor_total_loss
loss/q_loss
misc/q1_val
Finished
-
kongrui
4d 5h 50m 55s
-
numpy.float32
float32
torch.float32
0.0003
2
2.5
256
0.0003
false
["cuda:0","cuda:1"]
0.99
10
10
256
10
1
1000
corl
0.5
true
false
0.2
50
2.6
1000
walker2d-random-v2
0.005
lamda-rl
td3bc-Offline
115.68
13.47698
1.9516
0.46874
0.39603
-2.48377
-2.08774
46042628374711500
1638982016
Finished
-
kongrui
4d 5h 54m 11s
-
numpy.float32
float32
torch.float32
0.0003
2
2.5
256
0.0003
false
["cuda:0","cuda:1"]
0.99
10
10
256
10
1
1000
corl
0.5
true
false
0.2
50
2.6
1000
walker2d-medium-expert-v2
0.005
lamda-rl
td3bc-Offline
1000
0
110.39609
0.1022
0.031597
-2.49947
-2.46787
21.75717
393.51389
Finished
-
kongrui
4d 5h 52m 10s
-
numpy.float32
float32
torch.float32
0.0003
2
2.5
256
0.0003
false
cuda:0
0.99
10
10
256
10
1
1000
corl
0.5
true
false
0.2
50
2.6
1000
hopper-medium-replay-v2
0.005
lamda-rl
td3bc-Offline
556.7
197.9693
56.72938
19.75326
0.11822
-2.49921
-2.38099
22.94636
171.7374
Finished
-
kongrui
4d 5h 54m
-
numpy.float32
float32
torch.float32
0.0003
2
2.5
256
0.0003
false
cuda:0
0.99
10
10
256
10
1
1000
corl
0.5
true
false
0.2
50
2.6
1000
hopper-medium-expert-v2
0.005
lamda-rl
td3bc-Offline
914.58
144.42796
103.35221
16.46872
0.056737
-2.5
-2.44326
6.29005
296.89066
Finished
-
kongrui
4d 5h 54m 12s
-
numpy.float32
float32
torch.float32
0.0003
2
2.5
256
0.0003
false
cuda:0
0.99
10
10
256
10
1
1000
corl
0.5
true
false
0.2
50
2.6
1000
hopper-random-v2
0.005
lamda-rl
td3bc-Offline
137.36
0.53971
8.46602
0.037744
0.37303
-2.49662
-2.12359
28.93163
100.53061
Finished
-
kongrui
4d 5h 52m 54s
-
numpy.float32
float32
torch.float32
0.0003
2
2.5
256
0.0003
false
cuda:0
0.99
10
10
256
10
1
1000
corl
0.5
true
false
0.2
50
2.6
1000
halfcheetah-random-v2
0.005
lamda-rl
td3bc-Offline
1000
0
11.51115
0.057613
0.34888
-2.33284
-1.98395
2.03548
57.41666
Finished
-
kongrui
4d 5h 53m 49s
-
numpy.float32
float32
torch.float32
0.0003
2
2.5
256
0.0003
false
cuda:0
0.99
10
10
256
10
1
1000
corl
0.5
true
false
0.2
50
2.6
1000
hopper-medium-v2
0.005
lamda-rl
td3bc-Offline
613.48
117.94511
61.76177
12.23513
0.049256
-2.5
-2.45074
3.55423
236.24799
Finished
-
kongrui
4d 5h 55m 50s
-
numpy.float32
float32
torch.float32
0.0003
2
2.5
256
0.0003
false
["cuda:0","cuda:1"]
0.99
10
10
256
10
1
1000
corl
0.5
true
false
0.2
50
2.6
1000
walker2d-medium-replay-v2
0.005
lamda-rl
td3bc-Offline
836.66
210.72398
71.64811
20.04909
0.10923
-2.47509
-2.36587
58.22429
196.71161
Finished
-
kongrui
4d 5h 56m 1s
-
numpy.float32
float32
torch.float32
0.0003
2
2.5
256
0.0003
false
["cuda:0","cuda:1"]
0.99
10
10
256
10
1
1000
corl
0.5
true
false
0.2
50
2.6
1000
walker2d-medium-v2
0.005
lamda-rl
td3bc-Offline
985.6
43.2
84.3233
4.64105
0.033973
-2.49947
-2.4655
16.59302
308.57663
Finished
-
kongrui
4d 5h 57m 50s
-
numpy.float32
float32
torch.float32
0.0003
2
2.5
256
0.0003
false
cuda:0
0.99
10
10
256
10
1
1000
corl
0.5
true
false
0.2
50
2.6
1000
halfcheetah-medium-replay-v2
0.005
lamda-rl
td3bc-Offline
1000
0
44.72973
0.7568
0.092814
-2.39906
-2.30624
41.6676
310.00989
Finished
-
kongrui
4d 5h 57m
-
numpy.float32
float32
torch.float32
0.0003
2
2.5
256
0.0003
false
cuda:0
0.99
10
10
256
10
1
1000
corl
0.5
true
false
0.2
50
2.6
1000
halfcheetah-medium-v2
0.005
lamda-rl
td3bc-Offline
1000
0
48.50859
0.57966
0.028691
-2.49321
-2.46452
30.89496
449.04352
Finished
-
kongrui
4d 5h 54m 9s
-
numpy.float32
float32
torch.float32
0.0003
2
2.5
256
0.0003
false
cuda:0
0.99
10
10
256
10
1
1000
corl
0.5
true
false
0.2
50
2.6
1000
halfcheetah-medium-expert-v2
0.005
lamda-rl
td3bc-Offline
1000
0
91.70986
6.56786
0.032007
-2.49733
-2.46532
168.91643
725.14124
1-12
of 12