Whalswp commited on
Commit
5ea09f8
·
verified ·
1 Parent(s): 1143035

Add files using upload-large-folder tool

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +2 -0
  2. Baseline/action_space_manifest.json +141 -0
  3. Baseline/checkpoints/050000/action_space_manifest.json +141 -0
  4. Baseline/checkpoints/050000/prompt_manifest.json +40 -0
  5. Baseline/checkpoints/050000/resolved_config.yaml +24 -0
  6. Baseline/checkpoints/050000/source_config.yaml +63 -0
  7. Baseline/checkpoints/060000/action_space_manifest.json +141 -0
  8. Baseline/checkpoints/060000/prompt_manifest.json +40 -0
  9. Baseline/checkpoints/060000/source_config.yaml +63 -0
  10. Baseline/prompt_manifest.json +40 -0
  11. Baseline/resolved_config.yaml +24 -0
  12. Baseline/source_config.yaml +63 -0
  13. Baseline/wandb/debug-internal.log +0 -0
  14. Baseline/wandb/debug.log +46 -0
  15. Baseline/wandb/run-20260714_161240-1jmjb68j/files/output.log +269 -0
  16. Baseline/wandb/run-20260714_161240-1jmjb68j/files/requirements.txt +117 -0
  17. Baseline/wandb/run-20260714_161240-1jmjb68j/files/wandb-metadata.json +91 -0
  18. Baseline/wandb/run-20260714_161240-1jmjb68j/logs/debug-core.log +8 -0
  19. Baseline/wandb/run-20260714_161240-1jmjb68j/logs/debug-internal.log +0 -0
  20. Baseline/wandb/run-20260714_161240-1jmjb68j/logs/debug.log +19 -0
  21. Baseline/wandb/run-20260714_161240-1jmjb68j/run-1jmjb68j.wandb +3 -0
  22. Baseline/wandb/run-20260715_143250-1jmjb68j/files/config.yaml +351 -0
  23. Baseline/wandb/run-20260715_143250-1jmjb68j/files/output.log +74 -0
  24. Baseline/wandb/run-20260715_143250-1jmjb68j/files/requirements.txt +117 -0
  25. Baseline/wandb/run-20260715_143250-1jmjb68j/files/wandb-metadata.json +92 -0
  26. Baseline/wandb/run-20260715_143250-1jmjb68j/files/wandb-summary.json +1 -0
  27. Baseline/wandb/run-20260715_143250-1jmjb68j/logs/debug-core.log +16 -0
  28. Baseline/wandb/run-20260715_143250-1jmjb68j/logs/debug-internal.log +0 -0
  29. Baseline/wandb/run-20260715_143250-1jmjb68j/logs/debug.log +46 -0
  30. Baseline/wandb/run-20260715_143250-1jmjb68j/run-1jmjb68j.wandb +3 -0
  31. Baseline_gpu3_resume.log +0 -0
  32. Baseline_gpu3_resume.pid +1 -0
  33. Baseline_gpu3_watcher.log +17 -0
  34. Baseline_gpu3_watcher_status.json +5 -0
  35. Baseline_train.log +0 -0
  36. Baseline_train.pid +1 -0
  37. RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/debug.log +19 -0
  38. RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/run-20260715_011448-fbqzmp5m/files/requirements.txt +117 -0
  39. RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/run-20260715_011448-fbqzmp5m/files/wandb-metadata.json +91 -0
  40. RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/run-20260715_011448-fbqzmp5m/logs/debug-core.log +26 -0
  41. RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/run-20260715_011448-fbqzmp5m/logs/debug.log +19 -0
  42. RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/debug.log +19 -0
  43. RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/run-20260715_011448-pebh9ode/files/requirements.txt +117 -0
  44. RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/run-20260715_011448-pebh9ode/files/wandb-metadata.json +91 -0
  45. RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/run-20260715_011448-pebh9ode/logs/debug-core.log +26 -0
  46. RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/run-20260715_011448-pebh9ode/logs/debug.log +19 -0
  47. RKD_TimewarpVAE/RSCLstyle_cosine_60K/wandb/debug.log +19 -0
  48. RKD_TimewarpVAE/RSCLstyle_cosine_60K/wandb/run-20260715_102129-typqzcxc/files/requirements.txt +117 -0
  49. RKD_TimewarpVAE/RSCLstyle_cosine_60K/wandb/run-20260715_102129-typqzcxc/files/wandb-metadata.json +91 -0
  50. RKD_TimewarpVAE/RSCLstyle_cosine_60K/wandb/run-20260715_102129-typqzcxc/logs/debug-core.log +7 -0
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ Baseline/wandb/run-20260714_161240-1jmjb68j/run-1jmjb68j.wandb filter=lfs diff=lfs merge=lfs -text
37
+ Baseline/wandb/run-20260715_143250-1jmjb68j/run-1jmjb68j.wandb filter=lfs diff=lfs merge=lfs -text
Baseline/action_space_manifest.json ADDED
@@ -0,0 +1,141 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": 1,
3
+ "action_space": "ee_abs_rot6d",
4
+ "source_feature": "action.ee_abs_rot6d",
5
+ "canonical_feature": "action",
6
+ "dtype": "float32",
7
+ "shape": [
8
+ 10
9
+ ],
10
+ "names": [
11
+ "x",
12
+ "y",
13
+ "z",
14
+ "r_col0_x",
15
+ "r_col0_y",
16
+ "r_col0_z",
17
+ "r_col1_x",
18
+ "r_col1_y",
19
+ "r_col1_z",
20
+ "gripper"
21
+ ],
22
+ "dataset_repo_id": "Whalswp/INSIGHTfixposV4_filtered_multispace_v2",
23
+ "dataset_revision": "v3.0",
24
+ "dataset_total_frames": 218367,
25
+ "dataset_total_episodes": 4930,
26
+ "dataset_fps": 10,
27
+ "stats_sha256": "71698c7337070aa8267df5aea7ec08bc00d3925510f2649bc20c9201eeb669e4",
28
+ "stats": {
29
+ "min": [
30
+ -0.21784618496894836,
31
+ -0.4474811851978302,
32
+ 0.1702737659215927,
33
+ -0.9999986290931702,
34
+ -0.9999991655349731,
35
+ -0.9999999403953552,
36
+ -0.9999990463256836,
37
+ -0.9999998211860657,
38
+ -0.9999988079071045,
39
+ 0.0
40
+ ],
41
+ "max": [
42
+ 0.6283698678016663,
43
+ 0.4420371949672699,
44
+ 1.051027774810791,
45
+ 0.9999999403953552,
46
+ 0.9999995231628418,
47
+ 1.0,
48
+ 0.9999995231628418,
49
+ 1.0,
50
+ 0.9999996423721313,
51
+ 0.03999999910593033
52
+ ],
53
+ "mean": [
54
+ 0.4040228037735655,
55
+ -0.05247618696989652,
56
+ 0.6993459771218912,
57
+ 0.07011111991957117,
58
+ -0.03520558904460197,
59
+ 0.14470242846953205,
60
+ 0.05491166606739672,
61
+ -0.19073058893378805,
62
+ -0.06988117661871228,
63
+ 0.021167300266524632
64
+ ],
65
+ "std": [
66
+ 0.11861927913357073,
67
+ 0.11318943557281885,
68
+ 0.17849160646987516,
69
+ 0.4812528791886511,
70
+ 0.7487196258736938,
71
+ 0.4251126804030219,
72
+ 0.47797229807769137,
73
+ 0.6267695346407992,
74
+ 0.5782954306815554,
75
+ 0.01996590431050817
76
+ ],
77
+ "count": [
78
+ 218367
79
+ ],
80
+ "q01": [
81
+ 0.23988881827972064,
82
+ -0.10387461883034811,
83
+ 0.6297243923476871,
84
+ -0.43622841955353464,
85
+ -0.6885486527617446,
86
+ -0.14103555388039019,
87
+ -0.4327210063987595,
88
+ -0.9304790915716145,
89
+ -0.3465462258490417,
90
+ -1.000000013351432e-10
91
+ ],
92
+ "q10": [
93
+ 0.3042306436755216,
94
+ -0.09609991620622624,
95
+ 0.6394063072781145,
96
+ -0.3351306352998669,
97
+ -0.5751963015002924,
98
+ -0.08490621283307866,
99
+ -0.33849701676224275,
100
+ -0.8340557712529717,
101
+ -0.29760290790884514,
102
+ -1.000000013351432e-10
103
+ ],
104
+ "q50": [
105
+ 0.4160225263528291,
106
+ -0.06267459152989589,
107
+ 0.6811777491060801,
108
+ 0.0543536902230377,
109
+ -0.14012251058819217,
110
+ 0.04772394280004727,
111
+ 0.03838119138716219,
112
+ -0.14865169094195635,
113
+ -0.2073069401172425,
114
+ 0.020772594934870346
115
+ ],
116
+ "q90": [
117
+ 0.46641008755679864,
118
+ 0.020233143101890064,
119
+ 0.8095135405259374,
120
+ 0.48172401592474906,
121
+ 0.7398737043070319,
122
+ 0.5970841215636613,
123
+ 0.4674752875999555,
124
+ 0.36281754777672676,
125
+ 0.5033361883834496,
126
+ 0.039998362213352394
127
+ ],
128
+ "q99": [
129
+ 0.47320723171849244,
130
+ 0.04114262476135705,
131
+ 0.8467967259270389,
132
+ 0.5867781409980988,
133
+ 0.8529584859257228,
134
+ 0.7953936054341789,
135
+ 0.5768269858732034,
136
+ 0.49353349323389695,
137
+ 0.7168914440776504,
138
+ 0.03999983541667277
139
+ ]
140
+ }
141
+ }
Baseline/checkpoints/050000/action_space_manifest.json ADDED
@@ -0,0 +1,141 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": 1,
3
+ "action_space": "ee_abs_rot6d",
4
+ "source_feature": "action.ee_abs_rot6d",
5
+ "canonical_feature": "action",
6
+ "dtype": "float32",
7
+ "shape": [
8
+ 10
9
+ ],
10
+ "names": [
11
+ "x",
12
+ "y",
13
+ "z",
14
+ "r_col0_x",
15
+ "r_col0_y",
16
+ "r_col0_z",
17
+ "r_col1_x",
18
+ "r_col1_y",
19
+ "r_col1_z",
20
+ "gripper"
21
+ ],
22
+ "dataset_repo_id": "Whalswp/INSIGHTfixposV4_filtered_multispace_v2",
23
+ "dataset_revision": "v3.0",
24
+ "dataset_total_frames": 218367,
25
+ "dataset_total_episodes": 4930,
26
+ "dataset_fps": 10,
27
+ "stats_sha256": "71698c7337070aa8267df5aea7ec08bc00d3925510f2649bc20c9201eeb669e4",
28
+ "stats": {
29
+ "min": [
30
+ -0.21784618496894836,
31
+ -0.4474811851978302,
32
+ 0.1702737659215927,
33
+ -0.9999986290931702,
34
+ -0.9999991655349731,
35
+ -0.9999999403953552,
36
+ -0.9999990463256836,
37
+ -0.9999998211860657,
38
+ -0.9999988079071045,
39
+ 0.0
40
+ ],
41
+ "max": [
42
+ 0.6283698678016663,
43
+ 0.4420371949672699,
44
+ 1.051027774810791,
45
+ 0.9999999403953552,
46
+ 0.9999995231628418,
47
+ 1.0,
48
+ 0.9999995231628418,
49
+ 1.0,
50
+ 0.9999996423721313,
51
+ 0.03999999910593033
52
+ ],
53
+ "mean": [
54
+ 0.4040228037735655,
55
+ -0.05247618696989652,
56
+ 0.6993459771218912,
57
+ 0.07011111991957117,
58
+ -0.03520558904460197,
59
+ 0.14470242846953205,
60
+ 0.05491166606739672,
61
+ -0.19073058893378805,
62
+ -0.06988117661871228,
63
+ 0.021167300266524632
64
+ ],
65
+ "std": [
66
+ 0.11861927913357073,
67
+ 0.11318943557281885,
68
+ 0.17849160646987516,
69
+ 0.4812528791886511,
70
+ 0.7487196258736938,
71
+ 0.4251126804030219,
72
+ 0.47797229807769137,
73
+ 0.6267695346407992,
74
+ 0.5782954306815554,
75
+ 0.01996590431050817
76
+ ],
77
+ "count": [
78
+ 218367
79
+ ],
80
+ "q01": [
81
+ 0.23988881827972064,
82
+ -0.10387461883034811,
83
+ 0.6297243923476871,
84
+ -0.43622841955353464,
85
+ -0.6885486527617446,
86
+ -0.14103555388039019,
87
+ -0.4327210063987595,
88
+ -0.9304790915716145,
89
+ -0.3465462258490417,
90
+ -1.000000013351432e-10
91
+ ],
92
+ "q10": [
93
+ 0.3042306436755216,
94
+ -0.09609991620622624,
95
+ 0.6394063072781145,
96
+ -0.3351306352998669,
97
+ -0.5751963015002924,
98
+ -0.08490621283307866,
99
+ -0.33849701676224275,
100
+ -0.8340557712529717,
101
+ -0.29760290790884514,
102
+ -1.000000013351432e-10
103
+ ],
104
+ "q50": [
105
+ 0.4160225263528291,
106
+ -0.06267459152989589,
107
+ 0.6811777491060801,
108
+ 0.0543536902230377,
109
+ -0.14012251058819217,
110
+ 0.04772394280004727,
111
+ 0.03838119138716219,
112
+ -0.14865169094195635,
113
+ -0.2073069401172425,
114
+ 0.020772594934870346
115
+ ],
116
+ "q90": [
117
+ 0.46641008755679864,
118
+ 0.020233143101890064,
119
+ 0.8095135405259374,
120
+ 0.48172401592474906,
121
+ 0.7398737043070319,
122
+ 0.5970841215636613,
123
+ 0.4674752875999555,
124
+ 0.36281754777672676,
125
+ 0.5033361883834496,
126
+ 0.039998362213352394
127
+ ],
128
+ "q99": [
129
+ 0.47320723171849244,
130
+ 0.04114262476135705,
131
+ 0.8467967259270389,
132
+ 0.5867781409980988,
133
+ 0.8529584859257228,
134
+ 0.7953936054341789,
135
+ 0.5768269858732034,
136
+ 0.49353349323389695,
137
+ 0.7168914440776504,
138
+ 0.03999983541667277
139
+ ]
140
+ }
141
+ }
Baseline/checkpoints/050000/prompt_manifest.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": 1,
3
+ "prompt_set": "detailed",
4
+ "source_file": "task_prompts.json",
5
+ "source_sha256": "0ba48fb372da9f49dcfcd7dcc2da8cce6ce991d3661060e664a104778d1ffacc",
6
+ "mapping_sha256": "f71954786e94cdeceb135dde09fdc6b65db0b87ab3b09d25060129d0f0fde018",
7
+ "dataset_repo_id": "Whalswp/INSIGHTfixposV4_filtered_multispace_v2",
8
+ "dataset_revision": "v3.0",
9
+ "task_count": 13,
10
+ "task_codes": [
11
+ "5f",
12
+ "5g",
13
+ "1ext",
14
+ "3b",
15
+ "5b",
16
+ "3a",
17
+ "3c",
18
+ "5h",
19
+ "5d",
20
+ "3d",
21
+ "5e",
22
+ "5a",
23
+ "5c"
24
+ ],
25
+ "prompts": {
26
+ "5f": "Close the bottle in clockwise direction.",
27
+ "5g": "Grip the cap on the sides indicated by the 'squeeze' arrow and open the bottle in clockwise direction.",
28
+ "1ext": "Find the arrow guide and open the indicated drawer.",
29
+ "3b": "Open the door, rotate counter-clockwise and push.",
30
+ "5b": "Open the bottle in counter-clockwise direction.",
31
+ "3a": "Open the door, rotate clockwise and push.",
32
+ "3c": "Open the door, rotate clockwise and pull.",
33
+ "5h": "Close the bottle in counter-clockwise direction.",
34
+ "5d": "Close the bottle in clockwise direction.",
35
+ "3d": "Open the door, rotate counter-clockwise and pull.",
36
+ "5e": "Close the bottle in counter-clockwise direction.",
37
+ "5a": "Grip the cap on the sides indicated by the 'squeeze' arrow and open the bottle in counter-clockwise direction.",
38
+ "5c": "Open the bottle in clockwise direction."
39
+ }
40
+ }
Baseline/checkpoints/050000/resolved_config.yaml ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ dataset:
2
+ repo_id: Whalswp/INSIGHTfixposV4_filtered_multispace_v2
3
+ root: /home/ext_minje/INSIGHTfixposV4_filtered_multispace_v2
4
+ policy:
5
+ type: groot
6
+ device: cuda
7
+ chunk_size: 16
8
+ n_action_steps: 16
9
+ push_to_hub: false
10
+ tune_llm: false
11
+ tune_visual: false
12
+ tune_top_llm_layers: 0
13
+ tune_projector: true
14
+ tune_diffusion_model: true
15
+ tune_vlln: true
16
+ seed: 42
17
+ batch_size: 64
18
+ steps: 60000
19
+ log_freq: 200
20
+ output_dir: /home/ext_minje/groot_insight/Abs_6D/Baseline
21
+ job_name: INSIGHT_6D_baseline
22
+ wandb:
23
+ enable: true
24
+ disable_artifact: true
Baseline/checkpoints/050000/source_config.yaml ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # INSIGHT GR00T N1.7 training config
2
+ # INSIGHT wrapper-only fields are stripped before LeRobot config parsing.
3
+ insight:
4
+ # Options: joint_abs | ee_abs_quat | ee_abs_rot6d | ee_delta
5
+ action_space: ee_abs_rot6d
6
+ # Options: coarse | detailed
7
+ prompt_set: detailed
8
+ save_steps: ["50K"]
9
+
10
+ dataset:
11
+ repo_id: Whalswp/INSIGHTfixposV4_filtered_multispace_v2
12
+ root: ${HOME}/INSIGHTfixposV4_filtered_multispace_v2
13
+
14
+ policy:
15
+ type: groot
16
+ device: cuda
17
+ chunk_size: 16
18
+ n_action_steps: 16
19
+ push_to_hub: false
20
+ # Trainable scope. This baseline trains projector, diffusion, and VLLN; LLM/vision stay frozen.
21
+ # VLM
22
+ tune_llm: false
23
+ tune_visual: false
24
+ tune_top_llm_layers: 0
25
+ # Action Expert
26
+ tune_projector: true
27
+ tune_diffusion_model: true
28
+ tune_vlln: true
29
+
30
+
31
+ seed: 42
32
+ batch_size: 64
33
+ steps: 60000
34
+ log_freq: 200
35
+ output_dir: ${HOME}/groot_insight/Abs_6D/Baseline
36
+ job_name: INSIGHT_6D_baseline
37
+
38
+ wandb:
39
+ enable: true
40
+ disable_artifact: true
41
+
42
+ # Omitted because these match LeRobot/GR00T defaults or are unused in this setup:
43
+ #
44
+ # dataset:
45
+ # eval_split: 0.0
46
+ # image_transforms:
47
+ # enable: false # default; set true only for train-image augmentation.
48
+ #
49
+ # policy:
50
+ # base_model_path: nvidia/GR00T-N1.7-3B
51
+ # embodiment_tag: new_embodiment
52
+ # use_relative_actions: false
53
+ # use_bf16: true
54
+ # repo_id: Whalswp/INSIGHT # unused while push_to_hub is false
55
+ #
56
+ # save_checkpoint: true
57
+ # use_policy_training_preset: true
58
+ # eval_steps: 0
59
+ # env_eval_freq: 0
60
+ #
61
+ # insight.prompt_set options:
62
+ # coarse: High-level task text that requires following the visual guide.
63
+ # detailed: Direction- and interaction-specific text, including rotation and push/pull details.
Baseline/checkpoints/060000/action_space_manifest.json ADDED
@@ -0,0 +1,141 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": 1,
3
+ "action_space": "ee_abs_rot6d",
4
+ "source_feature": "action.ee_abs_rot6d",
5
+ "canonical_feature": "action",
6
+ "dtype": "float32",
7
+ "shape": [
8
+ 10
9
+ ],
10
+ "names": [
11
+ "x",
12
+ "y",
13
+ "z",
14
+ "r_col0_x",
15
+ "r_col0_y",
16
+ "r_col0_z",
17
+ "r_col1_x",
18
+ "r_col1_y",
19
+ "r_col1_z",
20
+ "gripper"
21
+ ],
22
+ "dataset_repo_id": "Whalswp/INSIGHTfixposV4_filtered_multispace_v2",
23
+ "dataset_revision": "v3.0",
24
+ "dataset_total_frames": 218367,
25
+ "dataset_total_episodes": 4930,
26
+ "dataset_fps": 10,
27
+ "stats_sha256": "71698c7337070aa8267df5aea7ec08bc00d3925510f2649bc20c9201eeb669e4",
28
+ "stats": {
29
+ "min": [
30
+ -0.21784618496894836,
31
+ -0.4474811851978302,
32
+ 0.1702737659215927,
33
+ -0.9999986290931702,
34
+ -0.9999991655349731,
35
+ -0.9999999403953552,
36
+ -0.9999990463256836,
37
+ -0.9999998211860657,
38
+ -0.9999988079071045,
39
+ 0.0
40
+ ],
41
+ "max": [
42
+ 0.6283698678016663,
43
+ 0.4420371949672699,
44
+ 1.051027774810791,
45
+ 0.9999999403953552,
46
+ 0.9999995231628418,
47
+ 1.0,
48
+ 0.9999995231628418,
49
+ 1.0,
50
+ 0.9999996423721313,
51
+ 0.03999999910593033
52
+ ],
53
+ "mean": [
54
+ 0.4040228037735655,
55
+ -0.05247618696989652,
56
+ 0.6993459771218912,
57
+ 0.07011111991957117,
58
+ -0.03520558904460197,
59
+ 0.14470242846953205,
60
+ 0.05491166606739672,
61
+ -0.19073058893378805,
62
+ -0.06988117661871228,
63
+ 0.021167300266524632
64
+ ],
65
+ "std": [
66
+ 0.11861927913357073,
67
+ 0.11318943557281885,
68
+ 0.17849160646987516,
69
+ 0.4812528791886511,
70
+ 0.7487196258736938,
71
+ 0.4251126804030219,
72
+ 0.47797229807769137,
73
+ 0.6267695346407992,
74
+ 0.5782954306815554,
75
+ 0.01996590431050817
76
+ ],
77
+ "count": [
78
+ 218367
79
+ ],
80
+ "q01": [
81
+ 0.23988881827972064,
82
+ -0.10387461883034811,
83
+ 0.6297243923476871,
84
+ -0.43622841955353464,
85
+ -0.6885486527617446,
86
+ -0.14103555388039019,
87
+ -0.4327210063987595,
88
+ -0.9304790915716145,
89
+ -0.3465462258490417,
90
+ -1.000000013351432e-10
91
+ ],
92
+ "q10": [
93
+ 0.3042306436755216,
94
+ -0.09609991620622624,
95
+ 0.6394063072781145,
96
+ -0.3351306352998669,
97
+ -0.5751963015002924,
98
+ -0.08490621283307866,
99
+ -0.33849701676224275,
100
+ -0.8340557712529717,
101
+ -0.29760290790884514,
102
+ -1.000000013351432e-10
103
+ ],
104
+ "q50": [
105
+ 0.4160225263528291,
106
+ -0.06267459152989589,
107
+ 0.6811777491060801,
108
+ 0.0543536902230377,
109
+ -0.14012251058819217,
110
+ 0.04772394280004727,
111
+ 0.03838119138716219,
112
+ -0.14865169094195635,
113
+ -0.2073069401172425,
114
+ 0.020772594934870346
115
+ ],
116
+ "q90": [
117
+ 0.46641008755679864,
118
+ 0.020233143101890064,
119
+ 0.8095135405259374,
120
+ 0.48172401592474906,
121
+ 0.7398737043070319,
122
+ 0.5970841215636613,
123
+ 0.4674752875999555,
124
+ 0.36281754777672676,
125
+ 0.5033361883834496,
126
+ 0.039998362213352394
127
+ ],
128
+ "q99": [
129
+ 0.47320723171849244,
130
+ 0.04114262476135705,
131
+ 0.8467967259270389,
132
+ 0.5867781409980988,
133
+ 0.8529584859257228,
134
+ 0.7953936054341789,
135
+ 0.5768269858732034,
136
+ 0.49353349323389695,
137
+ 0.7168914440776504,
138
+ 0.03999983541667277
139
+ ]
140
+ }
141
+ }
Baseline/checkpoints/060000/prompt_manifest.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": 1,
3
+ "prompt_set": "detailed",
4
+ "source_file": "task_prompts.json",
5
+ "source_sha256": "0ba48fb372da9f49dcfcd7dcc2da8cce6ce991d3661060e664a104778d1ffacc",
6
+ "mapping_sha256": "f71954786e94cdeceb135dde09fdc6b65db0b87ab3b09d25060129d0f0fde018",
7
+ "dataset_repo_id": "Whalswp/INSIGHTfixposV4_filtered_multispace_v2",
8
+ "dataset_revision": "v3.0",
9
+ "task_count": 13,
10
+ "task_codes": [
11
+ "5f",
12
+ "5g",
13
+ "1ext",
14
+ "3b",
15
+ "5b",
16
+ "3a",
17
+ "3c",
18
+ "5h",
19
+ "5d",
20
+ "3d",
21
+ "5e",
22
+ "5a",
23
+ "5c"
24
+ ],
25
+ "prompts": {
26
+ "5f": "Close the bottle in clockwise direction.",
27
+ "5g": "Grip the cap on the sides indicated by the 'squeeze' arrow and open the bottle in clockwise direction.",
28
+ "1ext": "Find the arrow guide and open the indicated drawer.",
29
+ "3b": "Open the door, rotate counter-clockwise and push.",
30
+ "5b": "Open the bottle in counter-clockwise direction.",
31
+ "3a": "Open the door, rotate clockwise and push.",
32
+ "3c": "Open the door, rotate clockwise and pull.",
33
+ "5h": "Close the bottle in counter-clockwise direction.",
34
+ "5d": "Close the bottle in clockwise direction.",
35
+ "3d": "Open the door, rotate counter-clockwise and pull.",
36
+ "5e": "Close the bottle in counter-clockwise direction.",
37
+ "5a": "Grip the cap on the sides indicated by the 'squeeze' arrow and open the bottle in counter-clockwise direction.",
38
+ "5c": "Open the bottle in clockwise direction."
39
+ }
40
+ }
Baseline/checkpoints/060000/source_config.yaml ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # INSIGHT GR00T N1.7 training config
2
+ # INSIGHT wrapper-only fields are stripped before LeRobot config parsing.
3
+ insight:
4
+ # Options: joint_abs | ee_abs_quat | ee_abs_rot6d | ee_delta
5
+ action_space: ee_abs_rot6d
6
+ # Options: coarse | detailed
7
+ prompt_set: detailed
8
+ save_steps: ["50K"]
9
+
10
+ dataset:
11
+ repo_id: Whalswp/INSIGHTfixposV4_filtered_multispace_v2
12
+ root: ${HOME}/INSIGHTfixposV4_filtered_multispace_v2
13
+
14
+ policy:
15
+ type: groot
16
+ device: cuda
17
+ chunk_size: 16
18
+ n_action_steps: 16
19
+ push_to_hub: false
20
+ # Trainable scope. This baseline trains projector, diffusion, and VLLN; LLM/vision stay frozen.
21
+ # VLM
22
+ tune_llm: false
23
+ tune_visual: false
24
+ tune_top_llm_layers: 0
25
+ # Action Expert
26
+ tune_projector: true
27
+ tune_diffusion_model: true
28
+ tune_vlln: true
29
+
30
+
31
+ seed: 42
32
+ batch_size: 64
33
+ steps: 60000
34
+ log_freq: 200
35
+ output_dir: ${HOME}/groot_insight/Abs_6D/Baseline
36
+ job_name: INSIGHT_6D_baseline
37
+
38
+ wandb:
39
+ enable: true
40
+ disable_artifact: true
41
+
42
+ # Omitted because these match LeRobot/GR00T defaults or are unused in this setup:
43
+ #
44
+ # dataset:
45
+ # eval_split: 0.0
46
+ # image_transforms:
47
+ # enable: false # default; set true only for train-image augmentation.
48
+ #
49
+ # policy:
50
+ # base_model_path: nvidia/GR00T-N1.7-3B
51
+ # embodiment_tag: new_embodiment
52
+ # use_relative_actions: false
53
+ # use_bf16: true
54
+ # repo_id: Whalswp/INSIGHT # unused while push_to_hub is false
55
+ #
56
+ # save_checkpoint: true
57
+ # use_policy_training_preset: true
58
+ # eval_steps: 0
59
+ # env_eval_freq: 0
60
+ #
61
+ # insight.prompt_set options:
62
+ # coarse: High-level task text that requires following the visual guide.
63
+ # detailed: Direction- and interaction-specific text, including rotation and push/pull details.
Baseline/prompt_manifest.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": 1,
3
+ "prompt_set": "detailed",
4
+ "source_file": "task_prompts.json",
5
+ "source_sha256": "0ba48fb372da9f49dcfcd7dcc2da8cce6ce991d3661060e664a104778d1ffacc",
6
+ "mapping_sha256": "f71954786e94cdeceb135dde09fdc6b65db0b87ab3b09d25060129d0f0fde018",
7
+ "dataset_repo_id": "Whalswp/INSIGHTfixposV4_filtered_multispace_v2",
8
+ "dataset_revision": "v3.0",
9
+ "task_count": 13,
10
+ "task_codes": [
11
+ "5f",
12
+ "5g",
13
+ "1ext",
14
+ "3b",
15
+ "5b",
16
+ "3a",
17
+ "3c",
18
+ "5h",
19
+ "5d",
20
+ "3d",
21
+ "5e",
22
+ "5a",
23
+ "5c"
24
+ ],
25
+ "prompts": {
26
+ "5f": "Close the bottle in clockwise direction.",
27
+ "5g": "Grip the cap on the sides indicated by the 'squeeze' arrow and open the bottle in clockwise direction.",
28
+ "1ext": "Find the arrow guide and open the indicated drawer.",
29
+ "3b": "Open the door, rotate counter-clockwise and push.",
30
+ "5b": "Open the bottle in counter-clockwise direction.",
31
+ "3a": "Open the door, rotate clockwise and push.",
32
+ "3c": "Open the door, rotate clockwise and pull.",
33
+ "5h": "Close the bottle in counter-clockwise direction.",
34
+ "5d": "Close the bottle in clockwise direction.",
35
+ "3d": "Open the door, rotate counter-clockwise and pull.",
36
+ "5e": "Close the bottle in counter-clockwise direction.",
37
+ "5a": "Grip the cap on the sides indicated by the 'squeeze' arrow and open the bottle in counter-clockwise direction.",
38
+ "5c": "Open the bottle in clockwise direction."
39
+ }
40
+ }
Baseline/resolved_config.yaml ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ dataset:
2
+ repo_id: Whalswp/INSIGHTfixposV4_filtered_multispace_v2
3
+ root: /home/ext_minje/INSIGHTfixposV4_filtered_multispace_v2
4
+ policy:
5
+ type: groot
6
+ device: cuda
7
+ chunk_size: 16
8
+ n_action_steps: 16
9
+ push_to_hub: false
10
+ tune_llm: false
11
+ tune_visual: false
12
+ tune_top_llm_layers: 0
13
+ tune_projector: true
14
+ tune_diffusion_model: true
15
+ tune_vlln: true
16
+ seed: 42
17
+ batch_size: 64
18
+ steps: 60000
19
+ log_freq: 200
20
+ output_dir: /home/ext_minje/groot_insight/Abs_6D/Baseline
21
+ job_name: INSIGHT_6D_baseline
22
+ wandb:
23
+ enable: true
24
+ disable_artifact: true
Baseline/source_config.yaml ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # INSIGHT GR00T N1.7 training config
2
+ # INSIGHT wrapper-only fields are stripped before LeRobot config parsing.
3
+ insight:
4
+ # Options: joint_abs | ee_abs_quat | ee_abs_rot6d | ee_delta
5
+ action_space: ee_abs_rot6d
6
+ # Options: coarse | detailed
7
+ prompt_set: detailed
8
+ save_steps: ["50K"]
9
+
10
+ dataset:
11
+ repo_id: Whalswp/INSIGHTfixposV4_filtered_multispace_v2
12
+ root: ${HOME}/INSIGHTfixposV4_filtered_multispace_v2
13
+
14
+ policy:
15
+ type: groot
16
+ device: cuda
17
+ chunk_size: 16
18
+ n_action_steps: 16
19
+ push_to_hub: false
20
+ # Trainable scope. This baseline trains projector, diffusion, and VLLN; LLM/vision stay frozen.
21
+ # VLM
22
+ tune_llm: false
23
+ tune_visual: false
24
+ tune_top_llm_layers: 0
25
+ # Action Expert
26
+ tune_projector: true
27
+ tune_diffusion_model: true
28
+ tune_vlln: true
29
+
30
+
31
+ seed: 42
32
+ batch_size: 64
33
+ steps: 60000
34
+ log_freq: 200
35
+ output_dir: ${HOME}/groot_insight/Abs_6D/Baseline
36
+ job_name: INSIGHT_6D_baseline
37
+
38
+ wandb:
39
+ enable: true
40
+ disable_artifact: true
41
+
42
+ # Omitted because these match LeRobot/GR00T defaults or are unused in this setup:
43
+ #
44
+ # dataset:
45
+ # eval_split: 0.0
46
+ # image_transforms:
47
+ # enable: false # default; set true only for train-image augmentation.
48
+ #
49
+ # policy:
50
+ # base_model_path: nvidia/GR00T-N1.7-3B
51
+ # embodiment_tag: new_embodiment
52
+ # use_relative_actions: false
53
+ # use_bf16: true
54
+ # repo_id: Whalswp/INSIGHT # unused while push_to_hub is false
55
+ #
56
+ # save_checkpoint: true
57
+ # use_policy_training_preset: true
58
+ # eval_steps: 0
59
+ # env_eval_freq: 0
60
+ #
61
+ # insight.prompt_set options:
62
+ # coarse: High-level task text that requires following the visual guide.
63
+ # detailed: Direction- and interaction-specific text, including rotation and push/pull details.
Baseline/wandb/debug-internal.log ADDED
The diff for this file is too large to render. See raw diff
 
Baseline/wandb/debug.log ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2026-07-15 14:32:50,279 INFO MainThread:1450496 [wandb_setup.py:_flush():81] Current SDK version is 0.27.2
2
+ 2026-07-15 14:32:50,279 INFO MainThread:1450496 [wandb_setup.py:_flush():81] Configure stats pid to 1450496
3
+ 2026-07-15 14:32:50,279 INFO MainThread:1450496 [wandb_setup.py:_flush():81] Loading settings from environment variables
4
+ 2026-07-15 14:32:50,279 INFO MainThread:1450496 [wandb_init.py:setup_run_log_directory():723] Logging user logs to /home/ext_minje/groot_insight/Abs_6D/Baseline/wandb/run-20260715_143250-1jmjb68j/logs/debug.log
5
+ 2026-07-15 14:32:50,279 INFO MainThread:1450496 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to /home/ext_minje/groot_insight/Abs_6D/Baseline/wandb/run-20260715_143250-1jmjb68j/logs/debug-internal.log
6
+ 2026-07-15 14:32:50,279 INFO MainThread:1450496 [wandb_init.py:init():766] calling init triggers
7
+ 2026-07-15 14:32:50,280 INFO MainThread:1450496 [wandb_init.py:init():771] wandb.init called with sweep_config: {}
8
+ config: {'dataset': {'repo_id': 'Whalswp/INSIGHTfixposV4_filtered_multispace_v2', 'root': '/home/ext_minje/INSIGHTfixposV4_filtered_multispace_v2', 'episodes': None, 'image_transforms': {'enable': False, 'max_num_transforms': 3, 'random_order': False, 'tfs': {'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}, 'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'affine': {'weight': 1.0, 'type': 'RandomAffine', 'kwargs': {'degrees': [-5.0, 5.0], 'translate': [0.05, 0.05]}}}}, 'revision': None, 'use_imagenet_stats': True, 'video_backend': 'torchcodec', 'return_uint8': False, 'depth_output_unit': 'mm', 'streaming': False, 'eval_split': 0.0}, 'env': None, 'policy': {'type': 'groot', 'n_obs_steps': 1, 'input_features': {'observation.state': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [16]}, 'observation.images.wrist': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 224, 224]}, 'observation.images.right_shoulder': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 224, 224]}, 'observation.images.guide': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 224, 224]}}, 'output_features': {'action': {'type': <FeatureType.ACTION: 'ACTION'>, 'shape': [10]}}, 'device': 'cuda', 'use_amp': False, 'use_peft': False, 'push_to_hub': False, 'repo_id': None, 'private': None, 'tags': None, 'license': None, 'pretrained_path': '/home/ext_minje/groot_insight/Abs_6D/Baseline/checkpoints/050000/pretrained_model', 'pretrained_revision': None, 'chunk_size': 16, 'n_action_steps': 16, 'max_state_dim': 132, 'max_action_dim': 132, 'normalization_mapping': {'VISUAL': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'STATE': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'ACTION': <NormalizationMode.IDENTITY: 'IDENTITY'>}, 'base_model_path': 'nvidia/GR00T-N1.7-3B', 'action_decode_transform': None, 'embodiment_tag': 'new_embodiment', 'tune_llm': False, 'tune_visual': False, 'tune_projector': True, 'tune_diffusion_model': True, 'tune_vlln': True, 'tune_top_llm_layers': 0, 'num_inference_timesteps': None, 'rtc_ramp_rate': None, 'use_flash_attention': False, 'use_relative_actions': False, 'relative_exclude_joints': [], 'optimizer_lr': 0.0001, 'optimizer_betas': [0.9, 0.999], 'optimizer_eps': 1e-08, 'optimizer_weight_decay': 1e-05, 'warmup_ratio': 0.05, 'use_bf16': True, 'model_params_fp32': True, 'image_size': [256, 256], 'tokenizer_assets_repo': None, 'lora_rank': 0, 'lora_alpha': 16, 'lora_dropout': 0.1, 'lora_full_model': False, 'video_backend': 'decord', 'balance_dataset_weights': True, 'balance_trajectory_weights': True, 'dataset_paths': None, 'output_dir': './tmp/gr00t', 'save_steps': 1000, 'max_steps': 10000, 'batch_size': 32, 'dataloader_num_workers': 8, 'report_to': 'wandb', 'resume': False}, 'reward_model': None, 'output_dir': '/home/ext_minje/groot_insight/Abs_6D/Baseline', 'job_name': 'INSIGHT_6D_baseline', 'resume': True, 'seed': 42, 'cudnn_deterministic': False, 'num_workers': 4, 'batch_size': 64, 'prefetch_factor': 4, 'persistent_workers': True, 'steps': 60000, 'env_eval_freq': 20000, 'log_freq': 200, 'eval_steps': 0, 'max_eval_samples': 0, 'tolerance_s': 0.0001, 'save_checkpoint': True, 'save_freq': 20000, 'use_policy_training_preset': True, 'optimizer': {'type': 'adamw', 'lr': 0.0001, 'weight_decay': 1e-05, 'grad_clip_norm': 1.0, 'betas': [0.9, 0.999], 'eps': 1e-08}, 'scheduler': {'type': 'diffuser', 'num_warmup_steps': 500, 'name': 'cosine'}, 'eval': {'n_episodes': 50, 'batch_size': 50, 'use_async_envs': True, 'recording': False, 'recording_repo_id': None, 'recording_private': False}, 'wandb': {'enable': True, 'disable_artifact': True, 'project': 'lerobot', 'entity': None, 'notes': None, 'run_id': '1jmjb68j', 'mode': None, 'add_tags': True}, 'peft': None, 'job': {'target': None, 'image': 'huggingface/lerobot-gpu:latest', 'timeout': '2d', 'detach': False, 'tags': []}, 'save_checkpoint_to_hub': False, 'sample_weighting': None, 'rename_map': {}, 'checkpoint_path': '/home/ext_minje/groot_insight/Abs_6D/Baseline/checkpoints/050000', '_wandb': {}}
9
+ 2026-07-15 14:32:50,280 INFO MainThread:1450496 [wandb_init.py:init():814] starting backend
10
+ 2026-07-15 14:32:50,495 INFO MainThread:1450496 [wandb_init.py:init():829] sending inform_init request
11
+ 2026-07-15 14:32:50,989 INFO MainThread:1450496 [wandb_init.py:init():834] backend started and connected
12
+ 2026-07-15 14:32:50,999 INFO MainThread:1450496 [wandb_init.py:init():904] updated telemetry
13
+ 2026-07-15 14:32:51,006 INFO MainThread:1450496 [wandb_init.py:init():927] communicating run to backend with 90.0 second timeout
14
+ 2026-07-15 14:32:51,687 INFO MainThread:1450496 [wandb_init.py:init():967] run resumed
15
+ 2026-07-15 14:32:51,692 INFO MainThread:1450496 [wandb_init.py:init():972] starting run threads in backend
16
+ 2026-07-15 14:32:51,783 INFO MainThread:1450496 [wandb_run.py:_console_start():2523] atexit reg
17
+ 2026-07-15 14:32:51,783 INFO MainThread:1450496 [wandb_run.py:_redirect():2373] redirect: wrap_raw
18
+ 2026-07-15 14:32:51,783 INFO MainThread:1450496 [wandb_run.py:_redirect():2442] Wrapping output streams.
19
+ 2026-07-15 14:32:51,783 INFO MainThread:1450496 [wandb_run.py:_redirect():2465] Redirects installed.
20
+ 2026-07-15 14:32:51,785 INFO MainThread:1450496 [wandb_init.py:init():1010] run started, returning control to user process
21
+ 2026-07-15 19:10:44,422 INFO wandb-AsyncioManager-main:1450496 [service_client.py:_forward_responses():122] Reached EOF.
22
+ 2026-07-15 19:10:44,422 INFO wandb-AsyncioManager-main:1450496 [mailbox.py:close():154] Closing mailbox, abandoning 2 handles.
23
+ 2026-07-15 19:10:44,423 ERROR wandb-AsyncioManager-main:1450496 [asyncio_manager.py:fn_wrap_exceptions():184] Uncaught exception in run_soon callback.
24
+ Traceback (most recent call last):
25
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/lib/asyncio_manager.py", line 182, in fn_wrap_exceptions
26
+ await fn()
27
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/lib/run_messages.py", line 90, in loop
28
+ await asyncio_compat.race(
29
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/lib/asyncio_compat.py", line 278, in race
30
+ async with open_task_group(race=True) as tg:
31
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^
32
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/contextlib.py", line 217, in __aexit__
33
+ await anext(self.gen)
34
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/lib/asyncio_compat.py", line 240, in open_task_group
35
+ await task_group._wait_all(race=race, timeout=exit_timeout)
36
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/lib/asyncio_compat.py", line 180, in _wait_all
37
+ raise exc
38
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/lib/run_messages.py", line 113, in _print_all
39
+ result = await handle.wait_async(timeout=None)
40
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
41
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/mailbox/mailbox_handle.py", line 127, in wait_async
42
+ response = await self._handle.wait_async(timeout=timeout)
43
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
44
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/mailbox/response_handle.py", line 123, in wait_async
45
+ raise HandleAbandonedError()
46
+ wandb.sdk.mailbox.mailbox_handle.HandleAbandonedError
Baseline/wandb/run-20260714_161240-1jmjb68j/files/output.log ADDED
@@ -0,0 +1,269 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ INFO 2026-07-14 16:12:42 db_utils.py:121 Logs will be synced with wandb.
2
+ INFO 2026-07-14 16:12:42 db_utils.py:122 Track this run --> https://wandb.ai/minje227_hyu-hanyang-university/lerobot/runs/1jmjb68j
3
+ INFO 2026-07-14 16:12:42 ot_train.py:277 Creating dataset
4
+ INFO 2026-07-14 16:12:45 ot_train.py:311 Creating policy
5
+ Fetching 27 files: 100%|██████████| 27/27 [00:00<00:00, 4171.90it/s]
6
+ `torch_dtype` is deprecated! Use `dtype` instead!
7
+ Loading weights: 100%|██████████| 1031/1031 [00:00<00:00, 2241.16it/s]
8
+ INFO 2026-07-14 16:12:51 ot_train.py:384 Creating optimizer and scheduler
9
+ INFO 2026-07-14 16:12:51 ot_train.py:416 Output dir: /home/ext_minje/groot_insight/Abs_6D/Baseline
10
+ INFO 2026-07-14 16:12:51 ot_train.py:423 cfg.steps=60000 (60K)
11
+ INFO 2026-07-14 16:12:51 ot_train.py:424 dataset.num_frames=218367 (218K)
12
+ INFO 2026-07-14 16:12:51 ot_train.py:425 dataset.num_episodes=4930
13
+ INFO 2026-07-14 16:12:51 ot_train.py:428 Effective batch size: 64 x 1 = 64
14
+ INFO 2026-07-14 16:12:51 ot_train.py:429 num_learnable_params=1620515968 (2B)
15
+ INFO 2026-07-14 16:12:51 ot_train.py:430 num_total_params=3144016000 (3B)
16
+ Training: 0%| | 0/60000 [00:00<?, ?step/s]INFO 2026-07-14 16:12:51 ot_train.py:581 Start offline training on a fixed dataset, with effective batch size: 64
17
+ Training: 0%| | 200/60000 [04:57<25:13:11, 1.52s/step]INFO 2026-07-14 16:17:48 ot_train.py:625 step:200 smpl:13K ep:289 epch:0.06 loss:0.924 grdn:1.272 lr:2.0e-05 updt_s:0.764 data_s:0.721 smp/s:43 mem_gb:39.61
18
+ Training: 1%| | 400/60000 [09:47<21:50:42, 1.32s/step]INFO 2026-07-14 16:22:38 ot_train.py:625 step:400 smpl:26K ep:578 epch:0.12 loss:0.258 grdn:1.552 lr:6.0e-05 updt_s:0.734 data_s:0.714 smp/s:44 mem_gb:39.63
19
+ Training: 1%| | 600/60000 [14:37<22:35:00, 1.37s/step]INFO 2026-07-14 16:27:28 ot_train.py:625 step:600 smpl:38K ep:867 epch:0.18 loss:0.199 grdn:1.034 lr:9.5e-05 updt_s:0.734 data_s:0.711 smp/s:44 mem_gb:39.63
20
+ Training: 1%|▏ | 800/60000 [19:30<23:53:10, 1.45s/step]INFO 2026-07-14 16:32:21 ot_train.py:625 step:800 smpl:51K ep:1K epch:0.23 loss:0.163 grdn:0.780 lr:1.0e-04 updt_s:0.731 data_s:0.731 smp/s:44 mem_gb:39.63
21
+ Training: 2%|▏ | 1000/60000 [24:22<26:08:09, 1.59s/step]INFO 2026-07-14 16:37:13 ot_train.py:625 step:1K smpl:64K ep:1K epch:0.29 loss:0.147 grdn:0.695 lr:1.0e-04 updt_s:0.735 data_s:0.721 smp/s:44 mem_gb:39.63
22
+ Training: 2%|▏ | 1200/60000 [29:14<23:30:02, 1.44s/step]INFO 2026-07-14 16:42:06 ot_train.py:625 step:1K smpl:77K ep:2K epch:0.35 loss:0.132 grdn:0.624 lr:1.0e-04 updt_s:0.741 data_s:0.719 smp/s:44 mem_gb:39.63
23
+ Training: 2%|▏ | 1400/60000 [34:02<22:35:14, 1.39s/step]INFO 2026-07-14 16:46:53 ot_train.py:625 step:1K smpl:90K ep:2K epch:0.41 loss:0.123 grdn:0.569 lr:1.0e-04 updt_s:0.723 data_s:0.712 smp/s:45 mem_gb:39.63
24
+ Training: 3%|▎ | 1600/60000 [38:48<24:19:48, 1.50s/step]INFO 2026-07-14 16:51:40 ot_train.py:625 step:2K smpl:102K ep:2K epch:0.47 loss:0.119 grdn:0.548 lr:1.0e-04 updt_s:0.731 data_s:0.698 smp/s:45 mem_gb:39.63
25
+ Training: 3%|▎ | 1800/60000 [43:34<21:35:33, 1.34s/step]INFO 2026-07-14 16:56:26 ot_train.py:625 step:2K smpl:115K ep:3K epch:0.53 loss:0.115 grdn:0.516 lr:1.0e-04 updt_s:0.728 data_s:0.700 smp/s:45 mem_gb:39.63
26
+ Training: 3%|▎ | 2000/60000 [48:21<22:37:52, 1.40s/step]INFO 2026-07-14 17:01:12 ot_train.py:625 step:2K smpl:128K ep:3K epch:0.59 loss:0.110 grdn:0.486 lr:1.0e-04 updt_s:0.734 data_s:0.697 smp/s:45 mem_gb:39.63
27
+ Training: 4%|▎ | 2200/60000 [53:16<27:27:37, 1.71s/step]INFO 2026-07-14 17:06:07 ot_train.py:625 step:2K smpl:141K ep:3K epch:0.64 loss:0.108 grdn:0.488 lr:1.0e-04 updt_s:0.736 data_s:0.735 smp/s:43 mem_gb:39.63
28
+ Training: 4%|▍ | 2400/60000 [57:52<20:49:34, 1.30s/step]INFO 2026-07-14 17:10:43 ot_train.py:625 step:2K smpl:154K ep:3K epch:0.70 loss:0.104 grdn:0.448 lr:1.0e-04 updt_s:0.722 data_s:0.654 smp/s:47 mem_gb:39.62
29
+ Training: 4%|▍ | 2600/60000 [1:02:13<20:53:12, 1.31s/step]INFO 2026-07-14 17:15:04 ot_train.py:625 step:3K smpl:166K ep:4K epch:0.76 loss:0.105 grdn:0.451 lr:1.0e-04 updt_s:0.684 data_s:0.617 smp/s:49 mem_gb:39.64
30
+ Training: 5%|▍ | 2800/60000 [1:07:04<23:58:13, 1.51s/step]INFO 2026-07-14 17:19:55 ot_train.py:625 step:3K smpl:179K ep:4K epch:0.82 loss:0.100 grdn:0.426 lr:1.0e-04 updt_s:0.736 data_s:0.715 smp/s:44 mem_gb:39.64
31
+ Training: 5%|▌ | 3000/60000 [1:12:20<24:00:49, 1.52s/step]INFO 2026-07-14 17:25:11 ot_train.py:625 step:3K smpl:192K ep:4K epch:0.88 loss:0.099 grdn:0.413 lr:1.0e-04 updt_s:0.772 data_s:0.803 smp/s:41 mem_gb:39.64
32
+ Training: 5%|▌ | 3200/60000 [1:16:55<22:01:57, 1.40s/step]INFO 2026-07-14 17:29:46 ot_train.py:625 step:3K smpl:205K ep:5K epch:0.94 loss:0.098 grdn:0.405 lr:1.0e-04 updt_s:0.708 data_s:0.667 smp/s:47 mem_gb:39.64
33
+ Training: 6%|▌ | 3400/60000 [1:22:12<25:28:17, 1.62s/step]INFO 2026-07-14 17:35:03 ot_train.py:625 step:3K smpl:218K ep:5K epch:1.00 loss:0.095 grdn:0.389 lr:9.9e-05 updt_s:0.767 data_s:0.811 smp/s:41 mem_gb:39.64
34
+ Training: 6%|▌ | 3600/60000 [1:27:35<22:56:31, 1.46s/step]INFO 2026-07-14 17:40:26 ot_train.py:625 step:4K smpl:230K ep:5K epch:1.06 loss:0.094 grdn:0.384 lr:9.9e-05 updt_s:0.778 data_s:0.835 smp/s:40 mem_gb:39.64
35
+ Training: 6%|▋ | 3800/60000 [1:32:56<25:08:26, 1.61s/step]INFO 2026-07-14 17:45:47 ot_train.py:625 step:4K smpl:243K ep:5K epch:1.11 loss:0.094 grdn:0.389 lr:9.9e-05 updt_s:0.780 data_s:0.822 smp/s:40 mem_gb:39.64
36
+ Training: 7%|▋ | 4000/60000 [1:38:22<26:54:50, 1.73s/step]INFO 2026-07-14 17:51:13 ot_train.py:625 step:4K smpl:256K ep:6K epch:1.17 loss:0.093 grdn:0.374 lr:9.9e-05 updt_s:0.792 data_s:0.836 smp/s:39 mem_gb:39.64
37
+ Training: 7%|▋ | 4200/60000 [1:43:51<25:40:34, 1.66s/step]INFO 2026-07-14 17:56:42 ot_train.py:625 step:4K smpl:269K ep:6K epch:1.23 loss:0.090 grdn:0.355 lr:9.9e-05 updt_s:0.785 data_s:0.854 smp/s:39 mem_gb:39.64
38
+ Training: 7%|▋ | 4400/60000 [1:49:18<25:19:11, 1.64s/step]INFO 2026-07-14 18:02:09 ot_train.py:625 step:4K smpl:282K ep:6K epch:1.29 loss:0.090 grdn:0.361 lr:9.9e-05 updt_s:0.788 data_s:0.843 smp/s:39 mem_gb:39.63
39
+ Training: 8%|▊ | 4600/60000 [1:54:41<25:34:40, 1.66s/step]INFO 2026-07-14 18:07:32 ot_train.py:625 step:5K smpl:294K ep:7K epch:1.35 loss:0.090 grdn:0.359 lr:9.9e-05 updt_s:0.777 data_s:0.834 smp/s:40 mem_gb:39.62
40
+ Training: 8%|▊ | 4800/60000 [2:00:00<24:06:01, 1.57s/step]INFO 2026-07-14 18:12:51 ot_train.py:625 step:5K smpl:307K ep:7K epch:1.41 loss:0.089 grdn:0.369 lr:9.9e-05 updt_s:0.769 data_s:0.823 smp/s:40 mem_gb:39.64
41
+ Training: 8%|▊ | 5000/60000 [2:04:54<18:21:51, 1.20s/step]INFO 2026-07-14 18:17:46 ot_train.py:625 step:5K smpl:320K ep:7K epch:1.47 loss:0.089 grdn:0.357 lr:9.9e-05 updt_s:0.745 data_s:0.725 smp/s:44 mem_gb:39.64
42
+ Training: 9%|▊ | 5200/60000 [2:09:11<22:04:04, 1.45s/step]INFO 2026-07-14 18:22:02 ot_train.py:625 step:5K smpl:333K ep:8K epch:1.52 loss:0.086 grdn:0.338 lr:9.9e-05 updt_s:0.688 data_s:0.589 smp/s:50 mem_gb:39.64
43
+ Training: 9%|▉ | 5400/60000 [2:13:52<18:51:33, 1.24s/step]INFO 2026-07-14 18:26:43 ot_train.py:625 step:5K smpl:346K ep:8K epch:1.58 loss:0.084 grdn:0.322 lr:9.8e-05 updt_s:0.728 data_s:0.675 smp/s:46 mem_gb:39.64
44
+ Training: 9%|▉ | 5600/60000 [2:18:31<21:32:04, 1.43s/step]INFO 2026-07-14 18:31:22 ot_train.py:625 step:6K smpl:358K ep:8K epch:1.64 loss:0.085 grdn:0.331 lr:9.8e-05 updt_s:0.719 data_s:0.674 smp/s:46 mem_gb:39.64
45
+ Training: 10%|▉ | 5800/60000 [2:23:00<19:34:55, 1.30s/step]INFO 2026-07-14 18:35:52 ot_train.py:625 step:6K smpl:371K ep:8K epch:1.70 loss:0.085 grdn:0.334 lr:9.8e-05 updt_s:0.702 data_s:0.642 smp/s:48 mem_gb:39.64
46
+ Training: 10%|█ | 6000/60000 [2:27:14<18:52:42, 1.26s/step]INFO 2026-07-14 18:40:05 ot_train.py:625 step:6K smpl:384K ep:9K epch:1.76 loss:0.082 grdn:0.329 lr:9.8e-05 updt_s:0.669 data_s:0.596 smp/s:51 mem_gb:39.63
47
+ Training: 10%|█ | 6200/60000 [2:31:29<18:57:12, 1.27s/step]INFO 2026-07-14 18:44:20 ot_train.py:625 step:6K smpl:397K ep:9K epch:1.82 loss:0.085 grdn:0.346 lr:9.8e-05 updt_s:0.673 data_s:0.600 smp/s:50 mem_gb:39.64
48
+ Training: 11%|█ | 6400/60000 [2:35:43<19:12:50, 1.29s/step]INFO 2026-07-14 18:48:34 ot_train.py:625 step:6K smpl:410K ep:9K epch:1.88 loss:0.084 grdn:0.332 lr:9.8e-05 updt_s:0.670 data_s:0.596 smp/s:51 mem_gb:39.64
49
+ Training: 11%|█ | 6600/60000 [2:39:53<18:06:25, 1.22s/step]INFO 2026-07-14 18:52:45 ot_train.py:625 step:7K smpl:422K ep:10K epch:1.93 loss:0.079 grdn:0.300 lr:9.8e-05 updt_s:0.669 data_s:0.582 smp/s:51 mem_gb:39.64
50
+ Training: 11%|█▏ | 6800/60000 [2:44:06<18:20:51, 1.24s/step]INFO 2026-07-14 18:56:57 ot_train.py:625 step:7K smpl:435K ep:10K epch:1.99 loss:0.080 grdn:0.314 lr:9.7e-05 updt_s:0.665 data_s:0.596 smp/s:51 mem_gb:39.62
51
+ Training: 12%|█▏ | 7000/60000 [2:48:25<18:42:17, 1.27s/step]INFO 2026-07-14 19:01:16 ot_train.py:625 step:7K smpl:448K ep:10K epch:2.05 loss:0.078 grdn:0.316 lr:9.7e-05 updt_s:0.682 data_s:0.609 smp/s:50 mem_gb:39.64
52
+ Training: 12%|█▏ | 7200/60000 [2:52:42<19:19:08, 1.32s/step]INFO 2026-07-14 19:05:33 ot_train.py:625 step:7K smpl:461K ep:10K epch:2.11 loss:0.078 grdn:0.307 lr:9.7e-05 updt_s:0.686 data_s:0.593 smp/s:50 mem_gb:39.64
53
+ Training: 12%|█▏ | 7400/60000 [2:57:04<18:49:23, 1.29s/step]INFO 2026-07-14 19:09:55 ot_train.py:625 step:7K smpl:474K ep:11K epch:2.17 loss:0.077 grdn:0.291 lr:9.7e-05 updt_s:0.693 data_s:0.617 smp/s:49 mem_gb:39.64
54
+ Training: 13%|█▎ | 7600/60000 [3:01:28<18:51:21, 1.30s/step]INFO 2026-07-14 19:14:19 ot_train.py:625 step:8K smpl:486K ep:11K epch:2.23 loss:0.079 grdn:0.329 lr:9.7e-05 updt_s:0.694 data_s:0.621 smp/s:49 mem_gb:39.64
55
+ Training: 13%|█▎ | 7800/60000 [3:05:46<18:19:01, 1.26s/step]INFO 2026-07-14 19:18:38 ot_train.py:625 step:8K smpl:499K ep:11K epch:2.29 loss:0.080 grdn:0.329 lr:9.6e-05 updt_s:0.683 data_s:0.606 smp/s:50 mem_gb:39.64
56
+ Training: 13%|█▎ | 8000/60000 [3:10:03<18:01:27, 1.25s/step]INFO 2026-07-14 19:22:54 ot_train.py:625 step:8K smpl:512K ep:12K epch:2.34 loss:0.081 grdn:0.327 lr:9.6e-05 updt_s:0.686 data_s:0.592 smp/s:50 mem_gb:39.64
57
+ Training: 14%|█▎ | 8200/60000 [3:14:22<18:28:54, 1.28s/step]INFO 2026-07-14 19:27:13 ot_train.py:625 step:8K smpl:525K ep:12K epch:2.40 loss:0.077 grdn:0.301 lr:9.6e-05 updt_s:0.689 data_s:0.604 smp/s:49 mem_gb:39.64
58
+ Training: 14%|█▍ | 8400/60000 [3:18:42<19:26:41, 1.36s/step]INFO 2026-07-14 19:31:34 ot_train.py:625 step:8K smpl:538K ep:12K epch:2.46 loss:0.075 grdn:0.294 lr:9.6e-05 updt_s:0.689 data_s:0.612 smp/s:49 mem_gb:39.64
59
+ Training: 14%|█▍ | 8600/60000 [3:23:03<18:46:25, 1.31s/step]INFO 2026-07-14 19:35:54 ot_train.py:625 step:9K smpl:550K ep:12K epch:2.52 loss:0.076 grdn:0.297 lr:9.6e-05 updt_s:0.687 data_s:0.613 smp/s:49 mem_gb:39.64
60
+ Training: 15%|█▍ | 8800/60000 [3:27:30<18:50:26, 1.32s/step]INFO 2026-07-14 19:40:21 ot_train.py:625 step:9K smpl:563K ep:13K epch:2.58 loss:0.075 grdn:0.305 lr:9.5e-05 updt_s:0.711 data_s:0.622 smp/s:48 mem_gb:39.63
61
+ Training: 15%|█▌ | 9000/60000 [3:31:54<18:56:39, 1.34s/step]INFO 2026-07-14 19:44:45 ot_train.py:625 step:9K smpl:576K ep:13K epch:2.64 loss:0.075 grdn:0.296 lr:9.5e-05 updt_s:0.690 data_s:0.626 smp/s:49 mem_gb:39.64
62
+ Training: 15%|█▌ | 9200/60000 [3:36:15<18:15:36, 1.29s/step]INFO 2026-07-14 19:49:06 ot_train.py:625 step:9K smpl:589K ep:13K epch:2.70 loss:0.072 grdn:0.288 lr:9.5e-05 updt_s:0.686 data_s:0.615 smp/s:49 mem_gb:39.62
63
+ Training: 16%|█▌ | 9400/60000 [3:40:27<17:40:11, 1.26s/step]INFO 2026-07-14 19:53:18 ot_train.py:625 step:9K smpl:602K ep:14K epch:2.75 loss:0.076 grdn:0.331 lr:9.5e-05 updt_s:0.670 data_s:0.587 smp/s:51 mem_gb:39.64
64
+ Training: 16%|█▌ | 9600/60000 [3:44:45<17:53:30, 1.28s/step]INFO 2026-07-14 19:57:36 ot_train.py:625 step:10K smpl:614K ep:14K epch:2.81 loss:0.073 grdn:0.299 lr:9.4e-05 updt_s:0.682 data_s:0.605 smp/s:50 mem_gb:39.64
65
+ Training: 16%|█▋ | 9800/60000 [3:49:05<17:46:19, 1.27s/step]INFO 2026-07-14 20:01:56 ot_train.py:625 step:10K smpl:627K ep:14K epch:2.87 loss:0.075 grdn:0.316 lr:9.4e-05 updt_s:0.682 data_s:0.615 smp/s:49 mem_gb:39.64
66
+ Training: 17%|█▋ | 10000/60000 [3:53:23<17:37:11, 1.27s/step]INFO 2026-07-14 20:06:14 ot_train.py:625 step:10K smpl:640K ep:14K epch:2.93 loss:0.071 grdn:0.292 lr:9.4e-05 updt_s:0.681 data_s:0.609 smp/s:50 mem_gb:39.64
67
+ Training: 17%|█▋ | 10200/60000 [3:57:43<18:16:00, 1.32s/step]INFO 2026-07-14 20:10:34 ot_train.py:625 step:10K smpl:653K ep:15K epch:2.99 loss:0.075 grdn:0.321 lr:9.4e-05 updt_s:0.681 data_s:0.616 smp/s:49 mem_gb:39.63
68
+ Training: 17%|█▋ | 10400/60000 [4:02:02<17:42:10, 1.28s/step]INFO 2026-07-14 20:14:53 ot_train.py:625 step:10K smpl:666K ep:15K epch:3.05 loss:0.072 grdn:0.289 lr:9.3e-05 updt_s:0.679 data_s:0.611 smp/s:50 mem_gb:39.64
69
+ Training: 18%|█▊ | 10600/60000 [4:06:08<16:46:22, 1.22s/step]INFO 2026-07-14 20:19:00 ot_train.py:625 step:11K smpl:678K ep:15K epch:3.11 loss:0.071 grdn:0.282 lr:9.3e-05 updt_s:0.660 data_s:0.571 smp/s:52 mem_gb:39.64
70
+ Training: 18%|█▊ | 10800/60000 [4:10:16<16:44:40, 1.23s/step]INFO 2026-07-14 20:23:07 ot_train.py:625 step:11K smpl:691K ep:16K epch:3.17 loss:0.071 grdn:0.279 lr:9.3e-05 updt_s:0.660 data_s:0.574 smp/s:52 mem_gb:39.63
71
+ Training: 18%|█▊ | 11000/60000 [4:14:24<16:49:58, 1.24s/step]INFO 2026-07-14 20:27:16 ot_train.py:625 step:11K smpl:704K ep:16K epch:3.22 loss:0.067 grdn:0.270 lr:9.3e-05 updt_s:0.663 data_s:0.578 smp/s:52 mem_gb:39.64
72
+ Training: 19%|█▊ | 11200/60000 [4:18:36<16:38:47, 1.23s/step]INFO 2026-07-14 20:31:27 ot_train.py:625 step:11K smpl:717K ep:16K epch:3.28 loss:0.067 grdn:0.274 lr:9.2e-05 updt_s:0.665 data_s:0.592 smp/s:51 mem_gb:39.64
73
+ Training: 19%|█▉ | 11400/60000 [4:22:47<16:26:56, 1.22s/step]INFO 2026-07-14 20:35:38 ot_train.py:625 step:11K smpl:730K ep:16K epch:3.34 loss:0.070 grdn:0.292 lr:9.2e-05 updt_s:0.658 data_s:0.591 smp/s:51 mem_gb:39.62
74
+ Training: 19%|█▉ | 11600/60000 [4:27:00<17:21:50, 1.29s/step]INFO 2026-07-14 20:39:51 ot_train.py:625 step:12K smpl:742K ep:17K epch:3.40 loss:0.067 grdn:0.283 lr:9.2e-05 updt_s:0.667 data_s:0.595 smp/s:51 mem_gb:39.64
75
+ Training: 20%|█▉ | 11800/60000 [4:31:17<16:16:30, 1.22s/step]INFO 2026-07-14 20:44:08 ot_train.py:625 step:12K smpl:755K ep:17K epch:3.46 loss:0.070 grdn:0.290 lr:9.2e-05 updt_s:0.675 data_s:0.608 smp/s:50 mem_gb:39.64
76
+ Training: 20%|██ | 12000/60000 [4:35:26<16:27:26, 1.23s/step]INFO 2026-07-14 20:48:17 ot_train.py:625 step:12K smpl:768K ep:17K epch:3.52 loss:0.071 grdn:0.295 lr:9.1e-05 updt_s:0.657 data_s:0.587 smp/s:51 mem_gb:39.64
77
+ Training: 20%|██ | 12200/60000 [4:39:32<17:11:03, 1.29s/step]INFO 2026-07-14 20:52:23 ot_train.py:625 step:12K smpl:781K ep:18K epch:3.58 loss:0.067 grdn:0.274 lr:9.1e-05 updt_s:0.655 data_s:0.570 smp/s:52 mem_gb:39.64
78
+ Training: 21%|█�� | 12400/60000 [4:43:39<16:03:48, 1.21s/step]INFO 2026-07-14 20:56:30 ot_train.py:625 step:12K smpl:794K ep:18K epch:3.63 loss:0.069 grdn:0.297 lr:9.1e-05 updt_s:0.657 data_s:0.576 smp/s:52 mem_gb:39.64
79
+ Training: 21%|██ | 12600/60000 [4:47:50<16:04:53, 1.22s/step]INFO 2026-07-14 21:00:41 ot_train.py:625 step:13K smpl:806K ep:18K epch:3.69 loss:0.066 grdn:0.270 lr:9.0e-05 updt_s:0.661 data_s:0.591 smp/s:51 mem_gb:39.64
80
+ Training: 21%|██▏ | 12800/60000 [4:52:03<16:09:50, 1.23s/step]INFO 2026-07-14 21:04:54 ot_train.py:625 step:13K smpl:819K ep:18K epch:3.75 loss:0.066 grdn:0.275 lr:9.0e-05 updt_s:0.666 data_s:0.596 smp/s:51 mem_gb:39.64
81
+ Training: 22%|██▏ | 13000/60000 [4:56:17<17:16:44, 1.32s/step]INFO 2026-07-14 21:09:08 ot_train.py:625 step:13K smpl:832K ep:19K epch:3.81 loss:0.067 grdn:0.284 lr:9.0e-05 updt_s:0.673 data_s:0.594 smp/s:50 mem_gb:39.63
82
+ Training: 22%|██▏ | 13200/60000 [5:00:26<16:13:10, 1.25s/step]INFO 2026-07-14 21:13:17 ot_train.py:625 step:13K smpl:845K ep:19K epch:3.87 loss:0.070 grdn:0.306 lr:8.9e-05 updt_s:0.661 data_s:0.583 smp/s:51 mem_gb:39.63
83
+ Training: 22%|██▏ | 13400/60000 [5:04:37<15:56:23, 1.23s/step]INFO 2026-07-14 21:17:29 ot_train.py:625 step:13K smpl:858K ep:19K epch:3.93 loss:0.068 grdn:0.301 lr:8.9e-05 updt_s:0.668 data_s:0.587 smp/s:51 mem_gb:39.64
84
+ Training: 23%|██▎ | 13600/60000 [5:08:44<15:34:44, 1.21s/step]INFO 2026-07-14 21:21:35 ot_train.py:625 step:14K smpl:870K ep:20K epch:3.99 loss:0.067 grdn:0.281 lr:8.9e-05 updt_s:0.652 data_s:0.578 smp/s:52 mem_gb:39.62
85
+ Training: 23%|██▎ | 13800/60000 [5:12:51<16:02:46, 1.25s/step]INFO 2026-07-14 21:25:42 ot_train.py:625 step:14K smpl:883K ep:20K epch:4.04 loss:0.065 grdn:0.268 lr:8.8e-05 updt_s:0.656 data_s:0.575 smp/s:52 mem_gb:39.64
86
+ Training: 23%|██▎ | 14000/60000 [5:17:03<15:27:28, 1.21s/step]INFO 2026-07-14 21:29:54 ot_train.py:625 step:14K smpl:896K ep:20K epch:4.10 loss:0.064 grdn:0.285 lr:8.8e-05 updt_s:0.665 data_s:0.591 smp/s:51 mem_gb:39.64
87
+ Training: 24%|██▎ | 14200/60000 [5:21:15<15:36:13, 1.23s/step]INFO 2026-07-14 21:34:06 ot_train.py:625 step:14K smpl:909K ep:21K epch:4.16 loss:0.066 grdn:0.281 lr:8.8e-05 updt_s:0.659 data_s:0.599 smp/s:51 mem_gb:39.64
88
+ Training: 24%|██▍ | 14400/60000 [5:25:33<16:34:18, 1.31s/step]INFO 2026-07-14 21:38:24 ot_train.py:625 step:14K smpl:922K ep:21K epch:4.22 loss:0.064 grdn:0.272 lr:8.7e-05 updt_s:0.682 data_s:0.605 smp/s:50 mem_gb:39.64
89
+ Training: 24%|██▍ | 14600/60000 [5:29:45<15:40:37, 1.24s/step]INFO 2026-07-14 21:42:36 ot_train.py:625 step:15K smpl:934K ep:21K epch:4.28 loss:0.067 grdn:0.315 lr:8.7e-05 updt_s:0.665 data_s:0.593 smp/s:51 mem_gb:39.63
90
+ Training: 25%|██▍ | 14800/60000 [5:34:00<15:40:37, 1.25s/step]INFO 2026-07-14 21:46:51 ot_train.py:625 step:15K smpl:947K ep:21K epch:4.34 loss:0.064 grdn:0.279 lr:8.7e-05 updt_s:0.671 data_s:0.599 smp/s:50 mem_gb:39.63
91
+ Training: 25%|██▌ | 15000/60000 [5:38:06<15:14:31, 1.22s/step]INFO 2026-07-14 21:50:57 ot_train.py:625 step:15K smpl:960K ep:22K epch:4.40 loss:0.064 grdn:0.267 lr:8.6e-05 updt_s:0.655 data_s:0.572 smp/s:52 mem_gb:39.64
92
+ Training: 25%|██▌ | 15200/60000 [5:42:12<14:50:06, 1.19s/step]INFO 2026-07-14 21:55:03 ot_train.py:625 step:15K smpl:973K ep:22K epch:4.45 loss:0.070 grdn:0.321 lr:8.6e-05 updt_s:0.657 data_s:0.571 smp/s:52 mem_gb:39.64
93
+ Training: 26%|██▌ | 15400/60000 [5:46:20<15:10:47, 1.23s/step]INFO 2026-07-14 21:59:12 ot_train.py:625 step:15K smpl:986K ep:22K epch:4.51 loss:0.064 grdn:0.269 lr:8.5e-05 updt_s:0.660 data_s:0.581 smp/s:52 mem_gb:39.64
94
+ Training: 26%|██▌ | 15600/60000 [5:50:34<15:03:47, 1.22s/step]INFO 2026-07-14 22:03:25 ot_train.py:625 step:16K smpl:998K ep:23K epch:4.57 loss:0.063 grdn:0.274 lr:8.5e-05 updt_s:0.662 data_s:0.604 smp/s:51 mem_gb:39.64
95
+ Training: 26%|██▋ | 15800/60000 [5:54:50<19:01:44, 1.55s/step]INFO 2026-07-14 22:07:42 ot_train.py:625 step:16K smpl:1M ep:23K epch:4.63 loss:0.062 grdn:0.271 lr:8.5e-05 updt_s:0.671 data_s:0.607 smp/s:50 mem_gb:39.62
96
+ Training: 27%|██▋ | 16000/60000 [5:59:04<14:59:09, 1.23s/step]INFO 2026-07-14 22:11:55 ot_train.py:625 step:16K smpl:1M ep:23K epch:4.69 loss:0.061 grdn:0.272 lr:8.4e-05 updt_s:0.674 data_s:0.592 smp/s:51 mem_gb:39.64
97
+ Training: 27%|██▋ | 16200/60000 [6:03:18<15:56:38, 1.31s/step]INFO 2026-07-14 22:16:09 ot_train.py:625 step:16K smpl:1M ep:23K epch:4.75 loss:0.060 grdn:0.258 lr:8.4e-05 updt_s:0.668 data_s:0.598 smp/s:51 mem_gb:39.64
98
+ Training: 27%|██▋ | 16400/60000 [6:07:27<15:43:33, 1.30s/step]INFO 2026-07-14 22:20:19 ot_train.py:625 step:16K smpl:1M ep:24K epch:4.81 loss:0.062 grdn:0.274 lr:8.4e-05 updt_s:0.660 data_s:0.585 smp/s:51 mem_gb:39.64
99
+ Training: 28%|██▊ | 16600/60000 [6:11:43<15:50:06, 1.31s/step]INFO 2026-07-14 22:24:35 ot_train.py:625 step:17K smpl:1M ep:24K epch:4.87 loss:0.060 grdn:0.270 lr:8.3e-05 updt_s:0.677 data_s:0.599 smp/s:50 mem_gb:39.63
100
+ Training: 28%|██▊ | 16800/60000 [6:16:05<15:21:36, 1.28s/step]INFO 2026-07-14 22:28:56 ot_train.py:625 step:17K smpl:1M ep:24K epch:4.92 loss:0.061 grdn:0.254 lr:8.3e-05 updt_s:0.685 data_s:0.618 smp/s:49 mem_gb:39.64
101
+ Training: 28%|██▊ | 17000/60000 [6:20:21<15:33:00, 1.30s/step]INFO 2026-07-14 22:33:12 ot_train.py:625 step:17K smpl:1M ep:25K epch:4.98 loss:0.061 grdn:0.262 lr:8.2e-05 updt_s:0.687 data_s:0.589 smp/s:50 mem_gb:39.64
102
+ Training: 29%|██▊ | 17200/60000 [6:24:31<14:26:42, 1.22s/step]INFO 2026-07-14 22:37:22 ot_train.py:625 step:17K smpl:1M ep:25K epch:5.04 loss:0.063 grdn:0.303 lr:8.2e-05 updt_s:0.676 data_s:0.573 smp/s:51 mem_gb:39.64
103
+ Training: 29%|██▉ | 17400/60000 [6:28:40<14:54:34, 1.26s/step]INFO 2026-07-14 22:41:32 ot_train.py:625 step:17K smpl:1M ep:25K epch:5.10 loss:0.060 grdn:0.271 lr:8.2e-05 updt_s:0.673 data_s:0.571 smp/s:51 mem_gb:39.63
104
+ Training: 29%|██▉ | 17600/60000 [6:32:50<15:00:30, 1.27s/step]INFO 2026-07-14 22:45:41 ot_train.py:625 step:18K smpl:1M ep:25K epch:5.16 loss:0.059 grdn:0.268 lr:8.1e-05 updt_s:0.676 data_s:0.568 smp/s:51 mem_gb:39.64
105
+ Training: 30%|██▉ | 17800/60000 [6:37:01<14:37:09, 1.25s/step]INFO 2026-07-14 22:49:52 ot_train.py:625 step:18K smpl:1M ep:26K epch:5.22 loss:0.060 grdn:0.260 lr:8.1e-05 updt_s:0.670 data_s:0.581 smp/s:51 mem_gb:39.64
106
+ Training: 30%|███ | 18000/60000 [6:41:10<15:23:22, 1.32s/step]INFO 2026-07-14 22:54:01 ot_train.py:625 step:18K smpl:1M ep:26K epch:5.28 loss:0.060 grdn:0.281 lr:8.0e-05 updt_s:0.674 data_s:0.571 smp/s:51 mem_gb:39.64
107
+ Training: 30%|███ | 18200/60000 [6:45:20<14:02:08, 1.21s/step]INFO 2026-07-14 22:58:11 ot_train.py:625 step:18K smpl:1M ep:26K epch:5.33 loss:0.060 grdn:0.265 lr:8.0e-05 updt_s:0.669 data_s:0.577 smp/s:51 mem_gb:39.62
108
+ Training: 31%|███ | 18400/60000 [6:49:43<14:36:21, 1.26s/step]INFO 2026-07-14 23:02:34 ot_train.py:625 step:18K smpl:1M ep:27K epch:5.39 loss:0.057 grdn:0.259 lr:7.9e-05 updt_s:0.701 data_s:0.611 smp/s:49 mem_gb:39.64
109
+ Training: 31%|███ | 18600/60000 [6:54:06<14:11:09, 1.23s/step]INFO 2026-07-14 23:06:57 ot_train.py:625 step:19K smpl:1M ep:27K epch:5.45 loss:0.059 grdn:0.263 lr:7.9e-05 updt_s:0.699 data_s:0.612 smp/s:49 mem_gb:39.63
110
+ Training: 31%|███▏ | 18800/60000 [6:58:35<15:03:04, 1.32s/step]INFO 2026-07-14 23:11:27 ot_train.py:625 step:19K smpl:1M ep:27K epch:5.51 loss:0.058 grdn:0.273 lr:7.9e-05 updt_s:0.712 data_s:0.632 smp/s:48 mem_gb:39.64
111
+ Training: 32%|███▏ | 19000/60000 [7:03:01<14:58:10, 1.31s/step]INFO 2026-07-14 23:15:52 ot_train.py:625 step:19K smpl:1M ep:27K epch:5.57 loss:0.059 grdn:0.265 lr:7.8e-05 updt_s:0.694 data_s:0.631 smp/s:48 mem_gb:39.64
112
+ Training: 32%|███▏ | 19200/60000 [7:07:23<13:38:45, 1.20s/step]INFO 2026-07-14 23:20:15 ot_train.py:625 step:19K smpl:1M ep:28K epch:5.63 loss:0.059 grdn:0.262 lr:7.8e-05 updt_s:0.688 data_s:0.620 smp/s:49 mem_gb:39.64
113
+ Training: 32%|███▏ | 19400/60000 [7:11:17<12:53:09, 1.14s/step]INFO 2026-07-14 23:24:08 ot_train.py:625 step:19K smpl:1M ep:28K epch:5.69 loss:0.058 grdn:0.265 lr:7.7e-05 updt_s:0.626 data_s:0.540 smp/s:55 mem_gb:39.63
114
+ Training: 33%|███▎ | 19600/60000 [7:15:11<14:19:26, 1.28s/step]INFO 2026-07-14 23:28:02 ot_train.py:625 step:20K smpl:1M ep:28K epch:5.74 loss:0.056 grdn:0.261 lr:7.7e-05 updt_s:0.630 data_s:0.535 smp/s:55 mem_gb:39.64
115
+ Training: 33%|███▎ | 19800/60000 [7:19:32<16:27:23, 1.47s/step]INFO 2026-07-14 23:32:23 ot_train.py:625 step:20K smpl:1M ep:29K epch:5.80 loss:0.058 grdn:0.265 lr:7.6e-05 updt_s:0.675 data_s:0.625 smp/s:49 mem_gb:39.64
116
+ Training: 33%|███▎ | 20000/60000 [7:24:10<14:40:40, 1.32s/step]INFO 2026-07-14 23:37:01 ot_train.py:625 step:20K smpl:1M ep:29K epch:5.86 loss:0.058 grdn:0.277 lr:7.6e-05 updt_s:0.723 data_s:0.666 smp/s:46 mem_gb:39.63
117
+ Training: 34%|███▎ | 20200/60000 [7:29:08<17:58:49, 1.63s/step]INFO 2026-07-14 23:41:59 ot_train.py:625 step:20K smpl:1M ep:29K epch:5.92 loss:0.060 grdn:0.285 lr:7.6e-05 updt_s:0.756 data_s:0.732 smp/s:43 mem_gb:39.64
118
+ Training: 34%|███▍ | 20400/60000 [7:34:25<16:39:59, 1.52s/step]INFO 2026-07-14 23:47:16 ot_train.py:625 step:20K smpl:1M ep:29K epch:5.98 loss:0.056 grdn:0.268 lr:7.5e-05 updt_s:0.799 data_s:0.781 smp/s:41 mem_gb:39.62
119
+ Training: 34%|███▍ | 20600/60000 [7:39:43<15:34:50, 1.42s/step]INFO 2026-07-14 23:52:34 ot_train.py:625 step:21K smpl:1M ep:30K epch:6.04 loss:0.055 grdn:0.262 lr:7.5e-05 updt_s:0.797 data_s:0.791 smp/s:40 mem_gb:39.64
120
+ Training: 35%|███▍ | 20800/60000 [7:45:03<18:00:15, 1.65s/step]INFO 2026-07-14 23:57:54 ot_train.py:625 step:21K smpl:1M ep:30K epch:6.10 loss:0.054 grdn:0.258 lr:7.4e-05 updt_s:0.795 data_s:0.801 smp/s:40 mem_gb:39.64
121
+ Training: 35%|███▌ | 21000/60000 [7:50:23<16:51:45, 1.56s/step]INFO 2026-07-15 00:03:14 ot_train.py:625 step:21K smpl:1M ep:30K epch:6.15 loss:0.056 grdn:0.258 lr:7.4e-05 updt_s:0.790 data_s:0.806 smp/s:40 mem_gb:39.64
122
+ Training: 35%|███▌ | 21200/60000 [7:55:39<17:37:43, 1.64s/step]INFO 2026-07-15 00:08:30 ot_train.py:625 step:21K smpl:1M ep:31K epch:6.21 loss:0.054 grdn:0.258 lr:7.3e-05 updt_s:0.785 data_s:0.793 smp/s:41 mem_gb:39.64
123
+ Training: 36%|███▌ | 21400/60000 [8:00:54<17:25:58, 1.63s/step]INFO 2026-07-15 00:13:46 ot_train.py:625 step:21K smpl:1M ep:31K epch:6.27 loss:0.054 grdn:0.258 lr:7.3e-05 updt_s:0.791 data_s:0.784 smp/s:41 mem_gb:39.64
124
+ Training: 36%|███▌ | 21600/60000 [8:06:01<17:31:34, 1.64s/step]INFO 2026-07-15 00:18:52 ot_train.py:625 step:22K smpl:1M ep:31K epch:6.33 loss:0.055 grdn:0.262 lr:7.2e-05 updt_s:0.773 data_s:0.755 smp/s:42 mem_gb:39.63
125
+ Training: 36%|███▋ | 21800/60000 [8:11:19<16:43:14, 1.58s/step]INFO 2026-07-15 00:24:10 ot_train.py:625 step:22K smpl:1M ep:31K epch:6.39 loss:0.054 grdn:0.263 lr:7.2e-05 updt_s:0.794 data_s:0.793 smp/s:40 mem_gb:39.64
126
+ Training: 37%|███▋ | 22000/60000 [8:16:41<17:11:39, 1.63s/step]INFO 2026-07-15 00:29:32 ot_train.py:625 step:22K smpl:1M ep:32K epch:6.45 loss:0.054 grdn:0.269 lr:7.1e-05 updt_s:0.800 data_s:0.806 smp/s:40 mem_gb:39.64
127
+ Training: 37%|███▋ | 22200/60000 [8:21:59<17:04:40, 1.63s/step]INFO 2026-07-15 00:34:51 ot_train.py:625 step:22K smpl:1M ep:32K epch:6.51 loss:0.054 grdn:0.260 lr:7.1e-05 updt_s:0.791 data_s:0.799 smp/s:40 mem_gb:39.64
128
+ Training: 37%|███▋ | 22400/60000 [8:27:18<16:23:29, 1.57s/step]INFO 2026-07-15 00:40:09 ot_train.py:625 step:22K smpl:1M ep:32K epch:6.57 loss:0.057 grdn:0.284 lr:7.0e-05 updt_s:0.791 data_s:0.799 smp/s:40 mem_gb:39.64
129
+ Training: 38%|███▊ | 22600/60000 [8:32:28<15:18:06, 1.47s/step]INFO 2026-07-15 00:45:19 ot_train.py:625 step:23K smpl:1M ep:33K epch:6.62 loss:0.053 grdn:0.251 lr:7.0e-05 updt_s:0.774 data_s:0.770 smp/s:41 mem_gb:39.62
130
+ Training: 38%|███▊ | 22800/60000 [8:36:49<11:38:23, 1.13s/step]INFO 2026-07-15 00:49:40 ot_train.py:625 step:23K smpl:1M ep:33K epch:6.68 loss:0.053 grdn:0.265 lr:6.9e-05 updt_s:0.685 data_s:0.619 smp/s:49 mem_gb:39.64
131
+ Training: 38%|███▊ | 23000/60000 [8:41:05<14:19:13, 1.39s/step]INFO 2026-07-15 00:53:56 ot_train.py:625 step:23K smpl:1M ep:33K epch:6.74 loss:0.051 grdn:0.249 lr:6.9e-05 updt_s:0.729 data_s:0.549 smp/s:50 mem_gb:39.64
132
+ Training: 39%|███▊ | 23200/60000 [8:45:41<13:59:08, 1.37s/step]INFO 2026-07-15 00:58:32 ot_train.py:625 step:23K smpl:1M ep:34K epch:6.80 loss:0.052 grdn:0.273 lr:6.8e-05 updt_s:0.806 data_s:0.572 smp/s:46 mem_gb:39.64
133
+ Training: 39%|███▉ | 23400/60000 [8:50:16<13:59:29, 1.38s/step]INFO 2026-07-15 01:03:07 ot_train.py:625 step:23K smpl:1M ep:34K epch:6.86 loss:0.051 grdn:0.266 lr:6.8e-05 updt_s:0.805 data_s:0.569 smp/s:47 mem_gb:39.64
134
+ Training: 39%|███▉ | 23600/60000 [8:54:51<13:59:56, 1.38s/step]INFO 2026-07-15 01:07:42 ot_train.py:625 step:24K smpl:2M ep:34K epch:6.92 loss:0.053 grdn:0.266 lr:6.7e-05 updt_s:0.806 data_s:0.564 smp/s:47 mem_gb:39.64
135
+ Training: 40%|███▉ | 23800/60000 [8:59:25<13:26:36, 1.34s/step]INFO 2026-07-15 01:12:16 ot_train.py:625 step:24K smpl:2M ep:34K epch:6.98 loss:0.051 grdn:0.250 lr:6.7e-05 updt_s:0.801 data_s:0.569 smp/s:47 mem_gb:39.64
136
+ Training: 40%|████ | 24000/60000 [9:03:54<13:29:52, 1.35s/step]INFO 2026-07-15 01:16:45 ot_train.py:625 step:24K smpl:2M ep:35K epch:7.03 loss:0.050 grdn:0.265 lr:6.6e-05 updt_s:0.765 data_s:0.576 smp/s:48 mem_gb:39.64
137
+ Training: 40%|████ | 24200/60000 [9:08:27<13:56:27, 1.40s/step]INFO 2026-07-15 01:21:18 ot_train.py:625 step:24K smpl:2M ep:35K epch:7.09 loss:0.052 grdn:0.261 lr:6.6e-05 updt_s:0.799 data_s:0.562 smp/s:47 mem_gb:39.63
138
+ Training: 41%|████ | 24400/60000 [9:13:08<13:10:45, 1.33s/step]INFO 2026-07-15 01:25:59 ot_train.py:625 step:24K smpl:2M ep:35K epch:7.15 loss:0.051 grdn:0.262 lr:6.5e-05 updt_s:0.834 data_s:0.569 smp/s:46 mem_gb:39.63
139
+ Training: 41%|████ | 24600/60000 [9:17:45<15:14:13, 1.55s/step]INFO 2026-07-15 01:30:36 ot_train.py:625 step:25K smpl:2M ep:36K epch:7.21 loss:0.052 grdn:0.259 lr:6.5e-05 updt_s:0.825 data_s:0.557 smp/s:46 mem_gb:39.64
140
+ Training: 41%|████▏ | 24800/60000 [9:22:23<13:03:16, 1.34s/step]INFO 2026-07-15 01:35:14 ot_train.py:625 step:25K smpl:2M ep:36K epch:7.27 loss:0.050 grdn:0.252 lr:6.4e-05 updt_s:0.817 data_s:0.569 smp/s:46 mem_gb:39.64
141
+ Training: 42%|████▏ | 25000/60000 [9:26:59<13:23:54, 1.38s/step]INFO 2026-07-15 01:39:50 ot_train.py:625 step:25K smpl:2M ep:36K epch:7.33 loss:0.051 grdn:0.264 lr:6.4e-05 updt_s:0.813 data_s:0.568 smp/s:46 mem_gb:39.62
142
+ Training: 42%|████▏ | 25200/60000 [9:31:36<12:46:56, 1.32s/step]INFO 2026-07-15 01:44:27 ot_train.py:625 step:25K smpl:2M ep:36K epch:7.39 loss:0.051 grdn:0.309 lr:6.3e-05 updt_s:0.821 data_s:0.559 smp/s:46 mem_gb:39.64
143
+ Training: 42%|████▏ | 25400/60000 [9:36:17<13:28:54, 1.40s/step]INFO 2026-07-15 01:49:08 ot_train.py:625 step:25K smpl:2M ep:37K epch:7.44 loss:0.048 grdn:0.251 lr:6.3e-05 updt_s:0.832 data_s:0.572 smp/s:46 mem_gb:39.64
144
+ Training: 43%|████▎ | 25600/60000 [9:40:57<12:45:29, 1.34s/step]INFO 2026-07-15 01:53:49 ot_train.py:625 step:26K smpl:2M ep:37K epch:7.50 loss:0.050 grdn:0.266 lr:6.2e-05 updt_s:0.843 data_s:0.555 smp/s:46 mem_gb:39.64
145
+ Training: 43%|████▎ | 25800/60000 [9:45:50<14:57:05, 1.57s/step]INFO 2026-07-15 01:58:42 ot_train.py:625 step:26K smpl:2M ep:37K epch:7.56 loss:0.049 grdn:0.269 lr:6.2e-05 updt_s:0.856 data_s:0.607 smp/s:44 mem_gb:39.64
146
+ Training: 43%|████▎ | 26000/60000 [9:51:14<14:39:08, 1.55s/step]INFO 2026-07-15 02:04:05 ot_train.py:625 step:26K smpl:2M ep:38K epch:7.62 loss:0.049 grdn:0.272 lr:6.1e-05 updt_s:0.906 data_s:0.710 smp/s:40 mem_gb:39.64
147
+ Training: 44%|████▎ | 26200/60000 [9:56:34<15:30:28, 1.65s/step]INFO 2026-07-15 02:09:26 ot_train.py:625 step:26K smpl:2M ep:38K epch:7.68 loss:0.049 grdn:0.263 lr:6.1e-05 updt_s:0.885 data_s:0.714 smp/s:40 mem_gb:39.63
148
+ Training: 44%|████▍ | 26400/60000 [10:02:02<14:44:53, 1.58s/step]INFO 2026-07-15 02:14:53 ot_train.py:625 step:26K smpl:2M ep:38K epch:7.74 loss:0.049 grdn:0.266 lr:6.0e-05 updt_s:0.889 data_s:0.746 smp/s:39 mem_gb:39.63
149
+ Training: 44%|████▍ | 26600/60000 [10:07:37<19:16:12, 2.08s/step]INFO 2026-07-15 02:20:28 ot_train.py:625 step:27K smpl:2M ep:38K epch:7.80 loss:0.049 grdn:0.274 lr:6.0e-05 updt_s:0.911 data_s:0.759 smp/s:38 mem_gb:39.64
150
+ Training: 45%|████▍ | 26800/60000 [10:13:10<15:20:56, 1.66s/step]INFO 2026-07-15 02:26:01 ot_train.py:625 step:27K smpl:2M ep:39K epch:7.85 loss:0.050 grdn:0.270 lr:5.9e-05 updt_s:0.884 data_s:0.779 smp/s:38 mem_gb:39.63
151
+ Training: 45%|████▌ | 27000/60000 [10:18:41<15:53:56, 1.73s/step]INFO 2026-07-15 02:31:32 ot_train.py:625 step:27K smpl:2M ep:39K epch:7.91 loss:0.047 grdn:0.259 lr:5.9e-05 updt_s:0.902 data_s:0.752 smp/s:39 mem_gb:39.63
152
+ Training: 45%|████▌ | 27200/60000 [10:24:13<14:53:00, 1.63s/step]INFO 2026-07-15 02:37:04 ot_train.py:625 step:27K smpl:2M ep:39K epch:7.97 loss:0.047 grdn:0.260 lr:5.8e-05 updt_s:0.876 data_s:0.780 smp/s:39 mem_gb:39.62
153
+ Training: 46%|████▌ | 27400/60000 [10:29:44<15:48:54, 1.75s/step]INFO 2026-07-15 02:42:36 ot_train.py:625 step:27K smpl:2M ep:40K epch:8.03 loss:0.047 grdn:0.273 lr:5.8e-05 updt_s:0.888 data_s:0.766 smp/s:39 mem_gb:39.63
154
+ Training: 46%|████▌ | 27600/60000 [10:35:17<15:05:31, 1.68s/step]INFO 2026-07-15 02:48:08 ot_train.py:625 step:28K smpl:2M ep:40K epch:8.09 loss:0.046 grdn:0.258 lr:5.7e-05 updt_s:0.918 data_s:0.740 smp/s:39 mem_gb:39.64
155
+ Training: 46%|████▋ | 27800/60000 [10:40:48<14:24:38, 1.61s/step]INFO 2026-07-15 02:53:39 ot_train.py:625 step:28K smpl:2M ep:40K epch:8.15 loss:0.046 grdn:0.258 lr:5.7e-05 updt_s:0.885 data_s:0.768 smp/s:39 mem_gb:39.64
156
+ Training: 47%|████▋ | 28000/60000 [10:46:18<14:09:46, 1.59s/step]INFO 2026-07-15 02:59:09 ot_train.py:625 step:28K smpl:2M ep:40K epch:8.21 loss:0.045 grdn:0.259 lr:5.6e-05 updt_s:0.906 data_s:0.742 smp/s:39 mem_gb:39.64
157
+ Training: 47%|████▋ | 28200/60000 [10:51:43<15:20:49, 1.74s/step]INFO 2026-07-15 03:04:34 ot_train.py:625 step:28K smpl:2M ep:41K epch:8.26 loss:0.046 grdn:0.271 lr:5.6e-05 updt_s:0.925 data_s:0.696 smp/s:39 mem_gb:39.64
158
+ Training: 47%|████▋ | 28400/60000 [10:57:14<15:10:37, 1.73s/step]INFO 2026-07-15 03:10:05 ot_train.py:625 step:28K smpl:2M ep:41K epch:8.32 loss:0.046 grdn:0.280 lr:5.5e-05 updt_s:0.877 data_s:0.776 smp/s:39 mem_gb:39.64
159
+ Training: 48%|████▊ | 28600/60000 [11:02:46<14:21:58, 1.65s/step]INFO 2026-07-15 03:15:37 ot_train.py:625 step:29K smpl:2M ep:41K epch:8.38 loss:0.046 grdn:0.291 lr:5.5e-05 updt_s:0.909 data_s:0.749 smp/s:39 mem_gb:39.64
160
+ Training: 48%|████▊ | 28800/60000 [11:08:18<14:46:01, 1.70s/step]INFO 2026-07-15 03:21:09 ot_train.py:625 step:29K smpl:2M ep:42K epch:8.44 loss:0.046 grdn:0.272 lr:5.4e-05 updt_s:0.902 data_s:0.752 smp/s:39 mem_gb:39.64
161
+ Training: 48%|████▊ | 29000/60000 [11:13:50<14:15:49, 1.66s/step]INFO 2026-07-15 03:26:41 ot_train.py:625 step:29K smpl:2M ep:42K epch:8.50 loss:0.046 grdn:0.273 lr:5.4e-05 updt_s:0.914 data_s:0.746 smp/s:39 mem_gb:39.64
162
+ Training: 49%|████▊ | 29200/60000 [11:19:24<14:15:17, 1.67s/step]INFO 2026-07-15 03:32:15 ot_train.py:625 step:29K smpl:2M ep:42K epch:8.56 loss:0.046 grdn:0.264 lr:5.3e-05 updt_s:0.879 data_s:0.787 smp/s:38 mem_gb:39.64
163
+ Training: 49%|████▉ | 29400/60000 [11:24:59<14:02:17, 1.65s/step]INFO 2026-07-15 03:37:50 ot_train.py:625 step:29K smpl:2M ep:42K epch:8.62 loss:0.045 grdn:0.278 lr:5.3e-05 updt_s:0.874 data_s:0.796 smp/s:38 mem_gb:39.62
164
+ Training: 49%|████▉ | 29600/60000 [11:30:29<13:54:15, 1.65s/step]INFO 2026-07-15 03:43:20 ot_train.py:625 step:30K smpl:2M ep:43K epch:8.68 loss:0.044 grdn:0.278 lr:5.2e-05 updt_s:0.867 data_s:0.779 smp/s:39 mem_gb:39.64
165
+ Training: 50%|██���█▉ | 29800/60000 [11:36:02<14:42:59, 1.75s/step]INFO 2026-07-15 03:48:53 ot_train.py:625 step:30K smpl:2M ep:43K epch:8.73 loss:0.043 grdn:0.268 lr:5.1e-05 updt_s:0.874 data_s:0.789 smp/s:38 mem_gb:39.64
166
+ Training: 50%|█████ | 30000/60000 [11:41:34<13:52:00, 1.66s/step]INFO 2026-07-15 03:54:25 ot_train.py:625 step:30K smpl:2M ep:43K epch:8.79 loss:0.043 grdn:0.278 lr:5.1e-05 updt_s:0.878 data_s:0.781 smp/s:39 mem_gb:39.64
167
+ Training: 50%|█████ | 30200/60000 [11:47:11<13:44:23, 1.66s/step]INFO 2026-07-15 04:00:02 ot_train.py:625 step:30K smpl:2M ep:44K epch:8.85 loss:0.043 grdn:0.276 lr:5.0e-05 updt_s:0.894 data_s:0.786 smp/s:38 mem_gb:39.64
168
+ Training: 51%|█████ | 30400/60000 [11:52:46<13:39:50, 1.66s/step]INFO 2026-07-15 04:05:37 ot_train.py:625 step:30K smpl:2M ep:44K epch:8.91 loss:0.043 grdn:0.264 lr:5.0e-05 updt_s:0.909 data_s:0.764 smp/s:38 mem_gb:39.64
169
+ Training: 51%|█████ | 30600/60000 [11:58:20<13:32:44, 1.66s/step]INFO 2026-07-15 04:11:11 ot_train.py:625 step:31K smpl:2M ep:44K epch:8.97 loss:0.042 grdn:0.270 lr:4.9e-05 updt_s:0.873 data_s:0.794 smp/s:38 mem_gb:39.64
170
+ Training: 51%|█████▏ | 30800/60000 [12:03:53<13:26:33, 1.66s/step]INFO 2026-07-15 04:16:45 ot_train.py:625 step:31K smpl:2M ep:45K epch:9.03 loss:0.043 grdn:0.267 lr:4.9e-05 updt_s:0.875 data_s:0.789 smp/s:38 mem_gb:39.64
171
+ Training: 52%|█████▏ | 31000/60000 [12:09:29<13:30:18, 1.68s/step]INFO 2026-07-15 04:22:20 ot_train.py:625 step:31K smpl:2M ep:45K epch:9.09 loss:0.042 grdn:0.270 lr:4.8e-05 updt_s:0.901 data_s:0.774 smp/s:38 mem_gb:39.64
172
+ Training: 52%|█████▏ | 31200/60000 [12:15:01<13:14:58, 1.66s/step]INFO 2026-07-15 04:27:52 ot_train.py:625 step:31K smpl:2M ep:45K epch:9.14 loss:0.042 grdn:0.266 lr:4.8e-05 updt_s:0.872 data_s:0.784 smp/s:39 mem_gb:39.64
173
+ Training: 52%|█████▏ | 31400/60000 [12:20:32<12:46:41, 1.61s/step]INFO 2026-07-15 04:33:23 ot_train.py:625 step:31K smpl:2M ep:45K epch:9.20 loss:0.043 grdn:0.267 lr:4.7e-05 updt_s:0.896 data_s:0.757 smp/s:39 mem_gb:39.64
174
+ Training: 53%|█████▎ | 31600/60000 [12:26:06<15:55:47, 2.02s/step]INFO 2026-07-15 04:38:57 ot_train.py:625 step:32K smpl:2M ep:46K epch:9.26 loss:0.042 grdn:0.269 lr:4.7e-05 updt_s:0.900 data_s:0.766 smp/s:38 mem_gb:39.62
175
+ Training: 53%|█████▎ | 31800/60000 [12:31:35<11:47:12, 1.50s/step]INFO 2026-07-15 04:44:26 ot_train.py:625 step:32K smpl:2M ep:46K epch:9.32 loss:0.040 grdn:0.260 lr:4.6e-05 updt_s:0.891 data_s:0.751 smp/s:39 mem_gb:39.64
176
+ Training: 53%|█████▎ | 32000/60000 [12:37:07<12:48:46, 1.65s/step]INFO 2026-07-15 04:49:58 ot_train.py:625 step:32K smpl:2M ep:46K epch:9.38 loss:0.041 grdn:0.275 lr:4.6e-05 updt_s:0.886 data_s:0.771 smp/s:39 mem_gb:39.64
177
+ Training: 54%|█████▎ | 32200/60000 [12:42:41<12:45:47, 1.65s/step]INFO 2026-07-15 04:55:32 ot_train.py:625 step:32K smpl:2M ep:47K epch:9.44 loss:0.041 grdn:0.286 lr:4.5e-05 updt_s:0.889 data_s:0.775 smp/s:38 mem_gb:39.64
178
+ Training: 54%|█████▍ | 32400/60000 [12:48:13<12:43:18, 1.66s/step]INFO 2026-07-15 05:01:04 ot_train.py:625 step:32K smpl:2M ep:47K epch:9.50 loss:0.041 grdn:0.277 lr:4.5e-05 updt_s:0.890 data_s:0.770 smp/s:39 mem_gb:39.63
179
+ Training: 54%|█████▍ | 32600/60000 [12:53:49<12:41:27, 1.67s/step]INFO 2026-07-15 05:06:40 ot_train.py:625 step:33K smpl:2M ep:47K epch:9.55 loss:0.040 grdn:0.277 lr:4.4e-05 updt_s:0.901 data_s:0.774 smp/s:38 mem_gb:39.64
180
+ Training: 55%|█████▍ | 32800/60000 [12:59:16<12:20:59, 1.63s/step]INFO 2026-07-15 05:12:07 ot_train.py:625 step:33K smpl:2M ep:47K epch:9.61 loss:0.040 grdn:0.271 lr:4.4e-05 updt_s:0.922 data_s:0.713 smp/s:39 mem_gb:39.63
181
+ Training: 55%|█████▌ | 33000/60000 [13:04:43<12:14:14, 1.63s/step]INFO 2026-07-15 05:17:34 ot_train.py:625 step:33K smpl:2M ep:48K epch:9.67 loss:0.041 grdn:0.286 lr:4.3e-05 updt_s:0.874 data_s:0.758 smp/s:39 mem_gb:39.63
182
+ Training: 55%|█████▌ | 33200/60000 [13:10:12<12:14:11, 1.64s/step]INFO 2026-07-15 05:23:03 ot_train.py:625 step:33K smpl:2M ep:48K epch:9.73 loss:0.040 grdn:0.286 lr:4.3e-05 updt_s:0.899 data_s:0.742 smp/s:39 mem_gb:39.64
183
+ Training: 56%|█████▌ | 33400/60000 [13:15:41<12:09:56, 1.65s/step]INFO 2026-07-15 05:28:32 ot_train.py:625 step:33K smpl:2M ep:48K epch:9.79 loss:0.040 grdn:0.279 lr:4.2e-05 updt_s:0.895 data_s:0.749 smp/s:39 mem_gb:39.64
184
+ Training: 56%|█████▌ | 33600/60000 [13:21:14<12:40:48, 1.73s/step]INFO 2026-07-15 05:34:05 ot_train.py:625 step:34K smpl:2M ep:49K epch:9.85 loss:0.040 grdn:0.287 lr:4.1e-05 updt_s:0.905 data_s:0.754 smp/s:39 mem_gb:39.64
185
+ Training: 56%|█████▋ | 33800/60000 [13:26:39<11:36:03, 1.59s/step]INFO 2026-07-15 05:39:30 ot_train.py:625 step:34K smpl:2M ep:49K epch:9.91 loss:0.040 grdn:0.281 lr:4.1e-05 updt_s:0.930 data_s:0.691 smp/s:39 mem_gb:39.64
186
+ Training: 57%|█████▋ | 34000/60000 [13:32:14<11:58:23, 1.66s/step]INFO 2026-07-15 05:45:05 ot_train.py:625 step:34K smpl:2M ep:49K epch:9.96 loss:0.038 grdn:0.264 lr:4.0e-05 updt_s:0.932 data_s:0.742 smp/s:38 mem_gb:39.62
187
+ Training: 57%|█████▋ | 34200/60000 [13:37:48<12:17:56, 1.72s/step]INFO 2026-07-15 05:50:39 ot_train.py:625 step:34K smpl:2M ep:49K epch:10.02 loss:0.037 grdn:0.260 lr:4.0e-05 updt_s:0.891 data_s:0.773 smp/s:38 mem_gb:39.64
188
+ Training: 57%|█████▋ | 34400/60000 [13:43:24<12:10:03, 1.71s/step]INFO 2026-07-15 05:56:15 ot_train.py:625 step:34K smpl:2M ep:50K epch:10.08 loss:0.037 grdn:0.274 lr:3.9e-05 updt_s:0.893 data_s:0.786 smp/s:38 mem_gb:39.63
189
+ Training: 58%|█████▊ | 34600/60000 [13:48:56<11:45:11, 1.67s/step]INFO 2026-07-15 06:01:47 ot_train.py:625 step:35K smpl:2M ep:50K epch:10.14 loss:0.037 grdn:0.273 lr:3.9e-05 updt_s:0.899 data_s:0.756 smp/s:39 mem_gb:39.63
190
+ Training: 58%|█████▊ | 34800/60000 [13:54:27<11:31:52, 1.65s/step]INFO 2026-07-15 06:07:18 ot_train.py:625 step:35K smpl:2M ep:50K epch:10.20 loss:0.038 grdn:0.269 lr:3.8e-05 updt_s:0.874 data_s:0.780 smp/s:39 mem_gb:39.64
191
+ Training: 58%|█████▊ | 35000/60000 [13:59:57<11:15:26, 1.62s/step]INFO 2026-07-15 06:12:48 ot_train.py:625 step:35K smpl:2M ep:51K epch:10.26 loss:0.037 grdn:0.277 lr:3.8e-05 updt_s:0.881 data_s:0.763 smp/s:39 mem_gb:39.63
192
+ Training: 59%|█████▊ | 35200/60000 [14:05:28<11:24:17, 1.66s/step]INFO 2026-07-15 06:18:19 ot_train.py:625 step:35K smpl:2M ep:51K epch:10.32 loss:0.037 grdn:0.300 lr:3.7e-05 updt_s:0.896 data_s:0.759 smp/s:39 mem_gb:39.64
193
+ Training: 59%|█████▉ | 35400/60000 [14:11:00<11:26:13, 1.67s/step]INFO 2026-07-15 06:23:51 ot_train.py:625 step:35K smpl:2M ep:51K epch:10.38 loss:0.037 grdn:0.274 lr:3.7e-05 updt_s:0.886 data_s:0.770 smp/s:39 mem_gb:39.64
194
+ Training: 59%|█████▉ | 35600/60000 [14:16:34<11:16:52, 1.66s/step]INFO 2026-07-15 06:29:25 ot_train.py:625 step:36K smpl:2M ep:51K epch:10.43 loss:0.036 grdn:0.276 lr:3.6e-05 updt_s:0.905 data_s:0.763 smp/s:38 mem_gb:39.63
195
+ Training: 60%|█████▉ | 35800/60000 [14:22:04<11:07:28, 1.65s/step]INFO 2026-07-15 06:34:56 ot_train.py:625 step:36K smpl:2M ep:52K epch:10.49 loss:0.036 grdn:0.273 lr:3.6e-05 updt_s:0.896 data_s:0.751 smp/s:39 mem_gb:39.64
196
+ Training: 60%|██████ | 36000/60000 [14:27:41<12:01:42, 1.80s/step]INFO 2026-07-15 06:40:32 ot_train.py:625 step:36K smpl:2M ep:52K epch:10.55 loss:0.037 grdn:0.277 lr:3.5e-05 updt_s:0.890 data_s:0.791 smp/s:38 mem_gb:39.64
197
+ Training: 60%|██████ | 36200/60000 [14:33:22<11:10:40, 1.69s/step]INFO 2026-07-15 06:46:13 ot_train.py:625 step:36K smpl:2M ep:52K epch:10.61 loss:0.036 grdn:0.278 lr:3.5e-05 updt_s:0.918 data_s:0.784 smp/s:38 mem_gb:39.62
198
+ Training: 61%|██████ | 36400/60000 [14:38:57<10:49:25, 1.65s/step]INFO 2026-07-15 06:51:48 ot_train.py:625 step:36K smpl:2M ep:53K epch:10.67 loss:0.035 grdn:0.279 lr:3.4e-05 updt_s:0.911 data_s:0.759 smp/s:38 mem_gb:39.64
199
+ Training: 61%|██████ | 36600/60000 [14:44:28<10:57:38, 1.69s/step]INFO 2026-07-15 06:57:19 ot_train.py:625 step:37K smpl:2M ep:53K epch:10.73 loss:0.034 grdn:0.267 lr:3.4e-05 updt_s:0.908 data_s:0.745 smp/s:39 mem_gb:39.63
200
+ Training: 61%|██████▏ | 36800/60000 [14:50:01<10:47:55, 1.68s/step]INFO 2026-07-15 07:02:52 ot_train.py:625 step:37K smpl:2M ep:53K epch:10.79 loss:0.034 grdn:0.272 lr:3.3e-05 updt_s:0.887 data_s:0.776 smp/s:38 mem_gb:39.63
201
+ Training: 62%|██████▏ | 37000/60000 [14:55:40<10:28:31, 1.64s/step]INFO 2026-07-15 07:08:31 ot_train.py:625 step:37K smpl:2M ep:53K epch:10.84 loss:0.034 grdn:0.274 lr:3.3e-05 updt_s:0.899 data_s:0.791 smp/s:38 mem_gb:39.64
202
+ Training: 62%|██████▏ | 37200/60000 [15:01:12<10:25:14, 1.65s/step]INFO 2026-07-15 07:14:03 ot_train.py:625 step:37K smpl:2M ep:54K epch:10.90 loss:0.034 grdn:0.287 lr:3.2e-05 updt_s:0.881 data_s:0.778 smp/s:39 mem_gb:39.64
203
+ Training: 62%|██████▏ | 37400/60000 [15:06:48<10:37:13, 1.69s/step]INFO 2026-07-15 07:19:39 ot_train.py:625 step:37K smpl:2M ep:54K epch:10.96 loss:0.035 grdn:0.311 lr:3.2e-05 updt_s:0.895 data_s:0.779 smp/s:38 mem_gb:39.63
204
+ Training: 63%|██████▎ | 37600/60000 [15:12:20<10:12:56, 1.64s/step]INFO 2026-07-15 07:25:11 ot_train.py:625 step:38K smpl:2M ep:54K epch:11.02 loss:0.034 grdn:0.274 lr:3.1e-05 updt_s:0.865 data_s:0.795 smp/s:39 mem_gb:39.64
205
+ Training: 63%|██████▎ | 37800/60000 [15:17:54<10:06:55, 1.64s/step]INFO 2026-07-15 07:30:45 ot_train.py:625 step:38K smpl:2M ep:55K epch:11.08 loss:0.034 grdn:0.285 lr:3.1e-05 updt_s:0.894 data_s:0.770 smp/s:38 mem_gb:39.64
206
+ Training: 63%|██████▎ | 38000/60000 [15:23:32<10:56:34, 1.79s/step]INFO 2026-07-15 07:36:23 ot_train.py:625 step:38K smpl:2M ep:55K epch:11.14 loss:0.034 grdn:0.276 lr:3.0e-05 updt_s:0.907 data_s:0.781 smp/s:38 mem_gb:39.63
207
+ Training: 64%|██████▎ | 38200/60000 [15:29:03<10:07:18, 1.67s/step]INFO 2026-07-15 07:41:54 ot_train.py:625 step:38K smpl:2M ep:55K epch:11.20 loss:0.034 grdn:0.279 lr:3.0e-05 updt_s:0.898 data_s:0.757 smp/s:39 mem_gb:39.64
208
+ Training: 64%|██████▍ | 38400/60000 [15:34:41<9:42:31, 1.62s/step]INFO 2026-07-15 07:47:32 ot_train.py:625 step:38K smpl:2M ep:55K epch:11.25 loss:0.034 grdn:0.278 lr:2.9e-05 updt_s:0.893 data_s:0.794 smp/s:38 mem_gb:39.62
209
+ Training: 64%|██████▍ | 38600/60000 [15:40:13<9:58:51, 1.68s/step]INFO 2026-07-15 07:53:04 ot_train.py:625 step:39K smpl:2M ep:56K epch:11.31 loss:0.031 grdn:0.273 lr:2.9e-05 updt_s:0.878 data_s:0.776 smp/s:39 mem_gb:39.64
210
+ Training: 65%|██████▍ | 38800/60000 [15:45:47<9:58:54, 1.70s/step]INFO 2026-07-15 07:58:38 ot_train.py:625 step:39K smpl:2M ep:56K epch:11.37 loss:0.032 grdn:0.269 lr:2.8e-05 updt_s:0.872 data_s:0.794 smp/s:38 mem_gb:39.64
211
+ Training: 65%|██████▌ | 39000/60000 [15:51:21<9:14:08, 1.58s/step]INFO 2026-07-15 08:04:12 ot_train.py:625 step:39K smpl:2M ep:56K epch:11.43 loss:0.032 grdn:0.268 lr:2.8e-05 updt_s:0.876 data_s:0.793 smp/s:38 mem_gb:39.64
212
+ Training: 65%|██████▌ | 39200/60000 [15:56:53<10:02:35, 1.74s/step]INFO 2026-07-15 08:09:44 ot_train.py:625 step:39K smpl:3M ep:57K epch:11.49 loss:0.030 grdn:0.278 lr:2.7e-05 updt_s:0.888 data_s:0.768 smp/s:39 mem_gb:39.64
213
+ Training: 66%|██████▌ | 39400/60000 [16:02:25<9:52:44, 1.73s/step]INFO 2026-07-15 08:15:16 ot_train.py:625 step:39K smpl:3M ep:57K epch:11.55 loss:0.032 grdn:0.276 lr:2.7e-05 updt_s:0.897 data_s:0.761 smp/s:39 mem_gb:39.63
214
+ Training: 66%|██████▌ | 39600/60000 [16:08:00<9:22:03, 1.65s/step]INFO 2026-07-15 08:20:51 ot_train.py:625 step:40K smpl:3M ep:57K epch:11.61 loss:0.032 grdn:0.283 lr:2.7e-05 updt_s:0.891 data_s:0.778 smp/s:38 mem_gb:39.64
215
+ Training: 66%|██████▋ | 39800/60000 [16:13:33<9:25:39, 1.68s/step]INFO 2026-07-15 08:26:24 ot_train.py:625 step:40K smpl:3M ep:58K epch:11.66 loss:0.032 grdn:0.284 lr:2.6e-05 updt_s:0.881 data_s:0.783 smp/s:38 mem_gb:39.64
216
+ Training: 67%|██████▋ | 40000/60000 [16:19:06<9:08:31, 1.65s/step]INFO 2026-07-15 08:31:57 ot_train.py:625 step:40K smpl:3M ep:58K epch:11.72 loss:0.031 grdn:0.275 lr:2.6e-05 updt_s:0.899 data_s:0.763 smp/s:38 mem_gb:39.64
217
+ Training: 67%|██████▋ | 40200/60000 [16:24:36<9:11:55, 1.67s/step]INFO 2026-07-15 08:37:27 ot_train.py:625 step:40K smpl:3M ep:58K epch:11.78 loss:0.032 grdn:0.277 lr:2.5e-05 updt_s:0.888 data_s:0.757 smp/s:39 mem_gb:39.64
218
+ Training: 67%|██████▋ | 40400/60000 [16:30:08<9:18:53, 1.71s/step]INFO 2026-07-15 08:42:59 ot_train.py:625 step:40K smpl:3M ep:58K epch:11.84 loss:0.030 grdn:0.271 lr:2.5e-05 updt_s:0.887 data_s:0.771 smp/s:39 mem_gb:39.64
219
+ Training: 68%|██████▊ | 40600/60000 [16:35:41<9:07:31, 1.69s/step]INFO 2026-07-15 08:48:32 ot_train.py:625 step:41K smpl:3M ep:59K epch:11.90 loss:0.030 grdn:0.292 lr:2.4e-05 updt_s:0.883 data_s:0.777 smp/s:39 mem_gb:39.64
220
+ Training: 68%|██████▊ | 40800/60000 [16:41:13<8:45:21, 1.64s/step]INFO 2026-07-15 08:54:04 ot_train.py:625 step:41K smpl:3M ep:59K epch:11.96 loss:0.029 grdn:0.272 lr:2.4e-05 updt_s:0.898 data_s:0.761 smp/s:39 mem_gb:39.62
221
+ Training: 68%|██████▊ | 41000/60000 [16:46:49<8:40:00, 1.64s/step]INFO 2026-07-15 08:59:40 ot_train.py:625 step:41K smpl:3M ep:59K epch:12.02 loss:0.030 grdn:0.284 lr:2.3e-05 updt_s:0.895 data_s:0.783 smp/s:38 mem_gb:39.64
222
+ Training: 69%|██████▊ | 41200/60000 [16:52:23<8:35:24, 1.64s/step]INFO 2026-07-15 09:05:14 ot_train.py:625 step:41K smpl:3M ep:60K epch:12.08 loss:0.029 grdn:0.289 lr:2.3e-05 updt_s:0.893 data_s:0.770 smp/s:38 mem_gb:39.63
223
+ Training: 69%|██████▉ | 41400/60000 [16:57:54<8:40:07, 1.68s/step]INFO 2026-07-15 09:10:45 ot_train.py:625 step:41K smpl:3M ep:60K epch:12.13 loss:0.029 grdn:0.262 lr:2.2e-05 updt_s:0.888 data_s:0.767 smp/s:39 mem_gb:39.64
224
+ Training: 69%|██████▉ | 41600/60000 [17:03:26<8:20:39, 1.63s/step]INFO 2026-07-15 09:16:17 ot_train.py:625 step:42K smpl:3M ep:60K epch:12.19 loss:0.027 grdn:0.266 lr:2.2e-05 updt_s:0.867 data_s:0.789 smp/s:39 mem_gb:39.64
225
+ Training: 70%|██████▉ | 41800/60000 [17:10:48<20:41:28, 4.09s/step]INFO 2026-07-15 09:23:39 ot_train.py:625 step:42K smpl:3M ep:60K epch:12.25 loss:0.029 grdn:0.273 lr:2.2e-05 updt_s:1.412 data_s:0.794 smp/s:29 mem_gb:39.64
226
+ Training: 70%|███████ | 42000/60000 [17:24:24<20:31:05, 4.10s/step]INFO 2026-07-15 09:37:15 ot_train.py:625 step:42K smpl:3M ep:61K epch:12.31 loss:0.029 grdn:0.269 lr:2.1e-05 updt_s:3.228 data_s:0.847 smp/s:16 mem_gb:39.64
227
+ Training: 70%|███████ | 42200/60000 [17:35:54<12:07:31, 2.45s/step]INFO 2026-07-15 09:48:45 ot_train.py:625 step:42K smpl:3M ep:61K epch:12.37 loss:0.029 grdn:0.271 lr:2.1e-05 updt_s:2.618 data_s:0.831 smp/s:19 mem_gb:39.64
228
+ Training: 71%|███████ | 42400/60000 [17:44:04<11:56:06, 2.44s/step]INFO 2026-07-15 09:56:56 ot_train.py:625 step:42K smpl:3M ep:61K epch:12.43 loss:0.029 grdn:0.279 lr:2.0e-05 updt_s:1.822 data_s:0.627 smp/s:26 mem_gb:39.64
229
+ Training: 71%|███████ | 42600/60000 [17:55:05<18:12:31, 3.77s/step]INFO 2026-07-15 10:07:56 ot_train.py:625 step:43K smpl:3M ep:62K epch:12.49 loss:0.029 grdn:0.269 lr:2.0e-05 updt_s:2.721 data_s:0.578 smp/s:19 mem_gb:39.64
230
+ Training: 71%|███████▏ | 42800/60000 [18:07:29<17:56:14, 3.75s/step]INFO 2026-07-15 10:20:20 ot_train.py:625 step:43K smpl:3M ep:62K epch:12.54 loss:0.029 grdn:0.267 lr:1.9e-05 updt_s:3.151 data_s:0.568 smp/s:17 mem_gb:39.64
231
+ Training: 72%|███████▏ | 43000/60000 [18:16:24<12:50:33, 2.72s/step]INFO 2026-07-15 10:29:15 ot_train.py:625 step:43K smpl:3M ep:62K epch:12.60 loss:0.027 grdn:0.270 lr:1.9e-05 updt_s:2.092 data_s:0.579 smp/s:24 mem_gb:39.62
232
+ Training: 72%|███████▏ | 43200/60000 [18:24:18<11:55:31, 2.56s/step]INFO 2026-07-15 10:37:10 ot_train.py:625 step:43K smpl:3M ep:62K epch:12.66 loss:0.027 grdn:0.273 lr:1.9e-05 updt_s:1.796 data_s:0.575 smp/s:27 mem_gb:39.64
233
+ Training: 72%|███████▏ | 43400/60000 [18:32:45<12:08:05, 2.63s/step]INFO 2026-07-15 10:45:36 ot_train.py:625 step:43K smpl:3M ep:63K epch:12.72 loss:0.028 grdn:0.274 lr:1.8e-05 updt_s:1.957 data_s:0.572 smp/s:25 mem_gb:39.63
234
+ Training: 73%|███████▎ | 43600/60000 [18:41:16<11:03:50, 2.43s/step]INFO 2026-07-15 10:54:07 ot_train.py:625 step:44K smpl:3M ep:63K epch:12.78 loss:0.026 grdn:0.259 lr:1.8e-05 updt_s:1.980 data_s:0.574 smp/s:25 mem_gb:39.64
235
+ Training: 73%|███████▎ | 43800/60000 [18:47:18<8:26:27, 1.88s/step]INFO 2026-07-15 11:00:09 ot_train.py:625 step:44K smpl:3M ep:63K epch:12.84 loss:0.027 grdn:0.273 lr:1.7e-05 updt_s:1.213 data_s:0.592 smp/s:35 mem_gb:39.64
236
+ Training: 73%|███████▎ | 44000/60000 [18:54:49<12:06:04, 2.72s/step]INFO 2026-07-15 11:07:40 ot_train.py:625 step:44K smpl:3M ep:64K epch:12.90 loss:0.027 grdn:0.257 lr:1.7e-05 updt_s:1.629 data_s:0.624 smp/s:28 mem_gb:39.64
237
+ Training: 74%|███████▎ | 44200/60000 [19:03:46<11:36:50, 2.65s/step]INFO 2026-07-15 11:16:37 ot_train.py:625 step:44K smpl:3M ep:64K epch:12.95 loss:0.027 grdn:0.286 lr:1.7e-05 updt_s:2.081 data_s:0.601 smp/s:24 mem_gb:39.64
238
+ Training: 74%|███████▍ | 44400/60000 [19:11:34<11:23:20, 2.63s/step]INFO 2026-07-15 11:24:25 ot_train.py:625 step:44K smpl:3M ep:64K epch:13.01 loss:0.027 grdn:0.281 lr:1.6e-05 updt_s:1.766 data_s:0.572 smp/s:27 mem_gb:39.64
239
+ Training: 74%|███████▍ | 44600/60000 [19:20:07<11:26:35, 2.68s/step]INFO 2026-07-15 11:32:58 ot_train.py:625 step:45K smpl:3M ep:64K epch:13.07 loss:0.026 grdn:0.264 lr:1.6e-05 updt_s:1.982 data_s:0.579 smp/s:25 mem_gb:39.64
240
+ Training: 75%|███████▍ | 44800/60000 [19:27:32<8:00:15, 1.90s/step]INFO 2026-07-15 11:40:23 ot_train.py:625 step:45K smpl:3M ep:65K epch:13.13 loss:0.026 grdn:0.263 lr:1.5e-05 updt_s:1.652 data_s:0.572 smp/s:29 mem_gb:39.64
241
+ Training: 75%|███████▌ | 45000/60000 [19:33:50<5:02:16, 1.21s/step]INFO 2026-07-15 11:46:41 ot_train.py:625 step:45K smpl:3M ep:65K epch:13.19 loss:0.026 grdn:0.264 lr:1.5e-05 updt_s:1.243 data_s:0.641 smp/s:34 mem_gb:39.63
242
+ Training: 75%|███████▌ | 45200/60000 [19:42:45<10:54:27, 2.65s/step]INFO 2026-07-15 11:55:36 ot_train.py:625 step:45K smpl:3M ep:65K epch:13.25 loss:0.026 grdn:0.262 lr:1.5e-05 updt_s:2.077 data_s:0.596 smp/s:24 mem_gb:39.62
243
+ Training: 76%|███████▌ | 45400/60000 [19:51:39<11:06:40, 2.74s/step]INFO 2026-07-15 12:04:30 ot_train.py:625 step:45K smpl:3M ep:66K epch:13.31 loss:0.025 grdn:0.251 lr:1.4e-05 updt_s:2.088 data_s:0.582 smp/s:24 mem_gb:39.64
244
+ Training: 76%|███████▌ | 45600/60000 [19:59:38<10:22:59, 2.60s/step]INFO 2026-07-15 12:12:29 ot_train.py:625 step:46K smpl:3M ep:66K epch:13.36 loss:0.026 grdn:0.295 lr:1.4e-05 updt_s:1.812 data_s:0.578 smp/s:27 mem_gb:39.64
245
+ Training: 76%|███████▋ | 45800/60000 [20:08:05<5:11:53, 1.32s/step]INFO 2026-07-15 12:20:56 ot_train.py:625 step:46K smpl:3M ep:66K epch:13.42 loss:0.025 grdn:0.257 lr:1.4e-05 updt_s:1.964 data_s:0.570 smp/s:25 mem_gb:39.64
246
+ Training: 77%|███████▋ | 46000/60000 [20:14:12<7:11:12, 1.85s/step]INFO 2026-07-15 12:27:03 ot_train.py:625 step:46K smpl:3M ep:66K epch:13.48 loss:0.024 grdn:0.257 lr:1.3e-05 updt_s:1.194 data_s:0.637 smp/s:35 mem_gb:39.63
247
+ Training: 77%|███████▋ | 46200/60000 [20:22:15<10:13:05, 2.67s/step]INFO 2026-07-15 12:35:06 ot_train.py:625 step:46K smpl:3M ep:67K epch:13.54 loss:0.024 grdn:0.258 lr:1.3e-05 updt_s:1.828 data_s:0.582 smp/s:27 mem_gb:39.64
248
+ Training: 77%|███████▋ | 46400/60000 [20:30:14<9:40:57, 2.56s/step]INFO 2026-07-15 12:43:05 ot_train.py:625 step:46K smpl:3M ep:67K epch:13.60 loss:0.024 grdn:0.266 lr:1.3e-05 updt_s:1.821 data_s:0.571 smp/s:27 mem_gb:39.63
249
+ Training: 78%|███████▊ | 46600/60000 [20:37:49<7:04:35, 1.90s/step]INFO 2026-07-15 12:50:40 ot_train.py:625 step:47K smpl:3M ep:67K epch:13.66 loss:0.024 grdn:0.260 lr:1.2e-05 updt_s:1.693 data_s:0.578 smp/s:28 mem_gb:39.64
250
+ Training: 78%|███████▊ | 46800/60000 [20:44:09<7:38:08, 2.08s/step]INFO 2026-07-15 12:57:00 ot_train.py:625 step:47K smpl:3M ep:68K epch:13.72 loss:0.024 grdn:0.256 lr:1.2e-05 updt_s:1.261 data_s:0.638 smp/s:34 mem_gb:39.64
251
+ Training: 78%|███████▊ | 47000/60000 [20:52:10<9:41:08, 2.68s/step]INFO 2026-07-15 13:05:01 ot_train.py:625 step:47K smpl:3M ep:68K epch:13.77 loss:0.024 grdn:0.265 lr:1.1e-05 updt_s:1.825 data_s:0.575 smp/s:27 mem_gb:39.63
252
+ Training: 79%|███████▊ | 47200/60000 [21:00:56<9:30:27, 2.67s/step]INFO 2026-07-15 13:13:47 ot_train.py:625 step:47K smpl:3M ep:68K epch:13.83 loss:0.024 grdn:0.264 lr:1.1e-05 updt_s:2.050 data_s:0.577 smp/s:24 mem_gb:39.63
253
+ Training: 79%|███████▉ | 47400/60000 [21:09:19<4:47:27, 1.37s/step]INFO 2026-07-15 13:22:10 ot_train.py:625 step:47K smpl:3M ep:68K epch:13.89 loss:0.024 grdn:0.276 lr:1.1e-05 updt_s:1.935 data_s:0.580 smp/s:25 mem_gb:39.62
254
+ Training: 79%|███████▉ | 47600/60000 [21:17:27<8:38:29, 2.51s/step]INFO 2026-07-15 13:30:19 ot_train.py:625 step:48K smpl:3M ep:69K epch:13.95 loss:0.023 grdn:0.254 lr:1.0e-05 updt_s:1.870 data_s:0.569 smp/s:26 mem_gb:39.64
255
+ Training: 80%|███████▉ | 47800/60000 [21:26:57<10:57:31, 3.23s/step]INFO 2026-07-15 13:39:48 ot_train.py:625 step:48K smpl:3M ep:69K epch:14.01 loss:0.023 grdn:0.274 lr:1.0e-05 updt_s:2.158 data_s:0.688 smp/s:22 mem_gb:39.63
256
+ Training: 80%|████████ | 48000/60000 [21:35:23<6:19:34, 1.90s/step]INFO 2026-07-15 13:48:14 ot_train.py:625 step:48K smpl:3M ep:69K epch:14.07 loss:0.023 grdn:0.253 lr:9.9e-06 updt_s:1.758 data_s:0.769 smp/s:25 mem_gb:39.63
257
+ Training: 80%|████████ | 48200/60000 [21:40:03<3:53:51, 1.19s/step]INFO 2026-07-15 13:52:54 ot_train.py:625 step:48K smpl:3M ep:70K epch:14.13 loss:0.023 grdn:0.259 lr:9.5e-06 updt_s:0.794 data_s:0.604 smp/s:46 mem_gb:39.64
258
+ Training: 81%|████████ | 48400/60000 [21:44:09<3:56:28, 1.22s/step]INFO 2026-07-15 13:57:00 ot_train.py:625 step:48K smpl:3M ep:70K epch:14.19 loss:0.022 grdn:0.250 lr:9.2e-06 updt_s:0.647 data_s:0.577 smp/s:52 mem_gb:39.63
259
+ Training: 81%|████████ | 48600/60000 [21:48:11<3:45:58, 1.19s/step]INFO 2026-07-15 14:01:02 ot_train.py:625 step:49K smpl:3M ep:70K epch:14.24 loss:0.023 grdn:0.260 lr:8.9e-06 updt_s:0.639 data_s:0.568 smp/s:53 mem_gb:39.64
260
+ Training: 81%|████████▏ | 48800/60000 [21:52:14<3:41:56, 1.19s/step]INFO 2026-07-15 14:05:05 ot_train.py:625 step:49K smpl:3M ep:71K epch:14.30 loss:0.024 grdn:0.253 lr:8.6e-06 updt_s:0.644 data_s:0.572 smp/s:53 mem_gb:39.63
261
+ Training: 82%|████████▏ | 49000/60000 [21:56:15<3:32:43, 1.16s/step]INFO 2026-07-15 14:09:06 ot_train.py:625 step:49K smpl:3M ep:71K epch:14.36 loss:0.023 grdn:0.245 lr:8.3e-06 updt_s:0.643 data_s:0.558 smp/s:53 mem_gb:39.63
262
+ Training: 82%|████████▏ | 49200/60000 [22:00:34<3:41:21, 1.23s/step]INFO 2026-07-15 14:13:25 ot_train.py:625 step:49K smpl:3M ep:71K epch:14.42 loss:0.022 grdn:0.235 lr:8.1e-06 updt_s:0.735 data_s:0.558 smp/s:49 mem_gb:39.63
263
+ Training: 82%|████████▏ | 49400/60000 [22:04:49<4:14:51, 1.44s/step]INFO 2026-07-15 14:17:40 ot_train.py:625 step:49K smpl:3M ep:71K epch:14.48 loss:0.023 grdn:0.253 lr:7.8e-06 updt_s:0.667 data_s:0.603 smp/s:50 mem_gb:39.64
264
+ Training: 83%|████████▎ | 49600/60000 [22:09:17<4:17:15, 1.48s/step]INFO 2026-07-15 14:22:08 ot_train.py:625 step:50K smpl:3M ep:72K epch:14.54 loss:0.022 grdn:0.249 lr:7.5e-06 updt_s:0.689 data_s:0.649 smp/s:48 mem_gb:39.64
265
+ Training: 83%|████████▎ | 49800/60000 [22:14:03<3:50:24, 1.36s/step]INFO 2026-07-15 14:26:54 ot_train.py:625 step:50K smpl:3M ep:72K epch:14.60 loss:0.022 grdn:0.252 lr:7.2e-06 updt_s:0.720 data_s:0.706 smp/s:45 mem_gb:39.62
266
+ Training: 83%|████████▎ | 50000/60000 [22:18:50<3:54:21, 1.41s/step]INFO 2026-07-15 14:31:41 ot_train.py:625 step:50K smpl:3M ep:72K epch:14.65 loss:0.022 grdn:0.246 lr:6.9e-06 updt_s:0.721 data_s:0.713 smp/s:45 mem_gb:39.64
267
+ INFO 2026-07-15 14:31:41 ot_train.py:670 Checkpoint policy after step 50000
268
+ INFO 2026-07-15 14:32:10 ain_yaml.py:662 Saved config, action-space, and prompt provenance in /home/ext_minje/groot_insight/Abs_6D/Baseline and /home/ext_minje/groot_insight/Abs_6D/Baseline/checkpoints/050000
269
+ Training: 83%|████████▎ | 50007/60000 [22:19:28<6:15:26, 2.25s/step]
Baseline/wandb/run-20260714_161240-1jmjb68j/files/requirements.txt ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ pyarrow==25.0.0
2
+ importlib_metadata==9.0.0
3
+ wheel==0.47.0
4
+ cloudpickle==3.1.2
5
+ nvidia-cusolver-cu12==11.7.3.90
6
+ multiprocess==0.70.19
7
+ torch==2.11.0+cu128
8
+ nvidia-nvjitlink-cu12==12.8.93
9
+ fsspec==2026.2.0
10
+ pytz==2026.2
11
+ hf-xet==1.5.1
12
+ tokenizers==0.22.2
13
+ opencv-python-headless==4.13.0.92
14
+ cuda-pathfinder==1.2.2
15
+ frozenlist==1.8.0
16
+ smmap==5.0.3
17
+ httpcore==1.0.9
18
+ nvidia-nvshmem-cu12==3.4.5
19
+ nvidia-cuda-nvrtc-cu12==12.8.93
20
+ annotated-types==0.7.0
21
+ platformdirs==4.10.0
22
+ click==8.4.2
23
+ pyyaml-include==1.4.1
24
+ MarkupSafe==3.0.3
25
+ Jinja2==3.1.6
26
+ urllib3==2.7.0
27
+ nvidia-curand-cu12==10.3.9.90
28
+ certifi==2026.6.17
29
+ nvidia-cuda-runtime-cu12==12.8.90
30
+ torchcodec==0.11.1
31
+ toml==0.10.2
32
+ av==15.1.0
33
+ nvidia-cufft-cu12==11.3.3.83
34
+ attrs==26.1.0
35
+ lerobot==0.6.0
36
+ PyYAML==6.0.3
37
+ safetensors==0.8.0
38
+ diffusers==0.35.2
39
+ requests==2.34.2
40
+ nvidia-nvtx-cu12==12.8.90
41
+ charset-normalizer==3.4.9
42
+ networkx==3.6.1
43
+ cmake==4.1.3
44
+ pydantic_core==2.46.4
45
+ timm==1.0.28
46
+ dm-tree==0.1.10
47
+ nvidia-cufile-cu12==1.13.1.3
48
+ termcolor==3.3.0
49
+ multidict==6.7.1
50
+ filelock==3.29.0
51
+ tzdata==2026.3
52
+ aiohappyeyeballs==2.7.1
53
+ httpx==0.28.1
54
+ nvidia-nccl-cu12==2.28.9
55
+ Pygments==2.20.0
56
+ einops==0.8.2
57
+ nvidia-cusparse-cu12==12.5.8.93
58
+ torchvision==0.26.0+cu128
59
+ dill==0.4.1
60
+ GitPython==3.1.51
61
+ nvidia-cuda-cupti-cu12==12.8.90
62
+ decord==0.6.0
63
+ sympy==1.14.0
64
+ cuda-toolkit==12.8.1
65
+ python-dateutil==2.9.0.post0
66
+ shellingham==1.5.4
67
+ propcache==0.5.2
68
+ Farama-Notifications==0.0.6
69
+ pillow==12.2.0
70
+ cuda-bindings==12.9.4
71
+ huggingface_hub==1.23.0
72
+ pandas==2.3.3
73
+ typing-inspect==0.9.0
74
+ packaging==25.0
75
+ rich==15.0.0
76
+ nvidia-cusparselt-cu12==0.7.1
77
+ setuptools==78.1.0
78
+ wrapt==2.2.2
79
+ mypy_extensions==1.1.0
80
+ tqdm==4.68.4
81
+ draccus==0.10.0
82
+ aiosignal==1.4.0
83
+ mpmath==1.3.0
84
+ numpy==2.2.6
85
+ typing_extensions==4.15.0
86
+ psutil==7.2.2
87
+ mdurl==0.1.2
88
+ nvidia-cublas-cu12==12.8.4.1
89
+ absl-py==2.5.0
90
+ gitdb==4.0.12
91
+ transformers==5.5.4
92
+ accelerate==1.14.0
93
+ pydantic==2.13.4
94
+ pip==26.1.2
95
+ markdown-it-py==4.2.0
96
+ typer==0.26.8
97
+ triton==3.6.0
98
+ wandb==0.27.2
99
+ regex==2026.7.10
100
+ yarl==1.24.2
101
+ jsonlines==4.0.0
102
+ mergedeep==1.3.4
103
+ aiohttp==3.14.1
104
+ datasets==4.8.5
105
+ gymnasium==1.3.0
106
+ peft==0.19.1
107
+ zipp==4.1.0
108
+ idna==3.18
109
+ nvidia-cudnn-cu12==9.19.0.56
110
+ h11==0.16.0
111
+ sentry-sdk==2.65.0
112
+ annotated-doc==0.0.4
113
+ protobuf==7.35.1
114
+ typing-inspection==0.4.2
115
+ anyio==4.14.2
116
+ xxhash==3.8.1
117
+ six==1.17.0
Baseline/wandb/run-20260714_161240-1jmjb68j/files/wandb-metadata.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "os": "Linux-6.8.0-107-generic-x86_64-with-glibc2.39",
3
+ "python": "CPython 3.12.13",
4
+ "startedAt": "2026-07-14T07:12:40.764055Z",
5
+ "args": [
6
+ "--config_path=/tmp/insight_lerobot_cfg_gtruilkj/baseline.yaml"
7
+ ],
8
+ "program": "/home/ext_minje/clvla/benchmarks/INSIGHT/my_scripts/lerobot_train_yaml.py",
9
+ "codePath": "benchmarks/INSIGHT/my_scripts/lerobot_train_yaml.py",
10
+ "codePathLocal": "my_scripts/lerobot_train_yaml.py",
11
+ "git": {
12
+ "remote": "git@github.com:paragon7060/clvla.git",
13
+ "commit": "ba1e47317c7f1aa2008714489345f8ee1bc52332"
14
+ },
15
+ "root": "/home/ext_minje/groot_insight/Abs_6D/Baseline",
16
+ "host": "DGX-H200-01",
17
+ "executable": "/home/ext_minje/miniforge3/envs/lerobot060_groot/bin/python",
18
+ "cpu_count": 112,
19
+ "cpu_count_logical": 224,
20
+ "gpu": "NVIDIA H200",
21
+ "gpu_count": 8,
22
+ "disk": {
23
+ "/": {
24
+ "total": "1888425144320",
25
+ "used": "1234837356544"
26
+ }
27
+ },
28
+ "memory": {
29
+ "total": "2164177248256"
30
+ },
31
+ "gpu_nvidia": [
32
+ {
33
+ "name": "NVIDIA H200",
34
+ "memoryTotal": "150754820096",
35
+ "cudaCores": 16896,
36
+ "architecture": "Hopper",
37
+ "uuid": "GPU-1e531a45-480e-3c1a-e9a0-2ba2cc27b0a3"
38
+ },
39
+ {
40
+ "name": "NVIDIA H200",
41
+ "memoryTotal": "150754820096",
42
+ "cudaCores": 16896,
43
+ "architecture": "Hopper",
44
+ "uuid": "GPU-5d3d5ae2-ffde-89d9-8818-dc19e11ad2fb"
45
+ },
46
+ {
47
+ "name": "NVIDIA H200",
48
+ "memoryTotal": "150754820096",
49
+ "cudaCores": 16896,
50
+ "architecture": "Hopper",
51
+ "uuid": "GPU-305c0dae-dcf5-d7bb-7dd3-5477d2194275"
52
+ },
53
+ {
54
+ "name": "NVIDIA H200",
55
+ "memoryTotal": "150754820096",
56
+ "cudaCores": 16896,
57
+ "architecture": "Hopper",
58
+ "uuid": "GPU-ed8b7920-8881-4e8f-3e88-be1bbecafa21"
59
+ },
60
+ {
61
+ "name": "NVIDIA H200",
62
+ "memoryTotal": "150754820096",
63
+ "cudaCores": 16896,
64
+ "architecture": "Hopper",
65
+ "uuid": "GPU-a91176db-55f2-b284-bbbf-56626596947d"
66
+ },
67
+ {
68
+ "name": "NVIDIA H200",
69
+ "memoryTotal": "150754820096",
70
+ "cudaCores": 16896,
71
+ "architecture": "Hopper",
72
+ "uuid": "GPU-922ff6d5-a948-205a-0426-dbee495fc3dd"
73
+ },
74
+ {
75
+ "name": "NVIDIA H200",
76
+ "memoryTotal": "150754820096",
77
+ "cudaCores": 16896,
78
+ "architecture": "Hopper",
79
+ "uuid": "GPU-1fde7c75-18fe-e3db-a4c1-85e67522f55f"
80
+ },
81
+ {
82
+ "name": "NVIDIA H200",
83
+ "memoryTotal": "150754820096",
84
+ "cudaCores": 16896,
85
+ "architecture": "Hopper",
86
+ "uuid": "GPU-a9b1c49b-d527-97d8-ce91-7e50bd02154c"
87
+ }
88
+ ],
89
+ "cudaVersion": "12.9",
90
+ "writerId": "njwh3jqwvkleypwcv23lu7svai2ouvzj"
91
+ }
Baseline/wandb/run-20260714_161240-1jmjb68j/logs/debug-core.log ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {"time":"2026-07-14T16:12:40.8218567+09:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp144ap2gv/port-201098.txt","pid":201098,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
2
+ {"time":"2026-07-14T16:12:40.822647836+09:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":201098}
3
+ {"time":"2026-07-14T16:12:40.822552298+09:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-201098-201620-3722653362/socket","Net":"unix"}}
4
+ {"time":"2026-07-14T16:12:40.992669022+09:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
5
+ {"time":"2026-07-14T16:12:41.001152537+09:00","level":"INFO","msg":"handleInformInit: received","streamId":"1jmjb68j","id":"1(@)"}
6
+ {"time":"2026-07-14T16:12:41.620084809+09:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"1jmjb68j","id":"1(@)"}
7
+ {"time":"2026-07-14T16:12:47.144909049+09:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"n7zozk137zt4"}
8
+ {"time":"2026-07-15T14:32:24.049418672+09:00","level":"INFO","msg":"server: parent process exited, terminating service process"}
Baseline/wandb/run-20260714_161240-1jmjb68j/logs/debug-internal.log ADDED
The diff for this file is too large to render. See raw diff
 
Baseline/wandb/run-20260714_161240-1jmjb68j/logs/debug.log ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2026-07-14 16:12:40,765 INFO MainThread:201098 [wandb_setup.py:_flush():81] Current SDK version is 0.27.2
2
+ 2026-07-14 16:12:40,765 INFO MainThread:201098 [wandb_setup.py:_flush():81] Configure stats pid to 201098
3
+ 2026-07-14 16:12:40,765 INFO MainThread:201098 [wandb_setup.py:_flush():81] Loading settings from environment variables
4
+ 2026-07-14 16:12:40,765 INFO MainThread:201098 [wandb_init.py:setup_run_log_directory():723] Logging user logs to /home/ext_minje/groot_insight/Abs_6D/Baseline/wandb/run-20260714_161240-1jmjb68j/logs/debug.log
5
+ 2026-07-14 16:12:40,765 INFO MainThread:201098 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to /home/ext_minje/groot_insight/Abs_6D/Baseline/wandb/run-20260714_161240-1jmjb68j/logs/debug-internal.log
6
+ 2026-07-14 16:12:40,765 INFO MainThread:201098 [wandb_init.py:init():766] calling init triggers
7
+ 2026-07-14 16:12:40,765 INFO MainThread:201098 [wandb_init.py:init():771] wandb.init called with sweep_config: {}
8
+ config: {'dataset': {'repo_id': 'Whalswp/INSIGHTfixposV4_filtered_multispace_v2', 'root': '/home/ext_minje/INSIGHTfixposV4_filtered_multispace_v2', 'episodes': None, 'image_transforms': {'enable': False, 'max_num_transforms': 3, 'random_order': False, 'tfs': {'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}, 'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'affine': {'weight': 1.0, 'type': 'RandomAffine', 'kwargs': {'degrees': [-5.0, 5.0], 'translate': [0.05, 0.05]}}}}, 'revision': None, 'use_imagenet_stats': True, 'video_backend': 'torchcodec', 'return_uint8': False, 'depth_output_unit': 'mm', 'streaming': False, 'eval_split': 0.0}, 'env': None, 'policy': {'type': 'groot', 'n_obs_steps': 1, 'input_features': {}, 'output_features': {}, 'device': 'cuda', 'use_amp': False, 'use_peft': False, 'push_to_hub': False, 'repo_id': None, 'private': None, 'tags': None, 'license': None, 'pretrained_path': None, 'pretrained_revision': None, 'chunk_size': 16, 'n_action_steps': 16, 'max_state_dim': 132, 'max_action_dim': 132, 'normalization_mapping': {'VISUAL': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'STATE': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'ACTION': <NormalizationMode.IDENTITY: 'IDENTITY'>}, 'base_model_path': 'nvidia/GR00T-N1.7-3B', 'action_decode_transform': None, 'embodiment_tag': 'new_embodiment', 'tune_llm': False, 'tune_visual': False, 'tune_projector': True, 'tune_diffusion_model': True, 'tune_vlln': True, 'tune_top_llm_layers': 0, 'num_inference_timesteps': None, 'rtc_ramp_rate': None, 'use_flash_attention': False, 'use_relative_actions': False, 'relative_exclude_joints': [], 'optimizer_lr': 0.0001, 'optimizer_betas': [0.9, 0.999], 'optimizer_eps': 1e-08, 'optimizer_weight_decay': 1e-05, 'warmup_ratio': 0.05, 'use_bf16': True, 'model_params_fp32': True, 'image_size': [256, 256], 'tokenizer_assets_repo': None, 'lora_rank': 0, 'lora_alpha': 16, 'lora_dropout': 0.1, 'lora_full_model': False, 'video_backend': 'decord', 'balance_dataset_weights': True, 'balance_trajectory_weights': True, 'dataset_paths': None, 'output_dir': './tmp/gr00t', 'save_steps': 1000, 'max_steps': 10000, 'batch_size': 32, 'dataloader_num_workers': 8, 'report_to': 'wandb', 'resume': False}, 'reward_model': None, 'output_dir': '/home/ext_minje/groot_insight/Abs_6D/Baseline', 'job_name': 'INSIGHT_6D_baseline', 'resume': False, 'seed': 42, 'cudnn_deterministic': False, 'num_workers': 4, 'batch_size': 64, 'prefetch_factor': 4, 'persistent_workers': True, 'steps': 60000, 'env_eval_freq': 20000, 'log_freq': 200, 'eval_steps': 0, 'max_eval_samples': 0, 'tolerance_s': 0.0001, 'save_checkpoint': True, 'save_freq': 20000, 'use_policy_training_preset': True, 'optimizer': {'type': 'adamw', 'lr': 0.0001, 'weight_decay': 1e-05, 'grad_clip_norm': 1.0, 'betas': [0.9, 0.999], 'eps': 1e-08}, 'scheduler': {'type': 'diffuser', 'num_warmup_steps': 500, 'name': 'cosine'}, 'eval': {'n_episodes': 50, 'batch_size': 50, 'use_async_envs': True, 'recording': False, 'recording_repo_id': None, 'recording_private': False}, 'wandb': {'enable': True, 'disable_artifact': True, 'project': 'lerobot', 'entity': None, 'notes': None, 'run_id': None, 'mode': None, 'add_tags': True}, 'peft': None, 'job': {'target': None, 'image': 'huggingface/lerobot-gpu:latest', 'timeout': '2d', 'detach': False, 'tags': []}, 'save_checkpoint_to_hub': False, 'sample_weighting': None, 'rename_map': {}, 'checkpoint_path': None, '_wandb': {}}
9
+ 2026-07-14 16:12:40,766 INFO MainThread:201098 [wandb_init.py:init():814] starting backend
10
+ 2026-07-14 16:12:40,992 INFO MainThread:201098 [wandb_init.py:init():829] sending inform_init request
11
+ 2026-07-14 16:12:41,621 INFO MainThread:201098 [wandb_init.py:init():834] backend started and connected
12
+ 2026-07-14 16:12:41,623 INFO MainThread:201098 [wandb_init.py:init():904] updated telemetry
13
+ 2026-07-14 16:12:41,626 INFO MainThread:201098 [wandb_init.py:init():927] communicating run to backend with 90.0 second timeout
14
+ 2026-07-14 16:12:42,075 INFO MainThread:201098 [wandb_init.py:init():972] starting run threads in backend
15
+ 2026-07-14 16:12:42,140 INFO MainThread:201098 [wandb_run.py:_console_start():2523] atexit reg
16
+ 2026-07-14 16:12:42,140 INFO MainThread:201098 [wandb_run.py:_redirect():2373] redirect: wrap_raw
17
+ 2026-07-14 16:12:42,140 INFO MainThread:201098 [wandb_run.py:_redirect():2442] Wrapping output streams.
18
+ 2026-07-14 16:12:42,141 INFO MainThread:201098 [wandb_run.py:_redirect():2465] Redirects installed.
19
+ 2026-07-14 16:12:42,144 INFO MainThread:201098 [wandb_init.py:init():1010] run started, returning control to user process
Baseline/wandb/run-20260714_161240-1jmjb68j/run-1jmjb68j.wandb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e451bed48e6d60249ec545de6e38bcaa9380330c18bd94b6ff0af55fb101116e
3
+ size 44302336
Baseline/wandb/run-20260715_143250-1jmjb68j/files/config.yaml ADDED
@@ -0,0 +1,351 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ _wandb:
2
+ value:
3
+ cli_version: 0.27.2
4
+ e:
5
+ c8d5ybgsruh4m5h3ie5jd18btesa3xht:
6
+ args:
7
+ - --config_path=/home/ext_minje/groot_insight/Abs_6D/Baseline/checkpoints/050000/pretrained_model/train_config.json
8
+ - --resume=true
9
+ codePath: benchmarks/INSIGHT/my_scripts/lerobot_train_yaml.py
10
+ codePathLocal: my_scripts/lerobot_train_yaml.py
11
+ cpu_count: 112
12
+ cpu_count_logical: 224
13
+ cudaVersion: "12.9"
14
+ disk:
15
+ /:
16
+ total: "1888425144320"
17
+ used: "1297062400000"
18
+ executable: /home/ext_minje/miniforge3/envs/lerobot060_groot/bin/python
19
+ git:
20
+ commit: 22b2faefeaab966ca2dc3688fe6a77f3bcd8de32
21
+ remote: git@github.com:paragon7060/clvla.git
22
+ gpu: NVIDIA H200
23
+ gpu_count: 8
24
+ gpu_nvidia:
25
+ - architecture: Hopper
26
+ cudaCores: 16896
27
+ memoryTotal: "150754820096"
28
+ name: NVIDIA H200
29
+ uuid: GPU-1e531a45-480e-3c1a-e9a0-2ba2cc27b0a3
30
+ - architecture: Hopper
31
+ cudaCores: 16896
32
+ memoryTotal: "150754820096"
33
+ name: NVIDIA H200
34
+ uuid: GPU-5d3d5ae2-ffde-89d9-8818-dc19e11ad2fb
35
+ - architecture: Hopper
36
+ cudaCores: 16896
37
+ memoryTotal: "150754820096"
38
+ name: NVIDIA H200
39
+ uuid: GPU-305c0dae-dcf5-d7bb-7dd3-5477d2194275
40
+ - architecture: Hopper
41
+ cudaCores: 16896
42
+ memoryTotal: "150754820096"
43
+ name: NVIDIA H200
44
+ uuid: GPU-ed8b7920-8881-4e8f-3e88-be1bbecafa21
45
+ - architecture: Hopper
46
+ cudaCores: 16896
47
+ memoryTotal: "150754820096"
48
+ name: NVIDIA H200
49
+ uuid: GPU-a91176db-55f2-b284-bbbf-56626596947d
50
+ - architecture: Hopper
51
+ cudaCores: 16896
52
+ memoryTotal: "150754820096"
53
+ name: NVIDIA H200
54
+ uuid: GPU-922ff6d5-a948-205a-0426-dbee495fc3dd
55
+ - architecture: Hopper
56
+ cudaCores: 16896
57
+ memoryTotal: "150754820096"
58
+ name: NVIDIA H200
59
+ uuid: GPU-1fde7c75-18fe-e3db-a4c1-85e67522f55f
60
+ - architecture: Hopper
61
+ cudaCores: 16896
62
+ memoryTotal: "150754820096"
63
+ name: NVIDIA H200
64
+ uuid: GPU-a9b1c49b-d527-97d8-ce91-7e50bd02154c
65
+ host: DGX-H200-01
66
+ memory:
67
+ total: "2164177248256"
68
+ os: Linux-6.8.0-107-generic-x86_64-with-glibc2.39
69
+ program: /home/ext_minje/clvla/benchmarks/INSIGHT/my_scripts/lerobot_train_yaml.py
70
+ python: CPython 3.12.13
71
+ root: /home/ext_minje/groot_insight/Abs_6D/Baseline
72
+ startedAt: "2026-07-15T05:32:50.279073Z"
73
+ writerId: c8d5ybgsruh4m5h3ie5jd18btesa3xht
74
+ m: []
75
+ python_version: 3.12.13
76
+ t:
77
+ "1":
78
+ - 1
79
+ - 11
80
+ - 41
81
+ - 49
82
+ - 51
83
+ - 71
84
+ - 83
85
+ - 98
86
+ "2":
87
+ - 1
88
+ - 11
89
+ - 41
90
+ - 49
91
+ - 51
92
+ - 71
93
+ - 83
94
+ - 98
95
+ "3":
96
+ - 5
97
+ - 13
98
+ - 14
99
+ - 15
100
+ - 16
101
+ - 61
102
+ - 62
103
+ "4": 3.12.13
104
+ "5": 0.27.2
105
+ "6": 5.5.4
106
+ "10":
107
+ - 21
108
+ "12": 0.27.2
109
+ "13": linux-x86_64
110
+ batch_size:
111
+ value: 64
112
+ checkpoint_path:
113
+ value: /home/ext_minje/groot_insight/Abs_6D/Baseline/checkpoints/050000
114
+ cudnn_deterministic:
115
+ value: false
116
+ dataset:
117
+ value:
118
+ depth_output_unit: mm
119
+ episodes: null
120
+ eval_split: 0
121
+ image_transforms:
122
+ enable: false
123
+ max_num_transforms: 3
124
+ random_order: false
125
+ tfs:
126
+ affine:
127
+ kwargs:
128
+ degrees:
129
+ - -5
130
+ - 5
131
+ translate:
132
+ - 0.05
133
+ - 0.05
134
+ type: RandomAffine
135
+ weight: 1
136
+ brightness:
137
+ kwargs:
138
+ brightness:
139
+ - 0.8
140
+ - 1.2
141
+ type: ColorJitter
142
+ weight: 1
143
+ contrast:
144
+ kwargs:
145
+ contrast:
146
+ - 0.8
147
+ - 1.2
148
+ type: ColorJitter
149
+ weight: 1
150
+ hue:
151
+ kwargs:
152
+ hue:
153
+ - -0.05
154
+ - 0.05
155
+ type: ColorJitter
156
+ weight: 1
157
+ saturation:
158
+ kwargs:
159
+ saturation:
160
+ - 0.5
161
+ - 1.5
162
+ type: ColorJitter
163
+ weight: 1
164
+ sharpness:
165
+ kwargs:
166
+ sharpness:
167
+ - 0.5
168
+ - 1.5
169
+ type: SharpnessJitter
170
+ weight: 1
171
+ repo_id: Whalswp/INSIGHTfixposV4_filtered_multispace_v2
172
+ return_uint8: false
173
+ revision: null
174
+ root: /home/ext_minje/INSIGHTfixposV4_filtered_multispace_v2
175
+ streaming: false
176
+ use_imagenet_stats: true
177
+ video_backend: torchcodec
178
+ env:
179
+ value: null
180
+ env_eval_freq:
181
+ value: 20000
182
+ eval:
183
+ value:
184
+ batch_size: 50
185
+ n_episodes: 50
186
+ recording: false
187
+ recording_private: false
188
+ recording_repo_id: null
189
+ use_async_envs: true
190
+ eval_steps:
191
+ value: 0
192
+ job:
193
+ value:
194
+ detach: false
195
+ image: huggingface/lerobot-gpu:latest
196
+ tags: []
197
+ target: null
198
+ timeout: 2d
199
+ job_name:
200
+ value: INSIGHT_6D_baseline
201
+ log_freq:
202
+ value: 200
203
+ max_eval_samples:
204
+ value: 0
205
+ num_workers:
206
+ value: 4
207
+ optimizer:
208
+ value:
209
+ betas:
210
+ - 0.9
211
+ - 0.999
212
+ eps: 1e-08
213
+ grad_clip_norm: 1
214
+ lr: 0.0001
215
+ type: adamw
216
+ weight_decay: 1e-05
217
+ output_dir:
218
+ value: /home/ext_minje/groot_insight/Abs_6D/Baseline
219
+ peft:
220
+ value: null
221
+ persistent_workers:
222
+ value: true
223
+ policy:
224
+ value:
225
+ action_decode_transform: null
226
+ balance_dataset_weights: true
227
+ balance_trajectory_weights: true
228
+ base_model_path: nvidia/GR00T-N1.7-3B
229
+ batch_size: 32
230
+ chunk_size: 16
231
+ dataloader_num_workers: 8
232
+ dataset_paths: null
233
+ device: cuda
234
+ embodiment_tag: new_embodiment
235
+ image_size:
236
+ - 256
237
+ - 256
238
+ input_features:
239
+ observation.images.guide:
240
+ shape:
241
+ - 3
242
+ - 224
243
+ - 224
244
+ type: VISUAL
245
+ observation.images.right_shoulder:
246
+ shape:
247
+ - 3
248
+ - 224
249
+ - 224
250
+ type: VISUAL
251
+ observation.images.wrist:
252
+ shape:
253
+ - 3
254
+ - 224
255
+ - 224
256
+ type: VISUAL
257
+ observation.state:
258
+ shape:
259
+ - 16
260
+ type: STATE
261
+ license: null
262
+ lora_alpha: 16
263
+ lora_dropout: 0.1
264
+ lora_full_model: false
265
+ lora_rank: 0
266
+ max_action_dim: 132
267
+ max_state_dim: 132
268
+ max_steps: 10000
269
+ model_params_fp32: true
270
+ n_action_steps: 16
271
+ n_obs_steps: 1
272
+ normalization_mapping:
273
+ ACTION: IDENTITY
274
+ STATE: IDENTITY
275
+ VISUAL: IDENTITY
276
+ num_inference_timesteps: null
277
+ optimizer_betas:
278
+ - 0.9
279
+ - 0.999
280
+ optimizer_eps: 1e-08
281
+ optimizer_lr: 0.0001
282
+ optimizer_weight_decay: 1e-05
283
+ output_dir: ./tmp/gr00t
284
+ output_features:
285
+ action:
286
+ shape:
287
+ - 10
288
+ type: ACTION
289
+ pretrained_path: /home/ext_minje/groot_insight/Abs_6D/Baseline/checkpoints/050000/pretrained_model
290
+ pretrained_revision: null
291
+ private: null
292
+ push_to_hub: false
293
+ relative_exclude_joints: []
294
+ repo_id: null
295
+ report_to: wandb
296
+ resume: false
297
+ rtc_ramp_rate: null
298
+ save_steps: 1000
299
+ tags: null
300
+ tokenizer_assets_repo: null
301
+ tune_diffusion_model: true
302
+ tune_llm: false
303
+ tune_projector: true
304
+ tune_top_llm_layers: 0
305
+ tune_visual: false
306
+ tune_vlln: true
307
+ type: groot
308
+ use_amp: false
309
+ use_bf16: true
310
+ use_flash_attention: false
311
+ use_peft: false
312
+ use_relative_actions: false
313
+ video_backend: decord
314
+ warmup_ratio: 0.05
315
+ prefetch_factor:
316
+ value: 4
317
+ resume:
318
+ value: true
319
+ reward_model:
320
+ value: null
321
+ sample_weighting:
322
+ value: null
323
+ save_checkpoint:
324
+ value: true
325
+ save_checkpoint_to_hub:
326
+ value: false
327
+ save_freq:
328
+ value: 20000
329
+ scheduler:
330
+ value:
331
+ name: cosine
332
+ num_warmup_steps: 500
333
+ type: diffuser
334
+ seed:
335
+ value: 42
336
+ steps:
337
+ value: 60000
338
+ tolerance_s:
339
+ value: 0.0001
340
+ use_policy_training_preset:
341
+ value: true
342
+ wandb:
343
+ value:
344
+ add_tags: true
345
+ disable_artifact: true
346
+ enable: true
347
+ entity: null
348
+ mode: null
349
+ notes: null
350
+ project: lerobot
351
+ run_id: 1jmjb68j
Baseline/wandb/run-20260715_143250-1jmjb68j/files/output.log ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ INFO 2026-07-15 14:32:51 db_utils.py:121 Logs will be synced with wandb.
2
+ INFO 2026-07-15 14:32:51 db_utils.py:122 Track this run --> https://wandb.ai/minje227_hyu-hanyang-university/lerobot/runs/1jmjb68j
3
+ INFO 2026-07-15 14:32:51 ot_train.py:298 Creating dataset
4
+ INFO 2026-07-15 14:32:54 ot_train.py:332 Creating policy
5
+ INFO 2026-07-15 14:32:54 ng_groot.py:191 The Groot policy wraps NVIDIA's GR00T n1.7 model. Loading pretrained model from: /home/ext_minje/groot_insight/Abs_6D/Baseline/checkpoints/050000/pretrained_model
6
+ INFO 2026-07-15 14:32:54 ng_groot.py:225 Detected fine-tuned LeRobot checkpoint, loading with state dict...
7
+ Fetching 27 files: 100%|██████████| 27/27 [00:00<00:00, 1554.62it/s]
8
+ `torch_dtype` is deprecated! Use `dtype` instead!
9
+ Loading weights: 100%|██████████| 1031/1031 [00:00<00:00, 2321.50it/s]
10
+ Loading weights from local directory
11
+ INFO 2026-07-15 14:33:09 ot_train.py:405 Creating optimizer and scheduler
12
+ INFO 2026-07-15 14:33:12 ot_train.py:437 Output dir: /home/ext_minje/groot_insight/Abs_6D/Baseline
13
+ INFO 2026-07-15 14:33:12 ot_train.py:444 cfg.steps=60000 (60K)
14
+ INFO 2026-07-15 14:33:12 ot_train.py:445 dataset.num_frames=218367 (218K)
15
+ INFO 2026-07-15 14:33:12 ot_train.py:446 dataset.num_episodes=4930
16
+ INFO 2026-07-15 14:33:12 ot_train.py:449 Effective batch size: 64 x 1 = 64
17
+ INFO 2026-07-15 14:33:12 ot_train.py:450 num_learnable_params=1620515968 (2B)
18
+ INFO 2026-07-15 14:33:12 ot_train.py:451 num_total_params=3144016000 (3B)
19
+ INFO 2026-07-15 14:33:12 ot_train.py:492 Resuming data order at epoch 22, sample 22144
20
+ Training: 0%| | 0/10000 [00:00<?, ?step/s]INFO 2026-07-15 14:33:12 ot_train.py:602 Start offline training on a fixed dataset, with effective batch size: 64
21
+ Training: 2%|▏ | 200/10000 [05:24<4:14:48, 1.56s/step]INFO 2026-07-15 14:38:37 ot_train.py:646 step:50K smpl:3M ep:73K epch:14.71 loss:0.021 grdn:0.236 lr:6.7e-06 updt_s:0.924 data_s:0.694 smp/s:40 mem_gb:39.59
22
+ Training: 4%|▍ | 400/10000 [10:52<4:26:31, 1.67s/step]INFO 2026-07-15 14:44:05 ot_train.py:646 step:50K smpl:3M ep:73K epch:14.77 loss:0.021 grdn:0.247 lr:6.4e-06 updt_s:0.915 data_s:0.723 smp/s:39 mem_gb:39.59
23
+ Training: 6%|▌ | 600/10000 [16:24<4:12:28, 1.61s/step]INFO 2026-07-15 14:49:37 ot_train.py:646 step:51K smpl:3M ep:73K epch:14.83 loss:0.022 grdn:0.230 lr:6.2e-06 updt_s:0.904 data_s:0.756 smp/s:39 mem_gb:39.59
24
+ Training: 8%|▊ | 800/10000 [21:58<4:17:42, 1.68s/step]INFO 2026-07-15 14:55:10 ot_train.py:646 step:51K smpl:3M ep:73K epch:14.89 loss:0.022 grdn:0.258 lr:5.9e-06 updt_s:0.879 data_s:0.784 smp/s:38 mem_gb:39.59
25
+ Training: 10%|█ | 1000/10000 [27:30<4:11:02, 1.67s/step]INFO 2026-07-15 15:00:43 ot_train.py:646 step:51K smpl:3M ep:74K epch:14.95 loss:0.021 grdn:0.242 lr:5.7e-06 updt_s:0.879 data_s:0.783 smp/s:39 mem_gb:39.59
26
+ Training: 12%|█▏ | 1200/10000 [33:03<4:00:36, 1.64s/step]INFO 2026-07-15 15:06:16 ot_train.py:646 step:51K smpl:3M ep:74K epch:15.01 loss:0.021 grdn:0.248 lr:5.4e-06 updt_s:0.889 data_s:0.772 smp/s:39 mem_gb:39.59
27
+ Training: 14%|█▍ | 1400/10000 [38:35<4:03:11, 1.70s/step]INFO 2026-07-15 15:11:48 ot_train.py:646 step:51K smpl:3M ep:74K epch:15.06 loss:0.021 grdn:0.238 lr:5.2e-06 updt_s:0.900 data_s:0.759 smp/s:39 mem_gb:39.59
28
+ Training: 16%|█▌ | 1600/10000 [44:10<3:54:05, 1.67s/step]INFO 2026-07-15 15:17:23 ot_train.py:646 step:52K smpl:3M ep:75K epch:15.12 loss:0.021 grdn:0.243 lr:5.0e-06 updt_s:0.892 data_s:0.779 smp/s:38 mem_gb:39.59
29
+ Training: 18%|█▊ | 1800/10000 [49:42<3:48:52, 1.67s/step]INFO 2026-07-15 15:22:55 ot_train.py:646 step:52K smpl:3M ep:75K epch:15.18 loss:0.022 grdn:0.237 lr:4.7e-06 updt_s:0.887 data_s:0.771 smp/s:39 mem_gb:39.58
30
+ Training: 20%|██ | 2000/10000 [55:12<3:33:15, 1.60s/step]INFO 2026-07-15 15:28:25 ot_train.py:646 step:52K smpl:3M ep:75K epch:15.24 loss:0.020 grdn:0.244 lr:4.5e-06 updt_s:0.955 data_s:0.691 smp/s:39 mem_gb:39.57
31
+ Training: 22%|██▏ | 2200/10000 [1:00:44<3:17:56, 1.52s/step]INFO 2026-07-15 15:33:57 ot_train.py:646 step:52K smpl:3M ep:75K epch:15.30 loss:0.021 grdn:0.240 lr:4.3e-06 updt_s:0.918 data_s:0.740 smp/s:39 mem_gb:39.59
32
+ Training: 24%|██▍ | 2400/10000 [1:06:15<3:27:54, 1.64s/step]INFO 2026-07-15 15:39:27 ot_train.py:646 step:52K smpl:3M ep:76K epch:15.36 loss:0.021 grdn:0.236 lr:4.1e-06 updt_s:0.895 data_s:0.754 smp/s:39 mem_gb:39.59
33
+ Training: 26%|██▌ | 2600/10000 [1:11:46<3:26:49, 1.68s/step]INFO 2026-07-15 15:44:59 ot_train.py:646 step:53K smpl:3M ep:76K epch:15.42 loss:0.020 grdn:0.243 lr:3.9e-06 updt_s:0.905 data_s:0.750 smp/s:39 mem_gb:39.59
34
+ Training: 28%|██▊ | 2800/10000 [1:17:20<3:10:47, 1.59s/step]INFO 2026-07-15 15:50:33 ot_train.py:646 step:53K smpl:3M ep:76K epch:15.47 loss:0.021 grdn:0.231 lr:3.7e-06 updt_s:0.921 data_s:0.745 smp/s:38 mem_gb:39.59
35
+ Training: 30%|███ | 3000/10000 [1:22:42<3:09:38, 1.63s/step]INFO 2026-07-15 15:55:55 ot_train.py:646 step:53K smpl:3M ep:77K epch:15.53 loss:0.020 grdn:0.237 lr:3.5e-06 updt_s:0.994 data_s:0.614 smp/s:40 mem_gb:39.59
36
+ Training: 32%|███▏ | 3200/10000 [1:28:06<3:08:09, 1.66s/step]INFO 2026-07-15 16:01:19 ot_train.py:646 step:53K smpl:3M ep:77K epch:15.59 loss:0.021 grdn:0.246 lr:3.3e-06 updt_s:1.000 data_s:0.618 smp/s:40 mem_gb:39.59
37
+ Training: 34%|███▍ | 3400/10000 [1:33:30<3:07:51, 1.71s/step]INFO 2026-07-15 16:06:43 ot_train.py:646 step:53K smpl:3M ep:77K epch:15.65 loss:0.020 grdn:0.235 lr:3.1e-06 updt_s:1.008 data_s:0.608 smp/s:40 mem_gb:39.59
38
+ Training: 36%|███▌ | 3600/10000 [1:38:57<2:58:53, 1.68s/step]INFO 2026-07-15 16:12:10 ot_train.py:646 step:54K smpl:3M ep:77K epch:15.71 loss:0.020 grdn:0.250 lr:2.9e-06 updt_s:1.015 data_s:0.621 smp/s:39 mem_gb:39.59
39
+ Training: 38%|███▊ | 3800/10000 [1:44:17<2:53:33, 1.68s/step]INFO 2026-07-15 16:17:30 ot_train.py:646 step:54K smpl:3M ep:78K epch:15.77 loss:0.020 grdn:0.244 lr:2.7e-06 updt_s:1.002 data_s:0.594 smp/s:40 mem_gb:39.59
40
+ Training: 40%|████ | 4000/10000 [1:49:34<2:32:27, 1.52s/step]INFO 2026-07-15 16:22:47 ot_train.py:646 step:54K smpl:3M ep:78K epch:15.83 loss:0.021 grdn:0.239 lr:2.6e-06 updt_s:0.994 data_s:0.588 smp/s:40 mem_gb:39.59
41
+ Training: 42%|████▏ | 4200/10000 [1:54:56<2:34:18, 1.60s/step]INFO 2026-07-15 16:28:09 ot_train.py:646 step:54K smpl:3M ep:78K epch:15.89 loss:0.020 grdn:0.230 lr:2.4e-06 updt_s:0.985 data_s:0.623 smp/s:40 mem_gb:39.58
42
+ Training: 44%|████▍ | 4400/10000 [2:00:31<2:35:38, 1.67s/step]INFO 2026-07-15 16:33:44 ot_train.py:646 step:54K smpl:3M ep:79K epch:15.94 loss:0.020 grdn:0.227 lr:2.2e-06 updt_s:0.913 data_s:0.757 smp/s:38 mem_gb:39.59
43
+ Training: 46%|████▌ | 4600/10000 [2:06:01<2:33:49, 1.71s/step]INFO 2026-07-15 16:39:14 ot_train.py:646 step:55K smpl:3M ep:79K epch:16.00 loss:0.020 grdn:0.230 lr:2.1e-06 updt_s:0.903 data_s:0.746 smp/s:39 mem_gb:39.59
44
+ Training: 48%|████▊ | 4800/10000 [2:11:35<2:21:30, 1.63s/step]INFO 2026-07-15 16:44:48 ot_train.py:646 step:55K smpl:4M ep:79K epch:16.06 loss:0.020 grdn:0.234 lr:1.9e-06 updt_s:0.928 data_s:0.740 smp/s:38 mem_gb:39.59
45
+ Training: 50%|█████ | 5000/10000 [2:17:09<2:18:50, 1.67s/step]INFO 2026-07-15 16:50:22 ot_train.py:646 step:55K smpl:4M ep:79K epch:16.12 loss:0.019 grdn:0.238 lr:1.8e-06 updt_s:0.920 data_s:0.749 smp/s:38 mem_gb:39.59
46
+ Training: 52%|█████▏ | 5200/10000 [2:22:43<2:19:05, 1.74s/step]INFO 2026-07-15 16:55:56 ot_train.py:646 step:55K smpl:4M ep:80K epch:16.18 loss:0.020 grdn:0.234 lr:1.7e-06 updt_s:0.891 data_s:0.774 smp/s:38 mem_gb:39.59
47
+ Training: 54%|█████▍ | 5400/10000 [2:28:15<2:09:20, 1.69s/step]INFO 2026-07-15 17:01:28 ot_train.py:646 step:55K smpl:4M ep:80K epch:16.24 loss:0.020 grdn:0.237 lr:1.5e-06 updt_s:0.890 data_s:0.771 smp/s:39 mem_gb:39.59
48
+ Training: 56%|█████▌ | 5600/10000 [2:33:49<2:00:51, 1.65s/step]INFO 2026-07-15 17:07:02 ot_train.py:646 step:56K smpl:4M ep:80K epch:16.30 loss:0.020 grdn:0.228 lr:1.4e-06 updt_s:0.900 data_s:0.764 smp/s:38 mem_gb:39.59
49
+ Training: 58%|█████▊ | 5800/10000 [2:39:22<1:56:06, 1.66s/step]INFO 2026-07-15 17:12:35 ot_train.py:646 step:56K smpl:4M ep:81K epch:16.35 loss:0.020 grdn:0.230 lr:1.3e-06 updt_s:0.921 data_s:0.744 smp/s:38 mem_gb:39.59
50
+ Training: 60%|██████ | 6000/10000 [2:45:01<1:53:35, 1.70s/step]INFO 2026-07-15 17:18:14 ot_train.py:646 step:56K smpl:4M ep:81K epch:16.41 loss:0.020 grdn:0.233 lr:1.2e-06 updt_s:0.928 data_s:0.763 smp/s:38 mem_gb:39.59
51
+ Training: 62%|██████▏ | 6200/10000 [2:50:40<1:42:26, 1.62s/step]INFO 2026-07-15 17:23:53 ot_train.py:646 step:56K smpl:4M ep:81K epch:16.47 loss:0.020 grdn:0.234 lr:1.1e-06 updt_s:0.905 data_s:0.789 smp/s:38 mem_gb:39.59
52
+ Training: 64%|██████▍ | 6400/10000 [2:56:17<2:00:10, 2.00s/step]INFO 2026-07-15 17:29:30 ot_train.py:646 step:56K smpl:4M ep:81K epch:16.53 loss:0.020 grdn:0.230 lr:9.5e-07 updt_s:0.897 data_s:0.785 smp/s:38 mem_gb:39.59
53
+ Training: 66%|██████▌ | 6600/10000 [3:01:57<1:35:11, 1.68s/step]INFO 2026-07-15 17:35:10 ot_train.py:646 step:57K smpl:4M ep:82K epch:16.59 loss:0.020 grdn:0.236 lr:8.5e-07 updt_s:0.917 data_s:0.782 smp/s:38 mem_gb:39.58
54
+ Training: 68%|██████▊ | 6800/10000 [3:07:35<1:31:24, 1.71s/step]INFO 2026-07-15 17:40:48 ot_train.py:646 step:57K smpl:4M ep:82K epch:16.65 loss:0.019 grdn:0.232 lr:7.6e-07 updt_s:0.954 data_s:0.731 smp/s:38 mem_gb:39.59
55
+ Training: 70%|███████ | 7000/10000 [3:13:08<1:25:46, 1.72s/step]INFO 2026-07-15 17:46:21 ot_train.py:646 step:57K smpl:4M ep:82K epch:16.71 loss:0.019 grdn:0.221 lr:6.7e-07 updt_s:0.890 data_s:0.772 smp/s:38 mem_gb:39.59
56
+ Training: 72%|███████▏ | 7200/10000 [3:18:42<1:18:00, 1.67s/step]INFO 2026-07-15 17:51:55 ot_train.py:646 step:57K smpl:4M ep:83K epch:16.76 loss:0.020 grdn:0.244 lr:5.9e-07 updt_s:0.898 data_s:0.771 smp/s:38 mem_gb:39.59
57
+ Training: 74%|███████�� | 7400/10000 [3:24:15<1:11:50, 1.66s/step]INFO 2026-07-15 17:57:28 ot_train.py:646 step:57K smpl:4M ep:83K epch:16.82 loss:0.019 grdn:0.220 lr:5.1e-07 updt_s:0.914 data_s:0.748 smp/s:39 mem_gb:39.59
58
+ Training: 76%|███████▌ | 7600/10000 [3:29:49<1:05:33, 1.64s/step]INFO 2026-07-15 18:03:02 ot_train.py:646 step:58K smpl:4M ep:83K epch:16.88 loss:0.019 grdn:0.225 lr:4.4e-07 updt_s:0.899 data_s:0.770 smp/s:38 mem_gb:39.59
59
+ Training: 78%|███████▊ | 7800/10000 [3:35:24<1:00:16, 1.64s/step]INFO 2026-07-15 18:08:37 ot_train.py:646 step:58K smpl:4M ep:84K epch:16.94 loss:0.019 grdn:0.230 lr:3.7e-07 updt_s:0.901 data_s:0.772 smp/s:38 mem_gb:39.59
60
+ Training: 80%|████████ | 8000/10000 [3:41:01<58:47, 1.76s/step]INFO 2026-07-15 18:14:14 ot_train.py:646 step:58K smpl:4M ep:84K epch:17.00 loss:0.019 grdn:0.219 lr:3.1e-07 updt_s:0.915 data_s:0.766 smp/s:38 mem_gb:39.59
61
+ Training: 82%|████████▏ | 8200/10000 [3:46:37<50:06, 1.67s/step]INFO 2026-07-15 18:19:50 ot_train.py:646 step:58K smpl:4M ep:84K epch:17.06 loss:0.020 grdn:0.230 lr:2.5e-07 updt_s:0.910 data_s:0.768 smp/s:38 mem_gb:39.59
62
+ Training: 84%|████████▍ | 8400/10000 [3:52:19<43:53, 1.65s/step]INFO 2026-07-15 18:25:32 ot_train.py:646 step:58K smpl:4M ep:84K epch:17.12 loss:0.020 grdn:0.231 lr:2.0e-07 updt_s:0.936 data_s:0.770 smp/s:38 mem_gb:39.59
63
+ Training: 86%|████████▌ | 8600/10000 [3:57:55<42:26, 1.82s/step]INFO 2026-07-15 18:31:08 ot_train.py:646 step:59K smpl:4M ep:85K epch:17.17 loss:0.019 grdn:0.228 lr:1.6e-07 updt_s:0.953 data_s:0.725 smp/s:38 mem_gb:39.59
64
+ Training: 88%|████████▊ | 8800/10000 [4:03:31<33:19, 1.67s/step]INFO 2026-07-15 18:36:44 ot_train.py:646 step:59K smpl:4M ep:85K epch:17.23 loss:0.020 grdn:0.234 lr:1.2e-07 updt_s:0.902 data_s:0.778 smp/s:38 mem_gb:39.58
65
+ Training: 90%|█████████ | 9000/10000 [4:09:07<30:05, 1.81s/step]INFO 2026-07-15 18:42:20 ot_train.py:646 step:59K smpl:4M ep:85K epch:17.29 loss:0.019 grdn:0.220 lr:8.4e-08 updt_s:0.904 data_s:0.771 smp/s:38 mem_gb:39.59
66
+ Training: 92%|█████████▏| 9200/10000 [4:14:44<23:27, 1.76s/step]INFO 2026-07-15 18:47:57 ot_train.py:646 step:59K smpl:4M ep:86K epch:17.35 loss:0.020 grdn:0.228 lr:5.7e-08 updt_s:0.899 data_s:0.782 smp/s:38 mem_gb:39.59
67
+ Training: 94%|█████████▍| 9400/10000 [4:20:17<16:53, 1.69s/step]INFO 2026-07-15 18:53:30 ot_train.py:646 step:59K smpl:4M ep:86K epch:17.41 loss:0.020 grdn:0.243 lr:3.4e-08 updt_s:0.901 data_s:0.764 smp/s:38 mem_gb:39.59
68
+ Training: 96%|█████████▌| 9600/10000 [4:25:50<10:42, 1.61s/step]INFO 2026-07-15 18:59:03 ot_train.py:646 step:60K smpl:4M ep:86K epch:17.47 loss:0.019 grdn:0.228 lr:1.8e-08 updt_s:0.909 data_s:0.752 smp/s:39 mem_gb:39.59
69
+ Training: 98%|█████████▊| 9800/10000 [4:31:24<05:26, 1.63s/step]INFO 2026-07-15 19:04:37 ot_train.py:646 step:60K smpl:4M ep:86K epch:17.53 loss:0.019 grdn:0.225 lr:6.5e-09 updt_s:0.889 data_s:0.778 smp/s:38 mem_gb:39.59
70
+ Training: 100%|██████████| 10000/10000 [4:37:01<00:00, 1.66s/step]INFO 2026-07-15 19:10:14 ot_train.py:646 step:60K smpl:4M ep:87K epch:17.59 loss:0.019 grdn:0.219 lr:9.2e-10 updt_s:0.909 data_s:0.773 smp/s:38 mem_gb:39.59
71
+ INFO 2026-07-15 19:10:14 ot_train.py:691 Checkpoint policy after step 60000
72
+ INFO 2026-07-15 19:10:44 ain_yaml.py:918 Saved config, action-space, prompt, and phase provenance in /home/ext_minje/groot_insight/Abs_6D/Baseline and /home/ext_minje/groot_insight/Abs_6D/Baseline/checkpoints/060000
73
+ Training: 100%|██████████| 10000/10000 [4:37:31<00:00, 1.67s/step]
74
+ INFO 2026-07-15 19:10:44 ot_train.py:777 End of training
Baseline/wandb/run-20260715_143250-1jmjb68j/files/requirements.txt ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ pyarrow==25.0.0
2
+ importlib_metadata==9.0.0
3
+ wheel==0.47.0
4
+ cloudpickle==3.1.2
5
+ nvidia-cusolver-cu12==11.7.3.90
6
+ multiprocess==0.70.19
7
+ torch==2.11.0+cu128
8
+ nvidia-nvjitlink-cu12==12.8.93
9
+ fsspec==2026.2.0
10
+ pytz==2026.2
11
+ hf-xet==1.5.1
12
+ tokenizers==0.22.2
13
+ opencv-python-headless==4.13.0.92
14
+ cuda-pathfinder==1.2.2
15
+ frozenlist==1.8.0
16
+ smmap==5.0.3
17
+ httpcore==1.0.9
18
+ nvidia-nvshmem-cu12==3.4.5
19
+ nvidia-cuda-nvrtc-cu12==12.8.93
20
+ annotated-types==0.7.0
21
+ platformdirs==4.10.0
22
+ click==8.4.2
23
+ pyyaml-include==1.4.1
24
+ MarkupSafe==3.0.3
25
+ Jinja2==3.1.6
26
+ urllib3==2.7.0
27
+ nvidia-curand-cu12==10.3.9.90
28
+ certifi==2026.6.17
29
+ nvidia-cuda-runtime-cu12==12.8.90
30
+ torchcodec==0.11.1
31
+ toml==0.10.2
32
+ av==15.1.0
33
+ nvidia-cufft-cu12==11.3.3.83
34
+ attrs==26.1.0
35
+ lerobot==0.6.0
36
+ PyYAML==6.0.3
37
+ safetensors==0.8.0
38
+ diffusers==0.35.2
39
+ requests==2.34.2
40
+ nvidia-nvtx-cu12==12.8.90
41
+ charset-normalizer==3.4.9
42
+ networkx==3.6.1
43
+ cmake==4.1.3
44
+ pydantic_core==2.46.4
45
+ timm==1.0.28
46
+ dm-tree==0.1.10
47
+ nvidia-cufile-cu12==1.13.1.3
48
+ termcolor==3.3.0
49
+ multidict==6.7.1
50
+ filelock==3.29.0
51
+ tzdata==2026.3
52
+ aiohappyeyeballs==2.7.1
53
+ httpx==0.28.1
54
+ nvidia-nccl-cu12==2.28.9
55
+ Pygments==2.20.0
56
+ einops==0.8.2
57
+ nvidia-cusparse-cu12==12.5.8.93
58
+ torchvision==0.26.0+cu128
59
+ dill==0.4.1
60
+ GitPython==3.1.51
61
+ nvidia-cuda-cupti-cu12==12.8.90
62
+ decord==0.6.0
63
+ sympy==1.14.0
64
+ cuda-toolkit==12.8.1
65
+ python-dateutil==2.9.0.post0
66
+ shellingham==1.5.4
67
+ propcache==0.5.2
68
+ Farama-Notifications==0.0.6
69
+ pillow==12.2.0
70
+ cuda-bindings==12.9.4
71
+ huggingface_hub==1.23.0
72
+ pandas==2.3.3
73
+ typing-inspect==0.9.0
74
+ packaging==25.0
75
+ rich==15.0.0
76
+ nvidia-cusparselt-cu12==0.7.1
77
+ setuptools==78.1.0
78
+ wrapt==2.2.2
79
+ mypy_extensions==1.1.0
80
+ tqdm==4.68.4
81
+ draccus==0.10.0
82
+ aiosignal==1.4.0
83
+ mpmath==1.3.0
84
+ numpy==2.2.6
85
+ typing_extensions==4.15.0
86
+ psutil==7.2.2
87
+ mdurl==0.1.2
88
+ nvidia-cublas-cu12==12.8.4.1
89
+ absl-py==2.5.0
90
+ gitdb==4.0.12
91
+ transformers==5.5.4
92
+ accelerate==1.14.0
93
+ pydantic==2.13.4
94
+ pip==26.1.2
95
+ markdown-it-py==4.2.0
96
+ typer==0.26.8
97
+ triton==3.6.0
98
+ wandb==0.27.2
99
+ regex==2026.7.10
100
+ yarl==1.24.2
101
+ jsonlines==4.0.0
102
+ mergedeep==1.3.4
103
+ aiohttp==3.14.1
104
+ datasets==4.8.5
105
+ gymnasium==1.3.0
106
+ peft==0.19.1
107
+ zipp==4.1.0
108
+ idna==3.18
109
+ nvidia-cudnn-cu12==9.19.0.56
110
+ h11==0.16.0
111
+ sentry-sdk==2.65.0
112
+ annotated-doc==0.0.4
113
+ protobuf==7.35.1
114
+ typing-inspection==0.4.2
115
+ anyio==4.14.2
116
+ xxhash==3.8.1
117
+ six==1.17.0
Baseline/wandb/run-20260715_143250-1jmjb68j/files/wandb-metadata.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "os": "Linux-6.8.0-107-generic-x86_64-with-glibc2.39",
3
+ "python": "CPython 3.12.13",
4
+ "startedAt": "2026-07-15T05:32:50.279073Z",
5
+ "args": [
6
+ "--config_path=/home/ext_minje/groot_insight/Abs_6D/Baseline/checkpoints/050000/pretrained_model/train_config.json",
7
+ "--resume=true"
8
+ ],
9
+ "program": "/home/ext_minje/clvla/benchmarks/INSIGHT/my_scripts/lerobot_train_yaml.py",
10
+ "codePath": "benchmarks/INSIGHT/my_scripts/lerobot_train_yaml.py",
11
+ "codePathLocal": "my_scripts/lerobot_train_yaml.py",
12
+ "git": {
13
+ "remote": "git@github.com:paragon7060/clvla.git",
14
+ "commit": "22b2faefeaab966ca2dc3688fe6a77f3bcd8de32"
15
+ },
16
+ "root": "/home/ext_minje/groot_insight/Abs_6D/Baseline",
17
+ "host": "DGX-H200-01",
18
+ "executable": "/home/ext_minje/miniforge3/envs/lerobot060_groot/bin/python",
19
+ "cpu_count": 112,
20
+ "cpu_count_logical": 224,
21
+ "gpu": "NVIDIA H200",
22
+ "gpu_count": 8,
23
+ "disk": {
24
+ "/": {
25
+ "total": "1888425144320",
26
+ "used": "1297062400000"
27
+ }
28
+ },
29
+ "memory": {
30
+ "total": "2164177248256"
31
+ },
32
+ "gpu_nvidia": [
33
+ {
34
+ "name": "NVIDIA H200",
35
+ "memoryTotal": "150754820096",
36
+ "cudaCores": 16896,
37
+ "architecture": "Hopper",
38
+ "uuid": "GPU-1e531a45-480e-3c1a-e9a0-2ba2cc27b0a3"
39
+ },
40
+ {
41
+ "name": "NVIDIA H200",
42
+ "memoryTotal": "150754820096",
43
+ "cudaCores": 16896,
44
+ "architecture": "Hopper",
45
+ "uuid": "GPU-5d3d5ae2-ffde-89d9-8818-dc19e11ad2fb"
46
+ },
47
+ {
48
+ "name": "NVIDIA H200",
49
+ "memoryTotal": "150754820096",
50
+ "cudaCores": 16896,
51
+ "architecture": "Hopper",
52
+ "uuid": "GPU-305c0dae-dcf5-d7bb-7dd3-5477d2194275"
53
+ },
54
+ {
55
+ "name": "NVIDIA H200",
56
+ "memoryTotal": "150754820096",
57
+ "cudaCores": 16896,
58
+ "architecture": "Hopper",
59
+ "uuid": "GPU-ed8b7920-8881-4e8f-3e88-be1bbecafa21"
60
+ },
61
+ {
62
+ "name": "NVIDIA H200",
63
+ "memoryTotal": "150754820096",
64
+ "cudaCores": 16896,
65
+ "architecture": "Hopper",
66
+ "uuid": "GPU-a91176db-55f2-b284-bbbf-56626596947d"
67
+ },
68
+ {
69
+ "name": "NVIDIA H200",
70
+ "memoryTotal": "150754820096",
71
+ "cudaCores": 16896,
72
+ "architecture": "Hopper",
73
+ "uuid": "GPU-922ff6d5-a948-205a-0426-dbee495fc3dd"
74
+ },
75
+ {
76
+ "name": "NVIDIA H200",
77
+ "memoryTotal": "150754820096",
78
+ "cudaCores": 16896,
79
+ "architecture": "Hopper",
80
+ "uuid": "GPU-1fde7c75-18fe-e3db-a4c1-85e67522f55f"
81
+ },
82
+ {
83
+ "name": "NVIDIA H200",
84
+ "memoryTotal": "150754820096",
85
+ "cudaCores": 16896,
86
+ "architecture": "Hopper",
87
+ "uuid": "GPU-a9b1c49b-d527-97d8-ce91-7e50bd02154c"
88
+ }
89
+ ],
90
+ "cudaVersion": "12.9",
91
+ "writerId": "c8d5ybgsruh4m5h3ie5jd18btesa3xht"
92
+ }
Baseline/wandb/run-20260715_143250-1jmjb68j/files/wandb-summary.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"_step":60000,"_wandb":{"runtime":97011},"_timestamp":1.7841102141629555e+09,"train/loss":0.020864611491560936,"train/update_s":0.909430973721901,"train/steps":60000,"train/lr":9.22312585679652e-10,"train/samples_per_s":38.035328348668635,"train/dataloading_s":0.773215207568137,"train/episodes":86694.41811262691,"train/grad_norm":0.2188143553212285,"train/epochs":17.58507466787564,"train/gpu_mem_gb":39.59298655033112,"train/samples":3840000,"_runtime":97011.728859976}
Baseline/wandb/run-20260715_143250-1jmjb68j/logs/debug-core.log ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"time":"2026-07-15T14:32:50.327781466+09:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpi0plx0gf/port-1450496.txt","pid":1450496,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
2
+ {"time":"2026-07-15T14:32:50.329194121+09:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":1450496}
3
+ {"time":"2026-07-15T14:32:50.329153734+09:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-1450496-1455451-2602571612/socket","Net":"unix"}}
4
+ {"time":"2026-07-15T14:32:50.495264696+09:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
5
+ {"time":"2026-07-15T14:32:50.504968593+09:00","level":"INFO","msg":"handleInformInit: received","streamId":"1jmjb68j","id":"1(@)"}
6
+ {"time":"2026-07-15T14:32:50.987680838+09:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"1jmjb68j","id":"1(@)"}
7
+ {"time":"2026-07-15T14:32:56.78660984+09:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"yyxntolayrj4"}
8
+ {"time":"2026-07-15T19:10:44.42243628+09:00","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
9
+ {"time":"2026-07-15T19:10:44.422612083+09:00","level":"INFO","msg":"connection: closing","id":"1(@)"}
10
+ {"time":"2026-07-15T19:10:44.422751956+09:00","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
11
+ {"time":"2026-07-15T19:10:44.422768684+09:00","level":"INFO","msg":"server is shutting down"}
12
+ {"time":"2026-07-15T19:10:44.422794402+09:00","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
13
+ {"time":"2026-07-15T19:10:44.422874606+09:00","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-1450496-1455451-2602571612/socket","Net":"unix"}}
14
+ {"time":"2026-07-15T19:10:46.287752374+09:00","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
15
+ {"time":"2026-07-15T19:10:46.287778824+09:00","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
16
+ {"time":"2026-07-15T19:10:46.287789838+09:00","level":"INFO","msg":"server is closed"}
Baseline/wandb/run-20260715_143250-1jmjb68j/logs/debug-internal.log ADDED
The diff for this file is too large to render. See raw diff
 
Baseline/wandb/run-20260715_143250-1jmjb68j/logs/debug.log ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2026-07-15 14:32:50,279 INFO MainThread:1450496 [wandb_setup.py:_flush():81] Current SDK version is 0.27.2
2
+ 2026-07-15 14:32:50,279 INFO MainThread:1450496 [wandb_setup.py:_flush():81] Configure stats pid to 1450496
3
+ 2026-07-15 14:32:50,279 INFO MainThread:1450496 [wandb_setup.py:_flush():81] Loading settings from environment variables
4
+ 2026-07-15 14:32:50,279 INFO MainThread:1450496 [wandb_init.py:setup_run_log_directory():723] Logging user logs to /home/ext_minje/groot_insight/Abs_6D/Baseline/wandb/run-20260715_143250-1jmjb68j/logs/debug.log
5
+ 2026-07-15 14:32:50,279 INFO MainThread:1450496 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to /home/ext_minje/groot_insight/Abs_6D/Baseline/wandb/run-20260715_143250-1jmjb68j/logs/debug-internal.log
6
+ 2026-07-15 14:32:50,279 INFO MainThread:1450496 [wandb_init.py:init():766] calling init triggers
7
+ 2026-07-15 14:32:50,280 INFO MainThread:1450496 [wandb_init.py:init():771] wandb.init called with sweep_config: {}
8
+ config: {'dataset': {'repo_id': 'Whalswp/INSIGHTfixposV4_filtered_multispace_v2', 'root': '/home/ext_minje/INSIGHTfixposV4_filtered_multispace_v2', 'episodes': None, 'image_transforms': {'enable': False, 'max_num_transforms': 3, 'random_order': False, 'tfs': {'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}, 'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'affine': {'weight': 1.0, 'type': 'RandomAffine', 'kwargs': {'degrees': [-5.0, 5.0], 'translate': [0.05, 0.05]}}}}, 'revision': None, 'use_imagenet_stats': True, 'video_backend': 'torchcodec', 'return_uint8': False, 'depth_output_unit': 'mm', 'streaming': False, 'eval_split': 0.0}, 'env': None, 'policy': {'type': 'groot', 'n_obs_steps': 1, 'input_features': {'observation.state': {'type': <FeatureType.STATE: 'STATE'>, 'shape': [16]}, 'observation.images.wrist': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 224, 224]}, 'observation.images.right_shoulder': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 224, 224]}, 'observation.images.guide': {'type': <FeatureType.VISUAL: 'VISUAL'>, 'shape': [3, 224, 224]}}, 'output_features': {'action': {'type': <FeatureType.ACTION: 'ACTION'>, 'shape': [10]}}, 'device': 'cuda', 'use_amp': False, 'use_peft': False, 'push_to_hub': False, 'repo_id': None, 'private': None, 'tags': None, 'license': None, 'pretrained_path': '/home/ext_minje/groot_insight/Abs_6D/Baseline/checkpoints/050000/pretrained_model', 'pretrained_revision': None, 'chunk_size': 16, 'n_action_steps': 16, 'max_state_dim': 132, 'max_action_dim': 132, 'normalization_mapping': {'VISUAL': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'STATE': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'ACTION': <NormalizationMode.IDENTITY: 'IDENTITY'>}, 'base_model_path': 'nvidia/GR00T-N1.7-3B', 'action_decode_transform': None, 'embodiment_tag': 'new_embodiment', 'tune_llm': False, 'tune_visual': False, 'tune_projector': True, 'tune_diffusion_model': True, 'tune_vlln': True, 'tune_top_llm_layers': 0, 'num_inference_timesteps': None, 'rtc_ramp_rate': None, 'use_flash_attention': False, 'use_relative_actions': False, 'relative_exclude_joints': [], 'optimizer_lr': 0.0001, 'optimizer_betas': [0.9, 0.999], 'optimizer_eps': 1e-08, 'optimizer_weight_decay': 1e-05, 'warmup_ratio': 0.05, 'use_bf16': True, 'model_params_fp32': True, 'image_size': [256, 256], 'tokenizer_assets_repo': None, 'lora_rank': 0, 'lora_alpha': 16, 'lora_dropout': 0.1, 'lora_full_model': False, 'video_backend': 'decord', 'balance_dataset_weights': True, 'balance_trajectory_weights': True, 'dataset_paths': None, 'output_dir': './tmp/gr00t', 'save_steps': 1000, 'max_steps': 10000, 'batch_size': 32, 'dataloader_num_workers': 8, 'report_to': 'wandb', 'resume': False}, 'reward_model': None, 'output_dir': '/home/ext_minje/groot_insight/Abs_6D/Baseline', 'job_name': 'INSIGHT_6D_baseline', 'resume': True, 'seed': 42, 'cudnn_deterministic': False, 'num_workers': 4, 'batch_size': 64, 'prefetch_factor': 4, 'persistent_workers': True, 'steps': 60000, 'env_eval_freq': 20000, 'log_freq': 200, 'eval_steps': 0, 'max_eval_samples': 0, 'tolerance_s': 0.0001, 'save_checkpoint': True, 'save_freq': 20000, 'use_policy_training_preset': True, 'optimizer': {'type': 'adamw', 'lr': 0.0001, 'weight_decay': 1e-05, 'grad_clip_norm': 1.0, 'betas': [0.9, 0.999], 'eps': 1e-08}, 'scheduler': {'type': 'diffuser', 'num_warmup_steps': 500, 'name': 'cosine'}, 'eval': {'n_episodes': 50, 'batch_size': 50, 'use_async_envs': True, 'recording': False, 'recording_repo_id': None, 'recording_private': False}, 'wandb': {'enable': True, 'disable_artifact': True, 'project': 'lerobot', 'entity': None, 'notes': None, 'run_id': '1jmjb68j', 'mode': None, 'add_tags': True}, 'peft': None, 'job': {'target': None, 'image': 'huggingface/lerobot-gpu:latest', 'timeout': '2d', 'detach': False, 'tags': []}, 'save_checkpoint_to_hub': False, 'sample_weighting': None, 'rename_map': {}, 'checkpoint_path': '/home/ext_minje/groot_insight/Abs_6D/Baseline/checkpoints/050000', '_wandb': {}}
9
+ 2026-07-15 14:32:50,280 INFO MainThread:1450496 [wandb_init.py:init():814] starting backend
10
+ 2026-07-15 14:32:50,495 INFO MainThread:1450496 [wandb_init.py:init():829] sending inform_init request
11
+ 2026-07-15 14:32:50,989 INFO MainThread:1450496 [wandb_init.py:init():834] backend started and connected
12
+ 2026-07-15 14:32:50,999 INFO MainThread:1450496 [wandb_init.py:init():904] updated telemetry
13
+ 2026-07-15 14:32:51,006 INFO MainThread:1450496 [wandb_init.py:init():927] communicating run to backend with 90.0 second timeout
14
+ 2026-07-15 14:32:51,687 INFO MainThread:1450496 [wandb_init.py:init():967] run resumed
15
+ 2026-07-15 14:32:51,692 INFO MainThread:1450496 [wandb_init.py:init():972] starting run threads in backend
16
+ 2026-07-15 14:32:51,783 INFO MainThread:1450496 [wandb_run.py:_console_start():2523] atexit reg
17
+ 2026-07-15 14:32:51,783 INFO MainThread:1450496 [wandb_run.py:_redirect():2373] redirect: wrap_raw
18
+ 2026-07-15 14:32:51,783 INFO MainThread:1450496 [wandb_run.py:_redirect():2442] Wrapping output streams.
19
+ 2026-07-15 14:32:51,783 INFO MainThread:1450496 [wandb_run.py:_redirect():2465] Redirects installed.
20
+ 2026-07-15 14:32:51,785 INFO MainThread:1450496 [wandb_init.py:init():1010] run started, returning control to user process
21
+ 2026-07-15 19:10:44,422 INFO wandb-AsyncioManager-main:1450496 [service_client.py:_forward_responses():122] Reached EOF.
22
+ 2026-07-15 19:10:44,422 INFO wandb-AsyncioManager-main:1450496 [mailbox.py:close():154] Closing mailbox, abandoning 2 handles.
23
+ 2026-07-15 19:10:44,423 ERROR wandb-AsyncioManager-main:1450496 [asyncio_manager.py:fn_wrap_exceptions():184] Uncaught exception in run_soon callback.
24
+ Traceback (most recent call last):
25
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/lib/asyncio_manager.py", line 182, in fn_wrap_exceptions
26
+ await fn()
27
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/lib/run_messages.py", line 90, in loop
28
+ await asyncio_compat.race(
29
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/lib/asyncio_compat.py", line 278, in race
30
+ async with open_task_group(race=True) as tg:
31
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^
32
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/contextlib.py", line 217, in __aexit__
33
+ await anext(self.gen)
34
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/lib/asyncio_compat.py", line 240, in open_task_group
35
+ await task_group._wait_all(race=race, timeout=exit_timeout)
36
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/lib/asyncio_compat.py", line 180, in _wait_all
37
+ raise exc
38
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/lib/run_messages.py", line 113, in _print_all
39
+ result = await handle.wait_async(timeout=None)
40
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
41
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/mailbox/mailbox_handle.py", line 127, in wait_async
42
+ response = await self._handle.wait_async(timeout=timeout)
43
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
44
+ File "/home/ext_minje/miniforge3/envs/lerobot060_groot/lib/python3.12/site-packages/wandb/sdk/mailbox/response_handle.py", line 123, in wait_async
45
+ raise HandleAbandonedError()
46
+ wandb.sdk.mailbox.mailbox_handle.HandleAbandonedError
Baseline/wandb/run-20260715_143250-1jmjb68j/run-1jmjb68j.wandb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6b40f8772cc119d7daa083c8228108c48fae49e5deb321ec0a3299944ff5322
3
+ size 9203799
Baseline_gpu3_resume.log ADDED
The diff for this file is too large to render. See raw diff
 
Baseline_gpu3_resume.pid ADDED
@@ -0,0 +1 @@
 
 
1
+ 1450496
Baseline_gpu3_watcher.log ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [2026-07-15T10:32:23+09:00] waiting for 50000: checkpoint directory not created
2
+ [2026-07-15T14:31:53+09:00] waiting for 50000: last checkpoint symlink not created
3
+ [2026-07-15T14:32:23+09:00] validated checkpoint 50000; GPU3 has 99644 MiB free
4
+ [2026-07-15T14:32:23+09:00] stopping GPU0 baseline PID 201098 with SIGTERM
5
+ [2026-07-15T14:32:29+09:00] GPU0 baseline process tree stopped cleanly
6
+ [2026-07-15T14:32:39+09:00] launched tmux session insight_baseline_g3 on GPU3
7
+ [2026-07-15T14:33:15+09:00] ERROR: resumed process environment mismatch: {'INSIGHT_LEROBOT_ACTION_SPACE': {'actual': None, 'expected': 'ee_abs_rot6d'}, 'INSIGHT_LEROBOT_PROMPT_SET': {'actual': None, 'expected': 'detailed'}, 'INSIGHT_LEROBOT_SAVE_STEPS': {'actual': None, 'expected': '60000'}}
8
+ Traceback (most recent call last):
9
+ File "/home/ext_minje/clvla/benchmarks/INSIGHT/my_scripts/watch_move_baseline_to_gpu3.py", line 475, in <module>
10
+ raise SystemExit(main())
11
+ ^^^^^^
12
+ File "/home/ext_minje/clvla/benchmarks/INSIGHT/my_scripts/watch_move_baseline_to_gpu3.py", line 452, in main
13
+ resume_details = verify_resume(
14
+ ^^^^^^^^^^^^^^
15
+ File "/home/ext_minje/clvla/benchmarks/INSIGHT/my_scripts/watch_move_baseline_to_gpu3.py", line 298, in verify_resume
16
+ raise RuntimeError(f"resumed process environment mismatch: {mismatches}")
17
+ RuntimeError: resumed process environment mismatch: {'INSIGHT_LEROBOT_ACTION_SPACE': {'actual': None, 'expected': 'ee_abs_rot6d'}, 'INSIGHT_LEROBOT_PROMPT_SET': {'actual': None, 'expected': 'detailed'}, 'INSIGHT_LEROBOT_SAVE_STEPS': {'actual': None, 'expected': '60000'}}
Baseline_gpu3_watcher_status.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "error": "resumed process environment mismatch: {'INSIGHT_LEROBOT_ACTION_SPACE': {'actual': None, 'expected': 'ee_abs_rot6d'}, 'INSIGHT_LEROBOT_PROMPT_SET': {'actual': None, 'expected': 'detailed'}, 'INSIGHT_LEROBOT_SAVE_STEPS': {'actual': None, 'expected': '60000'}}",
3
+ "state": "error",
4
+ "updated_at": "2026-07-15T14:33:15+09:00"
5
+ }
Baseline_train.log ADDED
The diff for this file is too large to render. See raw diff
 
Baseline_train.pid ADDED
@@ -0,0 +1 @@
 
 
1
+ 201098
RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/debug.log ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_setup.py:_flush():81] Current SDK version is 0.27.2
2
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_setup.py:_flush():81] Configure stats pid to 2695440
3
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_setup.py:_flush():81] Loading settings from environment variables
4
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_init.py:setup_run_log_directory():723] Logging user logs to /home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/run-20260715_011448-fbqzmp5m/logs/debug.log
5
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to /home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/run-20260715_011448-fbqzmp5m/logs/debug-internal.log
6
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_init.py:init():766] calling init triggers
7
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_init.py:init():771] wandb.init called with sweep_config: {}
8
+ config: {'dataset': {'repo_id': 'Whalswp/INSIGHTfixposV4_filtered_multispace_v2', 'root': '/home/ext_minje/INSIGHTfixposV4_filtered_multispace_v2', 'episodes': None, 'image_transforms': {'enable': False, 'max_num_transforms': 3, 'random_order': False, 'tfs': {'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}, 'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'affine': {'weight': 1.0, 'type': 'RandomAffine', 'kwargs': {'degrees': [-5.0, 5.0], 'translate': [0.05, 0.05]}}}}, 'revision': None, 'use_imagenet_stats': True, 'video_backend': 'torchcodec', 'return_uint8': False, 'depth_output_unit': 'mm', 'streaming': False, 'eval_split': 0.0}, 'env': None, 'policy': {'type': 'groot_rkd', 'n_obs_steps': 1, 'input_features': {}, 'output_features': {}, 'device': 'cuda', 'use_amp': False, 'use_peft': False, 'push_to_hub': False, 'repo_id': None, 'private': None, 'tags': None, 'license': None, 'pretrained_path': None, 'pretrained_revision': None, 'chunk_size': 16, 'n_action_steps': 16, 'max_state_dim': 132, 'max_action_dim': 132, 'normalization_mapping': {'VISUAL': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'STATE': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'ACTION': <NormalizationMode.IDENTITY: 'IDENTITY'>}, 'base_model_path': 'nvidia/GR00T-N1.7-3B', 'action_decode_transform': None, 'embodiment_tag': 'new_embodiment', 'tune_llm': False, 'tune_visual': False, 'tune_projector': True, 'tune_diffusion_model': True, 'tune_vlln': True, 'tune_top_llm_layers': 0, 'num_inference_timesteps': None, 'rtc_ramp_rate': None, 'use_flash_attention': False, 'use_relative_actions': False, 'relative_exclude_joints': [], 'optimizer_lr': 0.0001, 'optimizer_betas': [0.9, 0.999], 'optimizer_eps': 1e-08, 'optimizer_weight_decay': 1e-05, 'warmup_steps': 6000, 'warmup_ratio': 0.05, 'use_bf16': True, 'model_params_fp32': True, 'image_size': [256, 256], 'tokenizer_assets_repo': None, 'lora_rank': 0, 'lora_alpha': 16, 'lora_dropout': 0.1, 'lora_full_model': False, 'video_backend': 'decord', 'balance_dataset_weights': True, 'balance_trajectory_weights': True, 'dataset_paths': None, 'output_dir': './tmp/gr00t', 'save_steps': 1000, 'max_steps': 10000, 'batch_size': 32, 'dataloader_num_workers': 8, 'report_to': 'wandb', 'resume': False, 'rkd_enabled': True, 'rkd_vae_checkpoint': '/home/ext_minje/groot_insight/TimewarpVAE_combo_top8_30k/C8_01_C3_008_beta_5em05_latent_dim_16_granularity_4_decoder_hidden_dim_512/checkpoint-step-30000-epoch-39.11', 'rkd_student_pooling': 'attention', 'rkd_student_attention_hidden_dim': 512, 'rkd_student_attention_output_dim': 512, 'rkd_fm_loss_weight': 1.0, 'rkd_loss_weight': 1.0, 'rkd_distance_weight': 1.0, 'rkd_angle_weight': 2.0}, 'reward_model': None, 'output_dir': '/home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_10K', 'job_name': 'INSIGHT_6D_RKD_TimewarpVAE_LAPstyle_10K', 'resume': False, 'seed': 42, 'cudnn_deterministic': False, 'num_workers': 4, 'batch_size': 64, 'prefetch_factor': 4, 'persistent_workers': True, 'steps': 60000, 'env_eval_freq': 20000, 'log_freq': 200, 'eval_steps': 0, 'max_eval_samples': 0, 'tolerance_s': 0.0001, 'save_checkpoint': True, 'save_freq': 20000, 'use_policy_training_preset': True, 'optimizer': {'type': 'adamw', 'lr': 0.0001, 'weight_decay': 1e-05, 'grad_clip_norm': 1.0, 'betas': [0.9, 0.999], 'eps': 1e-08}, 'scheduler': {'type': 'diffuser', 'num_warmup_steps': 6000, 'name': 'cosine'}, 'eval': {'n_episodes': 50, 'batch_size': 50, 'use_async_envs': True, 'recording': False, 'recording_repo_id': None, 'recording_private': False}, 'wandb': {'enable': True, 'disable_artifact': True, 'project': 'lerobot', 'entity': None, 'notes': None, 'run_id': None, 'mode': None, 'add_tags': True}, 'peft': None, 'job': {'target': None, 'image': 'huggingface/lerobot-gpu:latest', 'timeout': '2d', 'detach': False, 'tags': []}, 'save_checkpoint_to_hub': False, 'sample_weighting': None, 'rename_map': {}, 'checkpoint_path': None, '_wandb': {}}
9
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_init.py:init():814] starting backend
10
+ 2026-07-15 01:14:48,572 INFO MainThread:2695440 [wandb_init.py:init():829] sending inform_init request
11
+ 2026-07-15 01:14:49,149 INFO MainThread:2695440 [wandb_init.py:init():834] backend started and connected
12
+ 2026-07-15 01:14:49,152 INFO MainThread:2695440 [wandb_init.py:init():904] updated telemetry
13
+ 2026-07-15 01:14:49,159 INFO MainThread:2695440 [wandb_init.py:init():927] communicating run to backend with 90.0 second timeout
14
+ 2026-07-15 01:14:49,692 INFO MainThread:2695440 [wandb_init.py:init():972] starting run threads in backend
15
+ 2026-07-15 01:14:49,747 INFO MainThread:2695440 [wandb_run.py:_console_start():2523] atexit reg
16
+ 2026-07-15 01:14:49,748 INFO MainThread:2695440 [wandb_run.py:_redirect():2373] redirect: wrap_raw
17
+ 2026-07-15 01:14:49,748 INFO MainThread:2695440 [wandb_run.py:_redirect():2442] Wrapping output streams.
18
+ 2026-07-15 01:14:49,748 INFO MainThread:2695440 [wandb_run.py:_redirect():2465] Redirects installed.
19
+ 2026-07-15 01:14:49,750 INFO MainThread:2695440 [wandb_init.py:init():1010] run started, returning control to user process
RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/run-20260715_011448-fbqzmp5m/files/requirements.txt ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ pyarrow==25.0.0
2
+ importlib_metadata==9.0.0
3
+ wheel==0.47.0
4
+ cloudpickle==3.1.2
5
+ nvidia-cusolver-cu12==11.7.3.90
6
+ multiprocess==0.70.19
7
+ torch==2.11.0+cu128
8
+ nvidia-nvjitlink-cu12==12.8.93
9
+ fsspec==2026.2.0
10
+ pytz==2026.2
11
+ hf-xet==1.5.1
12
+ tokenizers==0.22.2
13
+ opencv-python-headless==4.13.0.92
14
+ cuda-pathfinder==1.2.2
15
+ frozenlist==1.8.0
16
+ smmap==5.0.3
17
+ httpcore==1.0.9
18
+ nvidia-nvshmem-cu12==3.4.5
19
+ nvidia-cuda-nvrtc-cu12==12.8.93
20
+ annotated-types==0.7.0
21
+ platformdirs==4.10.0
22
+ click==8.4.2
23
+ pyyaml-include==1.4.1
24
+ MarkupSafe==3.0.3
25
+ Jinja2==3.1.6
26
+ urllib3==2.7.0
27
+ nvidia-curand-cu12==10.3.9.90
28
+ certifi==2026.6.17
29
+ nvidia-cuda-runtime-cu12==12.8.90
30
+ torchcodec==0.11.1
31
+ toml==0.10.2
32
+ av==15.1.0
33
+ nvidia-cufft-cu12==11.3.3.83
34
+ attrs==26.1.0
35
+ lerobot==0.6.0
36
+ PyYAML==6.0.3
37
+ safetensors==0.8.0
38
+ diffusers==0.35.2
39
+ requests==2.34.2
40
+ nvidia-nvtx-cu12==12.8.90
41
+ charset-normalizer==3.4.9
42
+ networkx==3.6.1
43
+ cmake==4.1.3
44
+ pydantic_core==2.46.4
45
+ timm==1.0.28
46
+ dm-tree==0.1.10
47
+ nvidia-cufile-cu12==1.13.1.3
48
+ termcolor==3.3.0
49
+ multidict==6.7.1
50
+ filelock==3.29.0
51
+ tzdata==2026.3
52
+ aiohappyeyeballs==2.7.1
53
+ httpx==0.28.1
54
+ nvidia-nccl-cu12==2.28.9
55
+ Pygments==2.20.0
56
+ einops==0.8.2
57
+ nvidia-cusparse-cu12==12.5.8.93
58
+ torchvision==0.26.0+cu128
59
+ dill==0.4.1
60
+ GitPython==3.1.51
61
+ nvidia-cuda-cupti-cu12==12.8.90
62
+ decord==0.6.0
63
+ sympy==1.14.0
64
+ cuda-toolkit==12.8.1
65
+ python-dateutil==2.9.0.post0
66
+ shellingham==1.5.4
67
+ propcache==0.5.2
68
+ Farama-Notifications==0.0.6
69
+ pillow==12.2.0
70
+ cuda-bindings==12.9.4
71
+ huggingface_hub==1.23.0
72
+ pandas==2.3.3
73
+ typing-inspect==0.9.0
74
+ packaging==25.0
75
+ rich==15.0.0
76
+ nvidia-cusparselt-cu12==0.7.1
77
+ setuptools==78.1.0
78
+ wrapt==2.2.2
79
+ mypy_extensions==1.1.0
80
+ tqdm==4.68.4
81
+ draccus==0.10.0
82
+ aiosignal==1.4.0
83
+ mpmath==1.3.0
84
+ numpy==2.2.6
85
+ typing_extensions==4.15.0
86
+ psutil==7.2.2
87
+ mdurl==0.1.2
88
+ nvidia-cublas-cu12==12.8.4.1
89
+ absl-py==2.5.0
90
+ gitdb==4.0.12
91
+ transformers==5.5.4
92
+ accelerate==1.14.0
93
+ pydantic==2.13.4
94
+ pip==26.1.2
95
+ markdown-it-py==4.2.0
96
+ typer==0.26.8
97
+ triton==3.6.0
98
+ wandb==0.27.2
99
+ regex==2026.7.10
100
+ yarl==1.24.2
101
+ jsonlines==4.0.0
102
+ mergedeep==1.3.4
103
+ aiohttp==3.14.1
104
+ datasets==4.8.5
105
+ gymnasium==1.3.0
106
+ peft==0.19.1
107
+ zipp==4.1.0
108
+ idna==3.18
109
+ nvidia-cudnn-cu12==9.19.0.56
110
+ h11==0.16.0
111
+ sentry-sdk==2.65.0
112
+ annotated-doc==0.0.4
113
+ protobuf==7.35.1
114
+ typing-inspection==0.4.2
115
+ anyio==4.14.2
116
+ xxhash==3.8.1
117
+ six==1.17.0
RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/run-20260715_011448-fbqzmp5m/files/wandb-metadata.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "os": "Linux-6.8.0-107-generic-x86_64-with-glibc2.39",
3
+ "python": "CPython 3.12.13",
4
+ "startedAt": "2026-07-14T16:14:48.356781Z",
5
+ "args": [
6
+ "--config_path=/tmp/insight_lerobot_cfg_t5yu3xzr/LAPstyle_RKD_linear_decay_10K.yaml"
7
+ ],
8
+ "program": "/home/ext_minje/clvla/benchmarks/INSIGHT/my_scripts/lerobot_train_yaml.py",
9
+ "codePath": "benchmarks/INSIGHT/my_scripts/lerobot_train_yaml.py",
10
+ "codePathLocal": "my_scripts/lerobot_train_yaml.py",
11
+ "git": {
12
+ "remote": "git@github.com:paragon7060/clvla.git",
13
+ "commit": "22b2faefeaab966ca2dc3688fe6a77f3bcd8de32"
14
+ },
15
+ "root": "/home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_10K",
16
+ "host": "DGX-H200-01",
17
+ "executable": "/home/ext_minje/miniforge3/envs/lerobot060_groot/bin/python",
18
+ "cpu_count": 112,
19
+ "cpu_count_logical": 224,
20
+ "gpu": "NVIDIA H200",
21
+ "gpu_count": 8,
22
+ "disk": {
23
+ "/": {
24
+ "total": "1888425144320",
25
+ "used": "1262682583040"
26
+ }
27
+ },
28
+ "memory": {
29
+ "total": "2164177248256"
30
+ },
31
+ "gpu_nvidia": [
32
+ {
33
+ "name": "NVIDIA H200",
34
+ "memoryTotal": "150754820096",
35
+ "cudaCores": 16896,
36
+ "architecture": "Hopper",
37
+ "uuid": "GPU-1e531a45-480e-3c1a-e9a0-2ba2cc27b0a3"
38
+ },
39
+ {
40
+ "name": "NVIDIA H200",
41
+ "memoryTotal": "150754820096",
42
+ "cudaCores": 16896,
43
+ "architecture": "Hopper",
44
+ "uuid": "GPU-5d3d5ae2-ffde-89d9-8818-dc19e11ad2fb"
45
+ },
46
+ {
47
+ "name": "NVIDIA H200",
48
+ "memoryTotal": "150754820096",
49
+ "cudaCores": 16896,
50
+ "architecture": "Hopper",
51
+ "uuid": "GPU-305c0dae-dcf5-d7bb-7dd3-5477d2194275"
52
+ },
53
+ {
54
+ "name": "NVIDIA H200",
55
+ "memoryTotal": "150754820096",
56
+ "cudaCores": 16896,
57
+ "architecture": "Hopper",
58
+ "uuid": "GPU-ed8b7920-8881-4e8f-3e88-be1bbecafa21"
59
+ },
60
+ {
61
+ "name": "NVIDIA H200",
62
+ "memoryTotal": "150754820096",
63
+ "cudaCores": 16896,
64
+ "architecture": "Hopper",
65
+ "uuid": "GPU-a91176db-55f2-b284-bbbf-56626596947d"
66
+ },
67
+ {
68
+ "name": "NVIDIA H200",
69
+ "memoryTotal": "150754820096",
70
+ "cudaCores": 16896,
71
+ "architecture": "Hopper",
72
+ "uuid": "GPU-922ff6d5-a948-205a-0426-dbee495fc3dd"
73
+ },
74
+ {
75
+ "name": "NVIDIA H200",
76
+ "memoryTotal": "150754820096",
77
+ "cudaCores": 16896,
78
+ "architecture": "Hopper",
79
+ "uuid": "GPU-1fde7c75-18fe-e3db-a4c1-85e67522f55f"
80
+ },
81
+ {
82
+ "name": "NVIDIA H200",
83
+ "memoryTotal": "150754820096",
84
+ "cudaCores": 16896,
85
+ "architecture": "Hopper",
86
+ "uuid": "GPU-a9b1c49b-d527-97d8-ce91-7e50bd02154c"
87
+ }
88
+ ],
89
+ "cudaVersion": "12.9",
90
+ "writerId": "pmztphx5rdwrxjq7nyo10wcg6qmd0739"
91
+ }
RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/run-20260715_011448-fbqzmp5m/logs/debug-core.log ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"time":"2026-07-15T01:14:48.303966233+09:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp60lspfb4/port-2695434.txt","pid":2695434,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
2
+ {"time":"2026-07-15T01:14:48.305408197+09:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":2695434}
3
+ {"time":"2026-07-15T01:14:48.305383274+09:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-2695434-2698143-952461375/socket","Net":"unix"}}
4
+ {"time":"2026-07-15T01:14:48.396793976+09:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpl1f6awb8/port-2695440.txt","pid":2695440,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
5
+ {"time":"2026-07-15T01:14:48.39732489+09:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":2695440}
6
+ {"time":"2026-07-15T01:14:48.397322762+09:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-2695440-2698230-1148333561/socket","Net":"unix"}}
7
+ {"time":"2026-07-15T01:14:48.397951076+09:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp8x_1y0jo/port-2695448.txt","pid":2695448,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
8
+ {"time":"2026-07-15T01:14:48.398416675+09:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":2695448}
9
+ {"time":"2026-07-15T01:14:48.398410488+09:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-2695448-2698229-2622114070/socket","Net":"unix"}}
10
+ {"time":"2026-07-15T01:14:48.485161862+09:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
11
+ {"time":"2026-07-15T01:14:48.493169073+09:00","level":"INFO","msg":"handleInformInit: received","streamId":"pebh9ode","id":"1(@)"}
12
+ {"time":"2026-07-15T01:14:48.572552182+09:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
13
+ {"time":"2026-07-15T01:14:48.573290779+09:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
14
+ {"time":"2026-07-15T01:14:48.581300802+09:00","level":"INFO","msg":"handleInformInit: received","streamId":"fbqzmp5m","id":"1(@)"}
15
+ {"time":"2026-07-15T01:14:48.584399672+09:00","level":"INFO","msg":"handleInformInit: received","streamId":"unz77u4h","id":"1(@)"}
16
+ {"time":"2026-07-15T01:14:48.991319873+09:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"pebh9ode","id":"1(@)"}
17
+ {"time":"2026-07-15T01:14:49.148475362+09:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"fbqzmp5m","id":"1(@)"}
18
+ {"time":"2026-07-15T01:14:49.154720114+09:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"unz77u4h","id":"1(@)"}
19
+ {"time":"2026-07-15T01:14:54.704195769+09:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"gboffkhx5odh"}
20
+ {"time":"2026-07-15T01:14:54.755700856+09:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ktloe72dm2ag"}
21
+ {"time":"2026-07-15T01:14:54.84707997+09:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"gveiruljishv"}
22
+ {"time":"2026-07-15T10:20:55.252683572+09:00","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
23
+ {"time":"2026-07-15T10:20:55.252742779+09:00","level":"INFO","msg":"connection: closing","id":"1(@)"}
24
+ {"time":"2026-07-15T10:20:55.252786889+09:00","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
25
+ {"time":"2026-07-15T10:20:55.25278977+09:00","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
26
+ {"time":"2026-07-15T10:20:56.037707662+09:00","level":"INFO","msg":"server: parent process exited, terminating service process"}
RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/run-20260715_011448-fbqzmp5m/logs/debug.log ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_setup.py:_flush():81] Current SDK version is 0.27.2
2
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_setup.py:_flush():81] Configure stats pid to 2695440
3
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_setup.py:_flush():81] Loading settings from environment variables
4
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_init.py:setup_run_log_directory():723] Logging user logs to /home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/run-20260715_011448-fbqzmp5m/logs/debug.log
5
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to /home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_10K/wandb/run-20260715_011448-fbqzmp5m/logs/debug-internal.log
6
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_init.py:init():766] calling init triggers
7
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_init.py:init():771] wandb.init called with sweep_config: {}
8
+ config: {'dataset': {'repo_id': 'Whalswp/INSIGHTfixposV4_filtered_multispace_v2', 'root': '/home/ext_minje/INSIGHTfixposV4_filtered_multispace_v2', 'episodes': None, 'image_transforms': {'enable': False, 'max_num_transforms': 3, 'random_order': False, 'tfs': {'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}, 'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'affine': {'weight': 1.0, 'type': 'RandomAffine', 'kwargs': {'degrees': [-5.0, 5.0], 'translate': [0.05, 0.05]}}}}, 'revision': None, 'use_imagenet_stats': True, 'video_backend': 'torchcodec', 'return_uint8': False, 'depth_output_unit': 'mm', 'streaming': False, 'eval_split': 0.0}, 'env': None, 'policy': {'type': 'groot_rkd', 'n_obs_steps': 1, 'input_features': {}, 'output_features': {}, 'device': 'cuda', 'use_amp': False, 'use_peft': False, 'push_to_hub': False, 'repo_id': None, 'private': None, 'tags': None, 'license': None, 'pretrained_path': None, 'pretrained_revision': None, 'chunk_size': 16, 'n_action_steps': 16, 'max_state_dim': 132, 'max_action_dim': 132, 'normalization_mapping': {'VISUAL': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'STATE': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'ACTION': <NormalizationMode.IDENTITY: 'IDENTITY'>}, 'base_model_path': 'nvidia/GR00T-N1.7-3B', 'action_decode_transform': None, 'embodiment_tag': 'new_embodiment', 'tune_llm': False, 'tune_visual': False, 'tune_projector': True, 'tune_diffusion_model': True, 'tune_vlln': True, 'tune_top_llm_layers': 0, 'num_inference_timesteps': None, 'rtc_ramp_rate': None, 'use_flash_attention': False, 'use_relative_actions': False, 'relative_exclude_joints': [], 'optimizer_lr': 0.0001, 'optimizer_betas': [0.9, 0.999], 'optimizer_eps': 1e-08, 'optimizer_weight_decay': 1e-05, 'warmup_steps': 6000, 'warmup_ratio': 0.05, 'use_bf16': True, 'model_params_fp32': True, 'image_size': [256, 256], 'tokenizer_assets_repo': None, 'lora_rank': 0, 'lora_alpha': 16, 'lora_dropout': 0.1, 'lora_full_model': False, 'video_backend': 'decord', 'balance_dataset_weights': True, 'balance_trajectory_weights': True, 'dataset_paths': None, 'output_dir': './tmp/gr00t', 'save_steps': 1000, 'max_steps': 10000, 'batch_size': 32, 'dataloader_num_workers': 8, 'report_to': 'wandb', 'resume': False, 'rkd_enabled': True, 'rkd_vae_checkpoint': '/home/ext_minje/groot_insight/TimewarpVAE_combo_top8_30k/C8_01_C3_008_beta_5em05_latent_dim_16_granularity_4_decoder_hidden_dim_512/checkpoint-step-30000-epoch-39.11', 'rkd_student_pooling': 'attention', 'rkd_student_attention_hidden_dim': 512, 'rkd_student_attention_output_dim': 512, 'rkd_fm_loss_weight': 1.0, 'rkd_loss_weight': 1.0, 'rkd_distance_weight': 1.0, 'rkd_angle_weight': 2.0}, 'reward_model': None, 'output_dir': '/home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_10K', 'job_name': 'INSIGHT_6D_RKD_TimewarpVAE_LAPstyle_10K', 'resume': False, 'seed': 42, 'cudnn_deterministic': False, 'num_workers': 4, 'batch_size': 64, 'prefetch_factor': 4, 'persistent_workers': True, 'steps': 60000, 'env_eval_freq': 20000, 'log_freq': 200, 'eval_steps': 0, 'max_eval_samples': 0, 'tolerance_s': 0.0001, 'save_checkpoint': True, 'save_freq': 20000, 'use_policy_training_preset': True, 'optimizer': {'type': 'adamw', 'lr': 0.0001, 'weight_decay': 1e-05, 'grad_clip_norm': 1.0, 'betas': [0.9, 0.999], 'eps': 1e-08}, 'scheduler': {'type': 'diffuser', 'num_warmup_steps': 6000, 'name': 'cosine'}, 'eval': {'n_episodes': 50, 'batch_size': 50, 'use_async_envs': True, 'recording': False, 'recording_repo_id': None, 'recording_private': False}, 'wandb': {'enable': True, 'disable_artifact': True, 'project': 'lerobot', 'entity': None, 'notes': None, 'run_id': None, 'mode': None, 'add_tags': True}, 'peft': None, 'job': {'target': None, 'image': 'huggingface/lerobot-gpu:latest', 'timeout': '2d', 'detach': False, 'tags': []}, 'save_checkpoint_to_hub': False, 'sample_weighting': None, 'rename_map': {}, 'checkpoint_path': None, '_wandb': {}}
9
+ 2026-07-15 01:14:48,357 INFO MainThread:2695440 [wandb_init.py:init():814] starting backend
10
+ 2026-07-15 01:14:48,572 INFO MainThread:2695440 [wandb_init.py:init():829] sending inform_init request
11
+ 2026-07-15 01:14:49,149 INFO MainThread:2695440 [wandb_init.py:init():834] backend started and connected
12
+ 2026-07-15 01:14:49,152 INFO MainThread:2695440 [wandb_init.py:init():904] updated telemetry
13
+ 2026-07-15 01:14:49,159 INFO MainThread:2695440 [wandb_init.py:init():927] communicating run to backend with 90.0 second timeout
14
+ 2026-07-15 01:14:49,692 INFO MainThread:2695440 [wandb_init.py:init():972] starting run threads in backend
15
+ 2026-07-15 01:14:49,747 INFO MainThread:2695440 [wandb_run.py:_console_start():2523] atexit reg
16
+ 2026-07-15 01:14:49,748 INFO MainThread:2695440 [wandb_run.py:_redirect():2373] redirect: wrap_raw
17
+ 2026-07-15 01:14:49,748 INFO MainThread:2695440 [wandb_run.py:_redirect():2442] Wrapping output streams.
18
+ 2026-07-15 01:14:49,748 INFO MainThread:2695440 [wandb_run.py:_redirect():2465] Redirects installed.
19
+ 2026-07-15 01:14:49,750 INFO MainThread:2695440 [wandb_init.py:init():1010] run started, returning control to user process
RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/debug.log ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_setup.py:_flush():81] Current SDK version is 0.27.2
2
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_setup.py:_flush():81] Configure stats pid to 2695434
3
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_setup.py:_flush():81] Loading settings from environment variables
4
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_init.py:setup_run_log_directory():723] Logging user logs to /home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/run-20260715_011448-pebh9ode/logs/debug.log
5
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to /home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/run-20260715_011448-pebh9ode/logs/debug-internal.log
6
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_init.py:init():766] calling init triggers
7
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_init.py:init():771] wandb.init called with sweep_config: {}
8
+ config: {'dataset': {'repo_id': 'Whalswp/INSIGHTfixposV4_filtered_multispace_v2', 'root': '/home/ext_minje/INSIGHTfixposV4_filtered_multispace_v2', 'episodes': None, 'image_transforms': {'enable': False, 'max_num_transforms': 3, 'random_order': False, 'tfs': {'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}, 'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'affine': {'weight': 1.0, 'type': 'RandomAffine', 'kwargs': {'degrees': [-5.0, 5.0], 'translate': [0.05, 0.05]}}}}, 'revision': None, 'use_imagenet_stats': True, 'video_backend': 'torchcodec', 'return_uint8': False, 'depth_output_unit': 'mm', 'streaming': False, 'eval_split': 0.0}, 'env': None, 'policy': {'type': 'groot_rkd', 'n_obs_steps': 1, 'input_features': {}, 'output_features': {}, 'device': 'cuda', 'use_amp': False, 'use_peft': False, 'push_to_hub': False, 'repo_id': None, 'private': None, 'tags': None, 'license': None, 'pretrained_path': None, 'pretrained_revision': None, 'chunk_size': 16, 'n_action_steps': 16, 'max_state_dim': 132, 'max_action_dim': 132, 'normalization_mapping': {'VISUAL': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'STATE': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'ACTION': <NormalizationMode.IDENTITY: 'IDENTITY'>}, 'base_model_path': 'nvidia/GR00T-N1.7-3B', 'action_decode_transform': None, 'embodiment_tag': 'new_embodiment', 'tune_llm': False, 'tune_visual': False, 'tune_projector': True, 'tune_diffusion_model': True, 'tune_vlln': True, 'tune_top_llm_layers': 0, 'num_inference_timesteps': None, 'rtc_ramp_rate': None, 'use_flash_attention': False, 'use_relative_actions': False, 'relative_exclude_joints': [], 'optimizer_lr': 0.0001, 'optimizer_betas': [0.9, 0.999], 'optimizer_eps': 1e-08, 'optimizer_weight_decay': 1e-05, 'warmup_steps': 6000, 'warmup_ratio': 0.05, 'use_bf16': True, 'model_params_fp32': True, 'image_size': [256, 256], 'tokenizer_assets_repo': None, 'lora_rank': 0, 'lora_alpha': 16, 'lora_dropout': 0.1, 'lora_full_model': False, 'video_backend': 'decord', 'balance_dataset_weights': True, 'balance_trajectory_weights': True, 'dataset_paths': None, 'output_dir': './tmp/gr00t', 'save_steps': 1000, 'max_steps': 10000, 'batch_size': 32, 'dataloader_num_workers': 8, 'report_to': 'wandb', 'resume': False, 'rkd_enabled': True, 'rkd_vae_checkpoint': '/home/ext_minje/groot_insight/TimewarpVAE_combo_top8_30k/C8_01_C3_008_beta_5em05_latent_dim_16_granularity_4_decoder_hidden_dim_512/checkpoint-step-30000-epoch-39.11', 'rkd_student_pooling': 'attention', 'rkd_student_attention_hidden_dim': 512, 'rkd_student_attention_output_dim': 512, 'rkd_fm_loss_weight': 1.0, 'rkd_loss_weight': 1.0, 'rkd_distance_weight': 1.0, 'rkd_angle_weight': 2.0}, 'reward_model': None, 'output_dir': '/home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_6K', 'job_name': 'INSIGHT_6D_RKD_TimewarpVAE_LAPstyle_6K', 'resume': False, 'seed': 42, 'cudnn_deterministic': False, 'num_workers': 4, 'batch_size': 64, 'prefetch_factor': 4, 'persistent_workers': True, 'steps': 60000, 'env_eval_freq': 20000, 'log_freq': 200, 'eval_steps': 0, 'max_eval_samples': 0, 'tolerance_s': 0.0001, 'save_checkpoint': True, 'save_freq': 20000, 'use_policy_training_preset': True, 'optimizer': {'type': 'adamw', 'lr': 0.0001, 'weight_decay': 1e-05, 'grad_clip_norm': 1.0, 'betas': [0.9, 0.999], 'eps': 1e-08}, 'scheduler': {'type': 'diffuser', 'num_warmup_steps': 6000, 'name': 'cosine'}, 'eval': {'n_episodes': 50, 'batch_size': 50, 'use_async_envs': True, 'recording': False, 'recording_repo_id': None, 'recording_private': False}, 'wandb': {'enable': True, 'disable_artifact': True, 'project': 'lerobot', 'entity': None, 'notes': None, 'run_id': None, 'mode': None, 'add_tags': True}, 'peft': None, 'job': {'target': None, 'image': 'huggingface/lerobot-gpu:latest', 'timeout': '2d', 'detach': False, 'tags': []}, 'save_checkpoint_to_hub': False, 'sample_weighting': None, 'rename_map': {}, 'checkpoint_path': None, '_wandb': {}}
9
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_init.py:init():814] starting backend
10
+ 2026-07-15 01:14:48,485 INFO MainThread:2695434 [wandb_init.py:init():829] sending inform_init request
11
+ 2026-07-15 01:14:48,992 INFO MainThread:2695434 [wandb_init.py:init():834] backend started and connected
12
+ 2026-07-15 01:14:48,995 INFO MainThread:2695434 [wandb_init.py:init():904] updated telemetry
13
+ 2026-07-15 01:14:49,005 INFO MainThread:2695434 [wandb_init.py:init():927] communicating run to backend with 90.0 second timeout
14
+ 2026-07-15 01:14:49,642 INFO MainThread:2695434 [wandb_init.py:init():972] starting run threads in backend
15
+ 2026-07-15 01:14:49,699 INFO MainThread:2695434 [wandb_run.py:_console_start():2523] atexit reg
16
+ 2026-07-15 01:14:49,699 INFO MainThread:2695434 [wandb_run.py:_redirect():2373] redirect: wrap_raw
17
+ 2026-07-15 01:14:49,699 INFO MainThread:2695434 [wandb_run.py:_redirect():2442] Wrapping output streams.
18
+ 2026-07-15 01:14:49,699 INFO MainThread:2695434 [wandb_run.py:_redirect():2465] Redirects installed.
19
+ 2026-07-15 01:14:49,701 INFO MainThread:2695434 [wandb_init.py:init():1010] run started, returning control to user process
RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/run-20260715_011448-pebh9ode/files/requirements.txt ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ pyarrow==25.0.0
2
+ importlib_metadata==9.0.0
3
+ wheel==0.47.0
4
+ cloudpickle==3.1.2
5
+ nvidia-cusolver-cu12==11.7.3.90
6
+ multiprocess==0.70.19
7
+ torch==2.11.0+cu128
8
+ nvidia-nvjitlink-cu12==12.8.93
9
+ fsspec==2026.2.0
10
+ pytz==2026.2
11
+ hf-xet==1.5.1
12
+ tokenizers==0.22.2
13
+ opencv-python-headless==4.13.0.92
14
+ cuda-pathfinder==1.2.2
15
+ frozenlist==1.8.0
16
+ smmap==5.0.3
17
+ httpcore==1.0.9
18
+ nvidia-nvshmem-cu12==3.4.5
19
+ nvidia-cuda-nvrtc-cu12==12.8.93
20
+ annotated-types==0.7.0
21
+ platformdirs==4.10.0
22
+ click==8.4.2
23
+ pyyaml-include==1.4.1
24
+ MarkupSafe==3.0.3
25
+ Jinja2==3.1.6
26
+ urllib3==2.7.0
27
+ nvidia-curand-cu12==10.3.9.90
28
+ certifi==2026.6.17
29
+ nvidia-cuda-runtime-cu12==12.8.90
30
+ torchcodec==0.11.1
31
+ toml==0.10.2
32
+ av==15.1.0
33
+ nvidia-cufft-cu12==11.3.3.83
34
+ attrs==26.1.0
35
+ lerobot==0.6.0
36
+ PyYAML==6.0.3
37
+ safetensors==0.8.0
38
+ diffusers==0.35.2
39
+ requests==2.34.2
40
+ nvidia-nvtx-cu12==12.8.90
41
+ charset-normalizer==3.4.9
42
+ networkx==3.6.1
43
+ cmake==4.1.3
44
+ pydantic_core==2.46.4
45
+ timm==1.0.28
46
+ dm-tree==0.1.10
47
+ nvidia-cufile-cu12==1.13.1.3
48
+ termcolor==3.3.0
49
+ multidict==6.7.1
50
+ filelock==3.29.0
51
+ tzdata==2026.3
52
+ aiohappyeyeballs==2.7.1
53
+ httpx==0.28.1
54
+ nvidia-nccl-cu12==2.28.9
55
+ Pygments==2.20.0
56
+ einops==0.8.2
57
+ nvidia-cusparse-cu12==12.5.8.93
58
+ torchvision==0.26.0+cu128
59
+ dill==0.4.1
60
+ GitPython==3.1.51
61
+ nvidia-cuda-cupti-cu12==12.8.90
62
+ decord==0.6.0
63
+ sympy==1.14.0
64
+ cuda-toolkit==12.8.1
65
+ python-dateutil==2.9.0.post0
66
+ shellingham==1.5.4
67
+ propcache==0.5.2
68
+ Farama-Notifications==0.0.6
69
+ pillow==12.2.0
70
+ cuda-bindings==12.9.4
71
+ huggingface_hub==1.23.0
72
+ pandas==2.3.3
73
+ typing-inspect==0.9.0
74
+ packaging==25.0
75
+ rich==15.0.0
76
+ nvidia-cusparselt-cu12==0.7.1
77
+ setuptools==78.1.0
78
+ wrapt==2.2.2
79
+ mypy_extensions==1.1.0
80
+ tqdm==4.68.4
81
+ draccus==0.10.0
82
+ aiosignal==1.4.0
83
+ mpmath==1.3.0
84
+ numpy==2.2.6
85
+ typing_extensions==4.15.0
86
+ psutil==7.2.2
87
+ mdurl==0.1.2
88
+ nvidia-cublas-cu12==12.8.4.1
89
+ absl-py==2.5.0
90
+ gitdb==4.0.12
91
+ transformers==5.5.4
92
+ accelerate==1.14.0
93
+ pydantic==2.13.4
94
+ pip==26.1.2
95
+ markdown-it-py==4.2.0
96
+ typer==0.26.8
97
+ triton==3.6.0
98
+ wandb==0.27.2
99
+ regex==2026.7.10
100
+ yarl==1.24.2
101
+ jsonlines==4.0.0
102
+ mergedeep==1.3.4
103
+ aiohttp==3.14.1
104
+ datasets==4.8.5
105
+ gymnasium==1.3.0
106
+ peft==0.19.1
107
+ zipp==4.1.0
108
+ idna==3.18
109
+ nvidia-cudnn-cu12==9.19.0.56
110
+ h11==0.16.0
111
+ sentry-sdk==2.65.0
112
+ annotated-doc==0.0.4
113
+ protobuf==7.35.1
114
+ typing-inspection==0.4.2
115
+ anyio==4.14.2
116
+ xxhash==3.8.1
117
+ six==1.17.0
RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/run-20260715_011448-pebh9ode/files/wandb-metadata.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "os": "Linux-6.8.0-107-generic-x86_64-with-glibc2.39",
3
+ "python": "CPython 3.12.13",
4
+ "startedAt": "2026-07-14T16:14:48.265745Z",
5
+ "args": [
6
+ "--config_path=/tmp/insight_lerobot_cfg_k5yvq_8p/LAPstyle_RKD_linear_decay_6K.yaml"
7
+ ],
8
+ "program": "/home/ext_minje/clvla/benchmarks/INSIGHT/my_scripts/lerobot_train_yaml.py",
9
+ "codePath": "benchmarks/INSIGHT/my_scripts/lerobot_train_yaml.py",
10
+ "codePathLocal": "my_scripts/lerobot_train_yaml.py",
11
+ "git": {
12
+ "remote": "git@github.com:paragon7060/clvla.git",
13
+ "commit": "22b2faefeaab966ca2dc3688fe6a77f3bcd8de32"
14
+ },
15
+ "root": "/home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_6K",
16
+ "host": "DGX-H200-01",
17
+ "executable": "/home/ext_minje/miniforge3/envs/lerobot060_groot/bin/python",
18
+ "cpu_count": 112,
19
+ "cpu_count_logical": 224,
20
+ "gpu": "NVIDIA H200",
21
+ "gpu_count": 8,
22
+ "disk": {
23
+ "/": {
24
+ "total": "1888425144320",
25
+ "used": "1262682492928"
26
+ }
27
+ },
28
+ "memory": {
29
+ "total": "2164177248256"
30
+ },
31
+ "gpu_nvidia": [
32
+ {
33
+ "name": "NVIDIA H200",
34
+ "memoryTotal": "150754820096",
35
+ "cudaCores": 16896,
36
+ "architecture": "Hopper",
37
+ "uuid": "GPU-1e531a45-480e-3c1a-e9a0-2ba2cc27b0a3"
38
+ },
39
+ {
40
+ "name": "NVIDIA H200",
41
+ "memoryTotal": "150754820096",
42
+ "cudaCores": 16896,
43
+ "architecture": "Hopper",
44
+ "uuid": "GPU-5d3d5ae2-ffde-89d9-8818-dc19e11ad2fb"
45
+ },
46
+ {
47
+ "name": "NVIDIA H200",
48
+ "memoryTotal": "150754820096",
49
+ "cudaCores": 16896,
50
+ "architecture": "Hopper",
51
+ "uuid": "GPU-305c0dae-dcf5-d7bb-7dd3-5477d2194275"
52
+ },
53
+ {
54
+ "name": "NVIDIA H200",
55
+ "memoryTotal": "150754820096",
56
+ "cudaCores": 16896,
57
+ "architecture": "Hopper",
58
+ "uuid": "GPU-ed8b7920-8881-4e8f-3e88-be1bbecafa21"
59
+ },
60
+ {
61
+ "name": "NVIDIA H200",
62
+ "memoryTotal": "150754820096",
63
+ "cudaCores": 16896,
64
+ "architecture": "Hopper",
65
+ "uuid": "GPU-a91176db-55f2-b284-bbbf-56626596947d"
66
+ },
67
+ {
68
+ "name": "NVIDIA H200",
69
+ "memoryTotal": "150754820096",
70
+ "cudaCores": 16896,
71
+ "architecture": "Hopper",
72
+ "uuid": "GPU-922ff6d5-a948-205a-0426-dbee495fc3dd"
73
+ },
74
+ {
75
+ "name": "NVIDIA H200",
76
+ "memoryTotal": "150754820096",
77
+ "cudaCores": 16896,
78
+ "architecture": "Hopper",
79
+ "uuid": "GPU-1fde7c75-18fe-e3db-a4c1-85e67522f55f"
80
+ },
81
+ {
82
+ "name": "NVIDIA H200",
83
+ "memoryTotal": "150754820096",
84
+ "cudaCores": 16896,
85
+ "architecture": "Hopper",
86
+ "uuid": "GPU-a9b1c49b-d527-97d8-ce91-7e50bd02154c"
87
+ }
88
+ ],
89
+ "cudaVersion": "12.9",
90
+ "writerId": "pdqtmahtvys2lc0hp4vh9nv4yumavqrr"
91
+ }
RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/run-20260715_011448-pebh9ode/logs/debug-core.log ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"time":"2026-07-15T01:14:48.303966233+09:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp60lspfb4/port-2695434.txt","pid":2695434,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
2
+ {"time":"2026-07-15T01:14:48.305408197+09:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":2695434}
3
+ {"time":"2026-07-15T01:14:48.305383274+09:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-2695434-2698143-952461375/socket","Net":"unix"}}
4
+ {"time":"2026-07-15T01:14:48.396793976+09:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpl1f6awb8/port-2695440.txt","pid":2695440,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
5
+ {"time":"2026-07-15T01:14:48.39732489+09:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":2695440}
6
+ {"time":"2026-07-15T01:14:48.397322762+09:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-2695440-2698230-1148333561/socket","Net":"unix"}}
7
+ {"time":"2026-07-15T01:14:48.397951076+09:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp8x_1y0jo/port-2695448.txt","pid":2695448,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
8
+ {"time":"2026-07-15T01:14:48.398416675+09:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":2695448}
9
+ {"time":"2026-07-15T01:14:48.398410488+09:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-2695448-2698229-2622114070/socket","Net":"unix"}}
10
+ {"time":"2026-07-15T01:14:48.485161862+09:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
11
+ {"time":"2026-07-15T01:14:48.493169073+09:00","level":"INFO","msg":"handleInformInit: received","streamId":"pebh9ode","id":"1(@)"}
12
+ {"time":"2026-07-15T01:14:48.572552182+09:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
13
+ {"time":"2026-07-15T01:14:48.573290779+09:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
14
+ {"time":"2026-07-15T01:14:48.581300802+09:00","level":"INFO","msg":"handleInformInit: received","streamId":"fbqzmp5m","id":"1(@)"}
15
+ {"time":"2026-07-15T01:14:48.584399672+09:00","level":"INFO","msg":"handleInformInit: received","streamId":"unz77u4h","id":"1(@)"}
16
+ {"time":"2026-07-15T01:14:48.991319873+09:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"pebh9ode","id":"1(@)"}
17
+ {"time":"2026-07-15T01:14:49.148475362+09:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"fbqzmp5m","id":"1(@)"}
18
+ {"time":"2026-07-15T01:14:49.154720114+09:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"unz77u4h","id":"1(@)"}
19
+ {"time":"2026-07-15T01:14:54.704195769+09:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"gboffkhx5odh"}
20
+ {"time":"2026-07-15T01:14:54.755700856+09:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ktloe72dm2ag"}
21
+ {"time":"2026-07-15T01:14:54.84707997+09:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"gveiruljishv"}
22
+ {"time":"2026-07-15T10:20:55.252683572+09:00","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
23
+ {"time":"2026-07-15T10:20:55.252742779+09:00","level":"INFO","msg":"connection: closing","id":"1(@)"}
24
+ {"time":"2026-07-15T10:20:55.252786889+09:00","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
25
+ {"time":"2026-07-15T10:20:55.25278977+09:00","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
26
+ {"time":"2026-07-15T10:20:56.037707662+09:00","level":"INFO","msg":"server: parent process exited, terminating service process"}
RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/run-20260715_011448-pebh9ode/logs/debug.log ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_setup.py:_flush():81] Current SDK version is 0.27.2
2
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_setup.py:_flush():81] Configure stats pid to 2695434
3
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_setup.py:_flush():81] Loading settings from environment variables
4
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_init.py:setup_run_log_directory():723] Logging user logs to /home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/run-20260715_011448-pebh9ode/logs/debug.log
5
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to /home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_6K/wandb/run-20260715_011448-pebh9ode/logs/debug-internal.log
6
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_init.py:init():766] calling init triggers
7
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_init.py:init():771] wandb.init called with sweep_config: {}
8
+ config: {'dataset': {'repo_id': 'Whalswp/INSIGHTfixposV4_filtered_multispace_v2', 'root': '/home/ext_minje/INSIGHTfixposV4_filtered_multispace_v2', 'episodes': None, 'image_transforms': {'enable': False, 'max_num_transforms': 3, 'random_order': False, 'tfs': {'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}, 'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'affine': {'weight': 1.0, 'type': 'RandomAffine', 'kwargs': {'degrees': [-5.0, 5.0], 'translate': [0.05, 0.05]}}}}, 'revision': None, 'use_imagenet_stats': True, 'video_backend': 'torchcodec', 'return_uint8': False, 'depth_output_unit': 'mm', 'streaming': False, 'eval_split': 0.0}, 'env': None, 'policy': {'type': 'groot_rkd', 'n_obs_steps': 1, 'input_features': {}, 'output_features': {}, 'device': 'cuda', 'use_amp': False, 'use_peft': False, 'push_to_hub': False, 'repo_id': None, 'private': None, 'tags': None, 'license': None, 'pretrained_path': None, 'pretrained_revision': None, 'chunk_size': 16, 'n_action_steps': 16, 'max_state_dim': 132, 'max_action_dim': 132, 'normalization_mapping': {'VISUAL': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'STATE': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'ACTION': <NormalizationMode.IDENTITY: 'IDENTITY'>}, 'base_model_path': 'nvidia/GR00T-N1.7-3B', 'action_decode_transform': None, 'embodiment_tag': 'new_embodiment', 'tune_llm': False, 'tune_visual': False, 'tune_projector': True, 'tune_diffusion_model': True, 'tune_vlln': True, 'tune_top_llm_layers': 0, 'num_inference_timesteps': None, 'rtc_ramp_rate': None, 'use_flash_attention': False, 'use_relative_actions': False, 'relative_exclude_joints': [], 'optimizer_lr': 0.0001, 'optimizer_betas': [0.9, 0.999], 'optimizer_eps': 1e-08, 'optimizer_weight_decay': 1e-05, 'warmup_steps': 6000, 'warmup_ratio': 0.05, 'use_bf16': True, 'model_params_fp32': True, 'image_size': [256, 256], 'tokenizer_assets_repo': None, 'lora_rank': 0, 'lora_alpha': 16, 'lora_dropout': 0.1, 'lora_full_model': False, 'video_backend': 'decord', 'balance_dataset_weights': True, 'balance_trajectory_weights': True, 'dataset_paths': None, 'output_dir': './tmp/gr00t', 'save_steps': 1000, 'max_steps': 10000, 'batch_size': 32, 'dataloader_num_workers': 8, 'report_to': 'wandb', 'resume': False, 'rkd_enabled': True, 'rkd_vae_checkpoint': '/home/ext_minje/groot_insight/TimewarpVAE_combo_top8_30k/C8_01_C3_008_beta_5em05_latent_dim_16_granularity_4_decoder_hidden_dim_512/checkpoint-step-30000-epoch-39.11', 'rkd_student_pooling': 'attention', 'rkd_student_attention_hidden_dim': 512, 'rkd_student_attention_output_dim': 512, 'rkd_fm_loss_weight': 1.0, 'rkd_loss_weight': 1.0, 'rkd_distance_weight': 1.0, 'rkd_angle_weight': 2.0}, 'reward_model': None, 'output_dir': '/home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/LAPstyle_linear_6K', 'job_name': 'INSIGHT_6D_RKD_TimewarpVAE_LAPstyle_6K', 'resume': False, 'seed': 42, 'cudnn_deterministic': False, 'num_workers': 4, 'batch_size': 64, 'prefetch_factor': 4, 'persistent_workers': True, 'steps': 60000, 'env_eval_freq': 20000, 'log_freq': 200, 'eval_steps': 0, 'max_eval_samples': 0, 'tolerance_s': 0.0001, 'save_checkpoint': True, 'save_freq': 20000, 'use_policy_training_preset': True, 'optimizer': {'type': 'adamw', 'lr': 0.0001, 'weight_decay': 1e-05, 'grad_clip_norm': 1.0, 'betas': [0.9, 0.999], 'eps': 1e-08}, 'scheduler': {'type': 'diffuser', 'num_warmup_steps': 6000, 'name': 'cosine'}, 'eval': {'n_episodes': 50, 'batch_size': 50, 'use_async_envs': True, 'recording': False, 'recording_repo_id': None, 'recording_private': False}, 'wandb': {'enable': True, 'disable_artifact': True, 'project': 'lerobot', 'entity': None, 'notes': None, 'run_id': None, 'mode': None, 'add_tags': True}, 'peft': None, 'job': {'target': None, 'image': 'huggingface/lerobot-gpu:latest', 'timeout': '2d', 'detach': False, 'tags': []}, 'save_checkpoint_to_hub': False, 'sample_weighting': None, 'rename_map': {}, 'checkpoint_path': None, '_wandb': {}}
9
+ 2026-07-15 01:14:48,266 INFO MainThread:2695434 [wandb_init.py:init():814] starting backend
10
+ 2026-07-15 01:14:48,485 INFO MainThread:2695434 [wandb_init.py:init():829] sending inform_init request
11
+ 2026-07-15 01:14:48,992 INFO MainThread:2695434 [wandb_init.py:init():834] backend started and connected
12
+ 2026-07-15 01:14:48,995 INFO MainThread:2695434 [wandb_init.py:init():904] updated telemetry
13
+ 2026-07-15 01:14:49,005 INFO MainThread:2695434 [wandb_init.py:init():927] communicating run to backend with 90.0 second timeout
14
+ 2026-07-15 01:14:49,642 INFO MainThread:2695434 [wandb_init.py:init():972] starting run threads in backend
15
+ 2026-07-15 01:14:49,699 INFO MainThread:2695434 [wandb_run.py:_console_start():2523] atexit reg
16
+ 2026-07-15 01:14:49,699 INFO MainThread:2695434 [wandb_run.py:_redirect():2373] redirect: wrap_raw
17
+ 2026-07-15 01:14:49,699 INFO MainThread:2695434 [wandb_run.py:_redirect():2442] Wrapping output streams.
18
+ 2026-07-15 01:14:49,699 INFO MainThread:2695434 [wandb_run.py:_redirect():2465] Redirects installed.
19
+ 2026-07-15 01:14:49,701 INFO MainThread:2695434 [wandb_init.py:init():1010] run started, returning control to user process
RKD_TimewarpVAE/RSCLstyle_cosine_60K/wandb/debug.log ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2026-07-15 10:21:29,547 INFO MainThread:1576313 [wandb_setup.py:_flush():81] Current SDK version is 0.27.2
2
+ 2026-07-15 10:21:29,548 INFO MainThread:1576313 [wandb_setup.py:_flush():81] Configure stats pid to 1576313
3
+ 2026-07-15 10:21:29,548 INFO MainThread:1576313 [wandb_setup.py:_flush():81] Loading settings from environment variables
4
+ 2026-07-15 10:21:29,548 INFO MainThread:1576313 [wandb_init.py:setup_run_log_directory():723] Logging user logs to /home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/RSCLstyle_cosine_60K/wandb/run-20260715_102129-typqzcxc/logs/debug.log
5
+ 2026-07-15 10:21:29,548 INFO MainThread:1576313 [wandb_init.py:setup_run_log_directory():724] Logging internal logs to /home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/RSCLstyle_cosine_60K/wandb/run-20260715_102129-typqzcxc/logs/debug-internal.log
6
+ 2026-07-15 10:21:29,548 INFO MainThread:1576313 [wandb_init.py:init():766] calling init triggers
7
+ 2026-07-15 10:21:29,548 INFO MainThread:1576313 [wandb_init.py:init():771] wandb.init called with sweep_config: {}
8
+ config: {'dataset': {'repo_id': 'Whalswp/INSIGHTfixposV4_filtered_multispace_v2', 'root': '/home/ext_minje/INSIGHTfixposV4_filtered_multispace_v2', 'episodes': None, 'image_transforms': {'enable': False, 'max_num_transforms': 3, 'random_order': False, 'tfs': {'brightness': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'brightness': [0.8, 1.2]}}, 'contrast': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'contrast': [0.8, 1.2]}}, 'saturation': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'saturation': [0.5, 1.5]}}, 'hue': {'weight': 1.0, 'type': 'ColorJitter', 'kwargs': {'hue': [-0.05, 0.05]}}, 'sharpness': {'weight': 1.0, 'type': 'SharpnessJitter', 'kwargs': {'sharpness': [0.5, 1.5]}}, 'affine': {'weight': 1.0, 'type': 'RandomAffine', 'kwargs': {'degrees': [-5.0, 5.0], 'translate': [0.05, 0.05]}}}}, 'revision': None, 'use_imagenet_stats': True, 'video_backend': 'torchcodec', 'return_uint8': False, 'depth_output_unit': 'mm', 'streaming': False, 'eval_split': 0.0}, 'env': None, 'policy': {'type': 'groot_rkd', 'n_obs_steps': 1, 'input_features': {}, 'output_features': {}, 'device': 'cuda', 'use_amp': False, 'use_peft': False, 'push_to_hub': False, 'repo_id': None, 'private': None, 'tags': None, 'license': None, 'pretrained_path': None, 'pretrained_revision': None, 'chunk_size': 16, 'n_action_steps': 16, 'max_state_dim': 132, 'max_action_dim': 132, 'normalization_mapping': {'VISUAL': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'STATE': <NormalizationMode.IDENTITY: 'IDENTITY'>, 'ACTION': <NormalizationMode.IDENTITY: 'IDENTITY'>}, 'base_model_path': 'nvidia/GR00T-N1.7-3B', 'action_decode_transform': None, 'embodiment_tag': 'new_embodiment', 'tune_llm': False, 'tune_visual': False, 'tune_projector': True, 'tune_diffusion_model': True, 'tune_vlln': True, 'tune_top_llm_layers': 0, 'num_inference_timesteps': None, 'rtc_ramp_rate': None, 'use_flash_attention': False, 'use_relative_actions': False, 'relative_exclude_joints': [], 'optimizer_lr': 0.0001, 'optimizer_betas': [0.9, 0.999], 'optimizer_eps': 1e-08, 'optimizer_weight_decay': 1e-05, 'warmup_steps': 3000, 'warmup_ratio': 0.05, 'use_bf16': True, 'model_params_fp32': True, 'image_size': [256, 256], 'tokenizer_assets_repo': None, 'lora_rank': 0, 'lora_alpha': 16, 'lora_dropout': 0.1, 'lora_full_model': False, 'video_backend': 'decord', 'balance_dataset_weights': True, 'balance_trajectory_weights': True, 'dataset_paths': None, 'output_dir': './tmp/gr00t', 'save_steps': 1000, 'max_steps': 10000, 'batch_size': 32, 'dataloader_num_workers': 8, 'report_to': 'wandb', 'resume': False, 'rkd_enabled': True, 'rkd_vae_checkpoint': '/home/ext_minje/groot_insight/TimewarpVAE_combo_top8_30k/C8_01_C3_008_beta_5em05_latent_dim_16_granularity_4_decoder_hidden_dim_512/checkpoint-step-30000-epoch-39.11', 'rkd_student_pooling': 'attention', 'rkd_student_attention_hidden_dim': 512, 'rkd_student_attention_output_dim': 512, 'rkd_fm_loss_weight': 1.0, 'rkd_loss_weight': 1.0, 'rkd_distance_weight': 1.0, 'rkd_angle_weight': 2.0}, 'reward_model': None, 'output_dir': '/home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/RSCLstyle_cosine_60K', 'job_name': 'INSIGHT_6D_RKD_TimewarpVAE_RSCLstyle_C60', 'resume': False, 'seed': 42, 'cudnn_deterministic': False, 'num_workers': 4, 'batch_size': 64, 'prefetch_factor': 4, 'persistent_workers': True, 'steps': 60000, 'env_eval_freq': 20000, 'log_freq': 200, 'eval_steps': 0, 'max_eval_samples': 0, 'tolerance_s': 0.0001, 'save_checkpoint': True, 'save_freq': 20000, 'use_policy_training_preset': True, 'optimizer': {'type': 'adamw', 'lr': 0.0001, 'weight_decay': 1e-05, 'grad_clip_norm': 1.0, 'betas': [0.9, 0.999], 'eps': 1e-08}, 'scheduler': {'type': 'diffuser', 'num_warmup_steps': 3000, 'name': 'cosine'}, 'eval': {'n_episodes': 50, 'batch_size': 50, 'use_async_envs': True, 'recording': False, 'recording_repo_id': None, 'recording_private': False}, 'wandb': {'enable': True, 'disable_artifact': True, 'project': 'lerobot', 'entity': None, 'notes': None, 'run_id': None, 'mode': None, 'add_tags': True}, 'peft': None, 'job': {'target': None, 'image': 'huggingface/lerobot-gpu:latest', 'timeout': '2d', 'detach': False, 'tags': []}, 'save_checkpoint_to_hub': False, 'sample_weighting': None, 'rename_map': {}, 'checkpoint_path': None, '_wandb': {}}
9
+ 2026-07-15 10:21:29,548 INFO MainThread:1576313 [wandb_init.py:init():814] starting backend
10
+ 2026-07-15 10:21:29,765 INFO MainThread:1576313 [wandb_init.py:init():829] sending inform_init request
11
+ 2026-07-15 10:21:30,340 INFO MainThread:1576313 [wandb_init.py:init():834] backend started and connected
12
+ 2026-07-15 10:21:30,342 INFO MainThread:1576313 [wandb_init.py:init():904] updated telemetry
13
+ 2026-07-15 10:21:30,346 INFO MainThread:1576313 [wandb_init.py:init():927] communicating run to backend with 90.0 second timeout
14
+ 2026-07-15 10:21:30,706 INFO MainThread:1576313 [wandb_init.py:init():972] starting run threads in backend
15
+ 2026-07-15 10:21:30,760 INFO MainThread:1576313 [wandb_run.py:_console_start():2523] atexit reg
16
+ 2026-07-15 10:21:30,760 INFO MainThread:1576313 [wandb_run.py:_redirect():2373] redirect: wrap_raw
17
+ 2026-07-15 10:21:30,760 INFO MainThread:1576313 [wandb_run.py:_redirect():2442] Wrapping output streams.
18
+ 2026-07-15 10:21:30,760 INFO MainThread:1576313 [wandb_run.py:_redirect():2465] Redirects installed.
19
+ 2026-07-15 10:21:30,761 INFO MainThread:1576313 [wandb_init.py:init():1010] run started, returning control to user process
RKD_TimewarpVAE/RSCLstyle_cosine_60K/wandb/run-20260715_102129-typqzcxc/files/requirements.txt ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ pyarrow==25.0.0
2
+ importlib_metadata==9.0.0
3
+ wheel==0.47.0
4
+ cloudpickle==3.1.2
5
+ nvidia-cusolver-cu12==11.7.3.90
6
+ multiprocess==0.70.19
7
+ torch==2.11.0+cu128
8
+ nvidia-nvjitlink-cu12==12.8.93
9
+ fsspec==2026.2.0
10
+ pytz==2026.2
11
+ hf-xet==1.5.1
12
+ tokenizers==0.22.2
13
+ opencv-python-headless==4.13.0.92
14
+ cuda-pathfinder==1.2.2
15
+ frozenlist==1.8.0
16
+ smmap==5.0.3
17
+ httpcore==1.0.9
18
+ nvidia-nvshmem-cu12==3.4.5
19
+ nvidia-cuda-nvrtc-cu12==12.8.93
20
+ annotated-types==0.7.0
21
+ platformdirs==4.10.0
22
+ click==8.4.2
23
+ pyyaml-include==1.4.1
24
+ MarkupSafe==3.0.3
25
+ Jinja2==3.1.6
26
+ urllib3==2.7.0
27
+ nvidia-curand-cu12==10.3.9.90
28
+ certifi==2026.6.17
29
+ nvidia-cuda-runtime-cu12==12.8.90
30
+ torchcodec==0.11.1
31
+ toml==0.10.2
32
+ av==15.1.0
33
+ nvidia-cufft-cu12==11.3.3.83
34
+ attrs==26.1.0
35
+ lerobot==0.6.0
36
+ PyYAML==6.0.3
37
+ safetensors==0.8.0
38
+ diffusers==0.35.2
39
+ requests==2.34.2
40
+ nvidia-nvtx-cu12==12.8.90
41
+ charset-normalizer==3.4.9
42
+ networkx==3.6.1
43
+ cmake==4.1.3
44
+ pydantic_core==2.46.4
45
+ timm==1.0.28
46
+ dm-tree==0.1.10
47
+ nvidia-cufile-cu12==1.13.1.3
48
+ termcolor==3.3.0
49
+ multidict==6.7.1
50
+ filelock==3.29.0
51
+ tzdata==2026.3
52
+ aiohappyeyeballs==2.7.1
53
+ httpx==0.28.1
54
+ nvidia-nccl-cu12==2.28.9
55
+ Pygments==2.20.0
56
+ einops==0.8.2
57
+ nvidia-cusparse-cu12==12.5.8.93
58
+ torchvision==0.26.0+cu128
59
+ dill==0.4.1
60
+ GitPython==3.1.51
61
+ nvidia-cuda-cupti-cu12==12.8.90
62
+ decord==0.6.0
63
+ sympy==1.14.0
64
+ cuda-toolkit==12.8.1
65
+ python-dateutil==2.9.0.post0
66
+ shellingham==1.5.4
67
+ propcache==0.5.2
68
+ Farama-Notifications==0.0.6
69
+ pillow==12.2.0
70
+ cuda-bindings==12.9.4
71
+ huggingface_hub==1.23.0
72
+ pandas==2.3.3
73
+ typing-inspect==0.9.0
74
+ packaging==25.0
75
+ rich==15.0.0
76
+ nvidia-cusparselt-cu12==0.7.1
77
+ setuptools==78.1.0
78
+ wrapt==2.2.2
79
+ mypy_extensions==1.1.0
80
+ tqdm==4.68.4
81
+ draccus==0.10.0
82
+ aiosignal==1.4.0
83
+ mpmath==1.3.0
84
+ numpy==2.2.6
85
+ typing_extensions==4.15.0
86
+ psutil==7.2.2
87
+ mdurl==0.1.2
88
+ nvidia-cublas-cu12==12.8.4.1
89
+ absl-py==2.5.0
90
+ gitdb==4.0.12
91
+ transformers==5.5.4
92
+ accelerate==1.14.0
93
+ pydantic==2.13.4
94
+ pip==26.1.2
95
+ markdown-it-py==4.2.0
96
+ typer==0.26.8
97
+ triton==3.6.0
98
+ wandb==0.27.2
99
+ regex==2026.7.10
100
+ yarl==1.24.2
101
+ jsonlines==4.0.0
102
+ mergedeep==1.3.4
103
+ aiohttp==3.14.1
104
+ datasets==4.8.5
105
+ gymnasium==1.3.0
106
+ peft==0.19.1
107
+ zipp==4.1.0
108
+ idna==3.18
109
+ nvidia-cudnn-cu12==9.19.0.56
110
+ h11==0.16.0
111
+ sentry-sdk==2.65.0
112
+ annotated-doc==0.0.4
113
+ protobuf==7.35.1
114
+ typing-inspection==0.4.2
115
+ anyio==4.14.2
116
+ xxhash==3.8.1
117
+ six==1.17.0
RKD_TimewarpVAE/RSCLstyle_cosine_60K/wandb/run-20260715_102129-typqzcxc/files/wandb-metadata.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "os": "Linux-6.8.0-107-generic-x86_64-with-glibc2.39",
3
+ "python": "CPython 3.12.13",
4
+ "startedAt": "2026-07-15T01:21:29.547181Z",
5
+ "args": [
6
+ "--config_path=/tmp/insight_lerobot_cfg_svz5bb41/RSCLstyle_RKD_cosine_decay_60K.yaml"
7
+ ],
8
+ "program": "/home/ext_minje/clvla/benchmarks/INSIGHT/my_scripts/lerobot_train_yaml.py",
9
+ "codePath": "benchmarks/INSIGHT/my_scripts/lerobot_train_yaml.py",
10
+ "codePathLocal": "my_scripts/lerobot_train_yaml.py",
11
+ "git": {
12
+ "remote": "git@github.com:paragon7060/clvla.git",
13
+ "commit": "22b2faefeaab966ca2dc3688fe6a77f3bcd8de32"
14
+ },
15
+ "root": "/home/ext_minje/groot_insight/Abs_6D/RKD_TimewarpVAE/RSCLstyle_cosine_60K",
16
+ "host": "DGX-H200-01",
17
+ "executable": "/home/ext_minje/miniforge3/envs/lerobot060_groot/bin/python",
18
+ "cpu_count": 112,
19
+ "cpu_count_logical": 224,
20
+ "gpu": "NVIDIA H200",
21
+ "gpu_count": 8,
22
+ "disk": {
23
+ "/": {
24
+ "total": "1888425144320",
25
+ "used": "1265248227328"
26
+ }
27
+ },
28
+ "memory": {
29
+ "total": "2164177248256"
30
+ },
31
+ "gpu_nvidia": [
32
+ {
33
+ "name": "NVIDIA H200",
34
+ "memoryTotal": "150754820096",
35
+ "cudaCores": 16896,
36
+ "architecture": "Hopper",
37
+ "uuid": "GPU-1e531a45-480e-3c1a-e9a0-2ba2cc27b0a3"
38
+ },
39
+ {
40
+ "name": "NVIDIA H200",
41
+ "memoryTotal": "150754820096",
42
+ "cudaCores": 16896,
43
+ "architecture": "Hopper",
44
+ "uuid": "GPU-5d3d5ae2-ffde-89d9-8818-dc19e11ad2fb"
45
+ },
46
+ {
47
+ "name": "NVIDIA H200",
48
+ "memoryTotal": "150754820096",
49
+ "cudaCores": 16896,
50
+ "architecture": "Hopper",
51
+ "uuid": "GPU-305c0dae-dcf5-d7bb-7dd3-5477d2194275"
52
+ },
53
+ {
54
+ "name": "NVIDIA H200",
55
+ "memoryTotal": "150754820096",
56
+ "cudaCores": 16896,
57
+ "architecture": "Hopper",
58
+ "uuid": "GPU-ed8b7920-8881-4e8f-3e88-be1bbecafa21"
59
+ },
60
+ {
61
+ "name": "NVIDIA H200",
62
+ "memoryTotal": "150754820096",
63
+ "cudaCores": 16896,
64
+ "architecture": "Hopper",
65
+ "uuid": "GPU-a91176db-55f2-b284-bbbf-56626596947d"
66
+ },
67
+ {
68
+ "name": "NVIDIA H200",
69
+ "memoryTotal": "150754820096",
70
+ "cudaCores": 16896,
71
+ "architecture": "Hopper",
72
+ "uuid": "GPU-922ff6d5-a948-205a-0426-dbee495fc3dd"
73
+ },
74
+ {
75
+ "name": "NVIDIA H200",
76
+ "memoryTotal": "150754820096",
77
+ "cudaCores": 16896,
78
+ "architecture": "Hopper",
79
+ "uuid": "GPU-1fde7c75-18fe-e3db-a4c1-85e67522f55f"
80
+ },
81
+ {
82
+ "name": "NVIDIA H200",
83
+ "memoryTotal": "150754820096",
84
+ "cudaCores": 16896,
85
+ "architecture": "Hopper",
86
+ "uuid": "GPU-a9b1c49b-d527-97d8-ce91-7e50bd02154c"
87
+ }
88
+ ],
89
+ "cudaVersion": "12.9",
90
+ "writerId": "10yarnxsl8z709it3esrx3bciyi8ka8b"
91
+ }
RKD_TimewarpVAE/RSCLstyle_cosine_60K/wandb/run-20260715_102129-typqzcxc/logs/debug-core.log ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {"time":"2026-07-15T10:21:29.588215737+09:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpopreo720/port-1576313.txt","pid":1576313,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
2
+ {"time":"2026-07-15T10:21:29.589043303+09:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-1576313-1581374-3157815362/socket","Net":"unix"}}
3
+ {"time":"2026-07-15T10:21:29.589550871+09:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":1576313}
4
+ {"time":"2026-07-15T10:21:29.765420504+09:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
5
+ {"time":"2026-07-15T10:21:29.771239791+09:00","level":"INFO","msg":"handleInformInit: received","streamId":"typqzcxc","id":"1(@)"}
6
+ {"time":"2026-07-15T10:21:30.339371442+09:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"typqzcxc","id":"1(@)"}
7
+ {"time":"2026-07-15T10:21:35.764649743+09:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ag5rdmsanx2o"}