CUMTHenry commited on
Commit
07775c2
·
verified ·
1 Parent(s): edfa92d

Upload 17 file(s)

Browse files
antmaze-giant-stitch-v0/chunk_consistency/og_antM_Gi_o2d_Cd_Stgl_PadBuf_Ft64_ts512_T512_ChunkConsistency_v1_restart_s200k/train_metrics.jsonl CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f490e57a6e9fa22a8327caf2b4373441bf39ee2be9fa4316320be110e01cf1c4
3
- size 36943973
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17efda7baba4f4018c41c257c03f2b8388bbed737f784f50854ba40b0d5f84cc
3
+ size 36980706
dql_results/cube-double-play-v0/actor_200.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6f939461a9ff79a6ee82924e177d1475a1e6a75d8023f9758be3c23420df955
3
+ size 607430
dql_results/cube-double-play-v0/critic_200.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1f3883be06241400cc1471282f35a22767bd57f2a14d7893492d43ba2f3550c
3
+ size 36769444
dql_results/cube-double-play-v0/critic_target_200.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92ede4f665d2fb5a4ce3fec9f6dbf7ded22d340a15817ea8714656e7440275ca
3
+ size 36769528
dql_results/cube-double-play-v0/variant.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "T": 5,
3
+ "action_dim": 5,
4
+ "algo": "ql",
5
+ "batch_size": 256,
6
+ "beta_schedule": "vp",
7
+ "device": "cuda:4",
8
+ "dir": "dql/results",
9
+ "discount": 0.99,
10
+ "early_stop": false,
11
+ "env_name": "cube-double-play-v0",
12
+ "eta": 1.0,
13
+ "eval_episodes": 100,
14
+ "eval_freq": 100,
15
+ "exp": "exp",
16
+ "gn": 7.0,
17
+ "goal_dim": 3,
18
+ "lcb_coef": 4.0,
19
+ "lr": 0.0003,
20
+ "lr_decay": true,
21
+ "max_action": 1.0,
22
+ "max_q_backup": false,
23
+ "ms": "offline",
24
+ "num_epochs": 2000,
25
+ "num_steps_per_epoch": 1000,
26
+ "output_dir": "dql/results",
27
+ "p": 0.2,
28
+ "replan": 15,
29
+ "reward_tune": "cql_antmaze",
30
+ "save_best_model": false,
31
+ "seed": 0,
32
+ "state_dim": 37,
33
+ "target_steps": 10,
34
+ "tau": 0.005,
35
+ "top_k": 1
36
+ }
dql_results/cube-quadruple-play-v0/actor_200.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70d17a620041fa9c0d6f268c43e51b9bdfd655628c7bc3541c768fadcd8773de
3
+ size 625862
dql_results/cube-quadruple-play-v0/critic_200.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fae515b79fdb46d41dab3d9ba43027eca0a0bdc788acd46dc7d0989bd4f86b03
3
+ size 37949092
dql_results/cube-quadruple-play-v0/critic_target_200.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:313700fdf238b2a12c4b132cff53298bb4372373dd092099bf9e39f07f59a8fb
3
+ size 37949176
dql_results/cube-quadruple-play-v0/variant.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "T": 5,
3
+ "action_dim": 5,
4
+ "algo": "ql",
5
+ "batch_size": 256,
6
+ "beta_schedule": "vp",
7
+ "device": "cuda:4",
8
+ "dir": "dql/results",
9
+ "discount": 0.99,
10
+ "early_stop": false,
11
+ "env_name": "cube-quadruple-play-v0",
12
+ "eta": 1.0,
13
+ "eval_episodes": 100,
14
+ "eval_freq": 100,
15
+ "exp": "exp",
16
+ "gn": 7.0,
17
+ "goal_dim": 3,
18
+ "lcb_coef": 4.0,
19
+ "lr": 0.0003,
20
+ "lr_decay": true,
21
+ "max_action": 1.0,
22
+ "max_q_backup": false,
23
+ "ms": "offline",
24
+ "num_epochs": 2000,
25
+ "num_steps_per_epoch": 1000,
26
+ "output_dir": "dql/results",
27
+ "p": 0.2,
28
+ "replan": 15,
29
+ "reward_tune": "cql_antmaze",
30
+ "save_best_model": false,
31
+ "seed": 0,
32
+ "state_dim": 55,
33
+ "target_steps": 10,
34
+ "tau": 0.005,
35
+ "top_k": 1
36
+ }
dql_results/cube-single-play-v0/actor_200.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb085b1a25be0611e9cff058addc27e45f0ee38ed64e3ba343bcf697bcb8821b
3
+ size 598214
dql_results/cube-single-play-v0/critic_200.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cba1f59fee700b5baeca9b1ea6f74e704cf1825ddc5dbc74ca055d227c0d47a4
3
+ size 36179620
dql_results/cube-single-play-v0/critic_target_200.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ba7fc77a2da6e08ea0a71f563e4d1c6e06a7d0e7bc823db4b8d664accce45dd
3
+ size 36179704
dql_results/cube-single-play-v0/variant.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "T": 5,
3
+ "action_dim": 5,
4
+ "algo": "ql",
5
+ "batch_size": 256,
6
+ "beta_schedule": "vp",
7
+ "device": "cuda:1",
8
+ "dir": "dql/results",
9
+ "discount": 0.99,
10
+ "early_stop": false,
11
+ "env_name": "cube-single-play-v0",
12
+ "eta": 1.0,
13
+ "eval_episodes": 100,
14
+ "eval_freq": 100,
15
+ "exp": "exp",
16
+ "gn": 7.0,
17
+ "goal_dim": 3,
18
+ "lcb_coef": 4.0,
19
+ "lr": 0.0003,
20
+ "lr_decay": true,
21
+ "max_action": 1.0,
22
+ "max_q_backup": false,
23
+ "ms": "offline",
24
+ "num_epochs": 200,
25
+ "num_steps_per_epoch": 1000,
26
+ "output_dir": "dql/results",
27
+ "p": 0.2,
28
+ "replan": 15,
29
+ "reward_tune": "cql_antmaze",
30
+ "save_best_model": false,
31
+ "seed": 0,
32
+ "state_dim": 28,
33
+ "target_steps": 10,
34
+ "tau": 0.005,
35
+ "top_k": 1
36
+ }
dql_results/cube-triple-play-v0/actor_200.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7ef4bb1ebcdec500ec011641bbd3f8c6f15901409039b28ec95b3645464db3d
3
+ size 616646
dql_results/cube-triple-play-v0/critic_200.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa9ea9f3aa76e2d9885d28f857b72bad50ec5a4cd7fecbd98270edf2eabacd3a
3
+ size 37359268
dql_results/cube-triple-play-v0/critic_target_200.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a470dcf37323ef5fbe7517db9b03dcf1eadb8cfe6da45abb6f597c84ea3e010
3
+ size 37359352
dql_results/cube-triple-play-v0/variant.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "T": 5,
3
+ "action_dim": 5,
4
+ "algo": "ql",
5
+ "batch_size": 256,
6
+ "beta_schedule": "vp",
7
+ "device": "cuda:4",
8
+ "dir": "dql/results",
9
+ "discount": 0.99,
10
+ "early_stop": false,
11
+ "env_name": "cube-triple-play-v0",
12
+ "eta": 1.0,
13
+ "eval_episodes": 100,
14
+ "eval_freq": 100,
15
+ "exp": "exp",
16
+ "gn": 7.0,
17
+ "goal_dim": 3,
18
+ "lcb_coef": 4.0,
19
+ "lr": 0.0003,
20
+ "lr_decay": true,
21
+ "max_action": 1.0,
22
+ "max_q_backup": false,
23
+ "ms": "offline",
24
+ "num_epochs": 2000,
25
+ "num_steps_per_epoch": 1000,
26
+ "output_dir": "dql/results",
27
+ "p": 0.2,
28
+ "replan": 15,
29
+ "reward_tune": "cql_antmaze",
30
+ "save_best_model": false,
31
+ "seed": 0,
32
+ "state_dim": 46,
33
+ "target_steps": 10,
34
+ "tau": 0.005,
35
+ "top_k": 1
36
+ }