Amshaker commited on
Commit
595b60f
·
verified ·
1 Parent(s): 504195f

Upload folder using huggingface_hub (part 4)

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000065_test__val_savanna__2__000016/metrics.json +80 -0
  2. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000066_test__val_iceplains__1__000027/frame_metrics.npz +3 -0
  3. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000066_test__val_iceplains__1__000027/metrics.json +80 -0
  4. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000067_test__val_iceplains__0__000048/frame_metrics.npz +3 -0
  5. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000067_test__val_iceplains__0__000048/metrics.json +80 -0
  6. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000068_test__val_plains__1__000034/frame_metrics.npz +3 -0
  7. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000068_test__val_plains__1__000034/metrics.json +80 -0
  8. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000069_test__val_desert__1__000004/frame_metrics.npz +3 -0
  9. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000069_test__val_desert__1__000004/metrics.json +80 -0
  10. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000070_test__val_desert__2__000001/frame_metrics.npz +3 -0
  11. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000070_test__val_desert__2__000001/metrics.json +80 -0
  12. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000071_test__val_iceplains__0__000005/frame_metrics.npz +3 -0
  13. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000071_test__val_iceplains__0__000005/metrics.json +80 -0
  14. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000072_test__val_iceplains__3__000006/frame_metrics.npz +3 -0
  15. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000072_test__val_iceplains__3__000006/metrics.json +80 -0
  16. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000073_test__val_savanna__0__000015/frame_metrics.npz +3 -0
  17. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000073_test__val_savanna__0__000015/metrics.json +80 -0
  18. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000074_test__val_iceplains__2__000036/frame_metrics.npz +3 -0
  19. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000074_test__val_iceplains__2__000036/metrics.json +80 -0
  20. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000075_test__val_desert__0__000036/frame_metrics.npz +3 -0
  21. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000075_test__val_desert__0__000036/metrics.json +80 -0
  22. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000076_test__val_iceplains__0__000013/frame_metrics.npz +3 -0
  23. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000076_test__val_iceplains__0__000013/metrics.json +80 -0
  24. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000077_test__val_desert__1__000010/frame_metrics.npz +3 -0
  25. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000077_test__val_desert__1__000010/metrics.json +80 -0
  26. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000078_test__val_desert__0__000029/frame_metrics.npz +3 -0
  27. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000078_test__val_desert__0__000029/metrics.json +80 -0
  28. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000079_test__val_iceplains__1__000019/frame_metrics.npz +3 -0
  29. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000079_test__val_iceplains__1__000019/metrics.json +80 -0
  30. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000080_test__val_desert__1__000014/frame_metrics.npz +3 -0
  31. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000080_test__val_desert__1__000014/metrics.json +80 -0
  32. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000081_test__val_savanna__1__000029/frame_metrics.npz +3 -0
  33. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000081_test__val_savanna__1__000029/metrics.json +80 -0
  34. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000082_test__val_savanna__0__000017/frame_metrics.npz +3 -0
  35. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000082_test__val_savanna__0__000017/metrics.json +80 -0
  36. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000083_test__val_plains__2__000017/frame_metrics.npz +3 -0
  37. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000083_test__val_plains__2__000017/metrics.json +80 -0
  38. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000084_test__val_savanna__0__000006/frame_metrics.npz +3 -0
  39. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000084_test__val_savanna__0__000006/metrics.json +80 -0
  40. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000085_test__val_savanna__1__000002/frame_metrics.npz +3 -0
  41. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000085_test__val_savanna__1__000002/metrics.json +80 -0
  42. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000086_test__val_iceplains__2__000039/frame_metrics.npz +3 -0
  43. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000086_test__val_iceplains__2__000039/metrics.json +80 -0
  44. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000087_test__val_savanna__0__000038/frame_metrics.npz +3 -0
  45. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000087_test__val_savanna__0__000038/metrics.json +80 -0
  46. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000088_test__val_plains__1__000013/frame_metrics.npz +3 -0
  47. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000088_test__val_plains__1__000013/metrics.json +80 -0
  48. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000089_test__val_savanna__2__000038/frame_metrics.npz +3 -0
  49. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000089_test__val_savanna__2__000038/metrics.json +80 -0
  50. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000090_test__val_plains__2__000044/frame_metrics.npz +3 -0
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000065_test__val_savanna__2__000016/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "65",
4
+ "global_test_index": "65",
5
+ "dataset_index": "169",
6
+ "relative_path": "test/val_savanna/2/000016.mp4",
7
+ "clip_id": "test/val_savanna/2/000016",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000065,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_savanna/2/000016_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.004014372825622559,
51
+ "psnr": 24.73114585876465,
52
+ "lpips": 0.21281827986240387
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.007330615073442459,
56
+ "psnr": 21.654787063598633,
57
+ "lpips": 0.33493173122406006
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.04523978382349014,
61
+ "psnr": 14.69363784790039,
62
+ "lpips": 0.49253830313682556
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.0367109440267086,
66
+ "psnr": 15.020256042480469,
67
+ "lpips": 0.4842808246612549
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.031806234270334244,
71
+ "psnr": 15.92253303527832,
72
+ "lpips": 0.5028233528137207
73
+ },
74
+ "overall": {
75
+ "mse": 0.025020388886332512,
76
+ "psnr": 18.404470443725586,
77
+ "lpips": 0.40547850728034973
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000066_test__val_iceplains__1__000027/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c84f5bec3385c04db85d5256e87fd559bf410e518269283bdd783a404e0da80b
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000066_test__val_iceplains__1__000027/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "66",
4
+ "global_test_index": "66",
5
+ "dataset_index": "312",
6
+ "relative_path": "test/val_iceplains/1/000027.mp4",
7
+ "clip_id": "test/val_iceplains/1/000027",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000066,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_iceplains/1/000027_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.03341749310493469,
51
+ "psnr": 15.692256927490234,
52
+ "lpips": 0.20027373731136322
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.045889515429735184,
56
+ "psnr": 13.689079284667969,
57
+ "lpips": 0.26659515500068665
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.03775228187441826,
61
+ "psnr": 14.451379776000977,
62
+ "lpips": 0.23427170515060425
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.03829583153128624,
66
+ "psnr": 14.485011100769043,
67
+ "lpips": 0.24214956164360046
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.023182572796940804,
71
+ "psnr": 16.42613983154297,
72
+ "lpips": 0.17161248624324799
73
+ },
74
+ "overall": {
75
+ "mse": 0.035707537084817886,
76
+ "psnr": 14.948773384094238,
77
+ "lpips": 0.2229805290699005
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000067_test__val_iceplains__0__000048/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d92d41fa3aa7cdfb3958038ff18a2ffd856386b902679ba7651b8250dd080425
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000067_test__val_iceplains__0__000048/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "67",
4
+ "global_test_index": "67",
5
+ "dataset_index": "368",
6
+ "relative_path": "test/val_iceplains/0/000048.mp4",
7
+ "clip_id": "test/val_iceplains/0/000048",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000067,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_iceplains/0/000048_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.018511664122343063,
51
+ "psnr": 17.558216094970703,
52
+ "lpips": 0.1909932643175125
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.018761316314339638,
56
+ "psnr": 17.441848754882812,
57
+ "lpips": 0.230560302734375
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.018833298236131668,
61
+ "psnr": 17.360692977905273,
62
+ "lpips": 0.18460948765277863
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.02397480420768261,
66
+ "psnr": 16.524255752563477,
67
+ "lpips": 0.20898346602916718
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.029439635574817657,
71
+ "psnr": 15.517582893371582,
72
+ "lpips": 0.26548025012016296
73
+ },
74
+ "overall": {
75
+ "mse": 0.021904144436120987,
76
+ "psnr": 16.88051986694336,
77
+ "lpips": 0.21612533926963806
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000068_test__val_plains__1__000034/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad72b4939888ed9979fd7a0816cd7d093270ca200e6b13189d17b98c4314c38d
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000068_test__val_plains__1__000034/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "68",
4
+ "global_test_index": "68",
5
+ "dataset_index": "449",
6
+ "relative_path": "test/val_plains/1/000034.mp4",
7
+ "clip_id": "test/val_plains/1/000034",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000068,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_plains/1/000034_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.02620217576622963,
51
+ "psnr": 18.96230125427246,
52
+ "lpips": 0.1933584213256836
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.0367947593331337,
56
+ "psnr": 15.132673263549805,
57
+ "lpips": 0.3913692533969879
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.020278511568903923,
61
+ "psnr": 17.276405334472656,
62
+ "lpips": 0.4713697135448456
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.011657017283141613,
66
+ "psnr": 19.586938858032227,
67
+ "lpips": 0.4975817799568176
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.01569548435509205,
71
+ "psnr": 18.3051815032959,
72
+ "lpips": 0.5181838870048523
73
+ },
74
+ "overall": {
75
+ "mse": 0.022125588729977608,
76
+ "psnr": 17.852699279785156,
77
+ "lpips": 0.41437259316444397
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000069_test__val_desert__1__000004/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f39fc5bd68c49adfc7156573e6253ca76525ef88972574a4144af8230f84a387
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000069_test__val_desert__1__000004/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "69",
4
+ "global_test_index": "69",
5
+ "dataset_index": "11",
6
+ "relative_path": "test/val_desert/1/000004.mp4",
7
+ "clip_id": "test/val_desert/1/000004",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000069,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_desert/1/000004_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.007419729605317116,
51
+ "psnr": 21.3951473236084,
52
+ "lpips": 0.23921996355056763
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.008625335060060024,
56
+ "psnr": 20.78408432006836,
57
+ "lpips": 0.367950439453125
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.016007713973522186,
61
+ "psnr": 18.248327255249023,
62
+ "lpips": 0.4495605230331421
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.03778526186943054,
66
+ "psnr": 15.024605751037598,
67
+ "lpips": 0.4829889237880707
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.04321294650435448,
71
+ "psnr": 14.007286071777344,
72
+ "lpips": 0.5312978029251099
73
+ },
74
+ "overall": {
75
+ "mse": 0.022610198706388474,
76
+ "psnr": 17.891891479492188,
77
+ "lpips": 0.4142035245895386
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000070_test__val_desert__2__000001/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d24e4c1cb7b736d5fe000494af0a9bbd1c77002798d2064803bd70cc00fe7f43
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000070_test__val_desert__2__000001/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "70",
4
+ "global_test_index": "70",
5
+ "dataset_index": "6",
6
+ "relative_path": "test/val_desert/2/000001.mp4",
7
+ "clip_id": "test/val_desert/2/000001",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000070,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_desert/2/000001_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.004076240584254265,
51
+ "psnr": 24.220115661621094,
52
+ "lpips": 0.06879425048828125
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.004011362791061401,
56
+ "psnr": 24.842662811279297,
57
+ "lpips": 0.05746771767735481
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.002503311960026622,
61
+ "psnr": 27.600914001464844,
62
+ "lpips": 0.038191843777894974
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.0008896967628970742,
66
+ "psnr": 30.556379318237305,
67
+ "lpips": 0.012969447299838066
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.0015840891283005476,
71
+ "psnr": 30.00088119506836,
72
+ "lpips": 0.015592323616147041
73
+ },
74
+ "overall": {
75
+ "mse": 0.0026129400357604027,
76
+ "psnr": 27.444189071655273,
77
+ "lpips": 0.03860311955213547
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000071_test__val_iceplains__0__000005/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8235982b3e0d07337687746116fc0c897d74267181d0ab7b146b647eb6a0ec38
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000071_test__val_iceplains__0__000005/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "71",
4
+ "global_test_index": "71",
5
+ "dataset_index": "257",
6
+ "relative_path": "test/val_iceplains/0/000005.mp4",
7
+ "clip_id": "test/val_iceplains/0/000005",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000071,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_iceplains/0/000005_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.02564489282667637,
51
+ "psnr": 16.70078468322754,
52
+ "lpips": 0.2122913897037506
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.007004851475358009,
56
+ "psnr": 22.290752410888672,
57
+ "lpips": 0.23672369122505188
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.02094113826751709,
61
+ "psnr": 17.696422576904297,
62
+ "lpips": 0.31616321206092834
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.005959112662822008,
66
+ "psnr": 22.855588912963867,
67
+ "lpips": 0.2691815197467804
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.018363988026976585,
71
+ "psnr": 17.680204391479492,
72
+ "lpips": 0.38818472623825073
73
+ },
74
+ "overall": {
75
+ "mse": 0.0155827971175313,
76
+ "psnr": 19.44474983215332,
77
+ "lpips": 0.28450891375541687
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000072_test__val_iceplains__3__000006/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a831a6f0f3d2dec3d4fe0d7c72f7d9c5eeca344bb4f55c91db949027b35ca6c
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000072_test__val_iceplains__3__000006/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "72",
4
+ "global_test_index": "72",
5
+ "dataset_index": "262",
6
+ "relative_path": "test/val_iceplains/3/000006.mp4",
7
+ "clip_id": "test/val_iceplains/3/000006",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000072,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_iceplains/3/000006_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.004939154256135225,
51
+ "psnr": 24.171674728393555,
52
+ "lpips": 0.15284396708011627
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.018846845254302025,
56
+ "psnr": 17.99581527709961,
57
+ "lpips": 0.1997644305229187
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.029573652893304825,
61
+ "psnr": 16.305177688598633,
62
+ "lpips": 0.17197054624557495
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.01824493706226349,
66
+ "psnr": 18.500707626342773,
67
+ "lpips": 0.1221550852060318
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.046771831810474396,
71
+ "psnr": 13.485700607299805,
72
+ "lpips": 0.32107117772102356
73
+ },
74
+ "overall": {
75
+ "mse": 0.023675285279750824,
76
+ "psnr": 18.091815948486328,
77
+ "lpips": 0.19356104731559753
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000073_test__val_savanna__0__000015/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c8e7edd3fe0a4af2774e4b0c3b65b36a284288a4d445e10f026a8a328c2b476
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000073_test__val_savanna__0__000015/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "73",
4
+ "global_test_index": "73",
5
+ "dataset_index": "164",
6
+ "relative_path": "test/val_savanna/0/000015.mp4",
7
+ "clip_id": "test/val_savanna/0/000015",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000073,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_savanna/0/000015_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.026894740760326385,
51
+ "psnr": 18.513084411621094,
52
+ "lpips": 0.1595374345779419
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.19733141362667084,
56
+ "psnr": 8.006949424743652,
57
+ "lpips": 0.6579327583312988
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.29380542039871216,
61
+ "psnr": 5.346463203430176,
62
+ "lpips": 0.8078112006187439
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.26710739731788635,
66
+ "psnr": 5.820764064788818,
67
+ "lpips": 0.7569387555122375
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.30158257484436035,
71
+ "psnr": 5.236715793609619,
72
+ "lpips": 0.8018047213554382
73
+ },
74
+ "overall": {
75
+ "mse": 0.21734432876110077,
76
+ "psnr": 8.584794998168945,
77
+ "lpips": 0.63680499792099
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000074_test__val_iceplains__2__000036/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98fd882fd142d781c713c2b2130c8bdb0763051b7fa7fce48c03419e14208581
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000074_test__val_iceplains__2__000036/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "74",
4
+ "global_test_index": "74",
5
+ "dataset_index": "337",
6
+ "relative_path": "test/val_iceplains/2/000036.mp4",
7
+ "clip_id": "test/val_iceplains/2/000036",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000074,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_iceplains/2/000036_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.009120816364884377,
51
+ "psnr": 20.750158309936523,
52
+ "lpips": 0.10336937010288239
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.010461344383656979,
56
+ "psnr": 20.395462036132812,
57
+ "lpips": 0.08742618560791016
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.008670277893543243,
61
+ "psnr": 21.52127456665039,
62
+ "lpips": 0.08238144218921661
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.005135504994541407,
66
+ "psnr": 23.335512161254883,
67
+ "lpips": 0.05467966943979263
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.0017478882800787687,
71
+ "psnr": 29.054672241210938,
72
+ "lpips": 0.023778129369020462
73
+ },
74
+ "overall": {
75
+ "mse": 0.007027166895568371,
76
+ "psnr": 23.011415481567383,
77
+ "lpips": 0.07032694667577744
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000075_test__val_desert__0__000036/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0ee32ff9f4a304d0d4e461040acff7646a2ef8710c2039f07d4527570171fc9
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000075_test__val_desert__0__000036/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "75",
4
+ "global_test_index": "75",
5
+ "dataset_index": "95",
6
+ "relative_path": "test/val_desert/0/000036.mp4",
7
+ "clip_id": "test/val_desert/0/000036",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000075,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_desert/0/000036_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.0016190209425985813,
51
+ "psnr": 28.893476486206055,
52
+ "lpips": 0.025067996233701706
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.0007734189857728779,
56
+ "psnr": 31.128684997558594,
57
+ "lpips": 0.009813263081014156
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.0018020159332081676,
61
+ "psnr": 28.51152992248535,
62
+ "lpips": 0.02812228724360466
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.007088941987603903,
66
+ "psnr": 22.067663192749023,
67
+ "lpips": 0.13363191485404968
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.016006767749786377,
71
+ "psnr": 18.305557250976562,
72
+ "lpips": 0.3150196075439453
73
+ },
74
+ "overall": {
75
+ "mse": 0.005458032712340355,
76
+ "psnr": 25.781381607055664,
77
+ "lpips": 0.1023310124874115
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000076_test__val_iceplains__0__000013/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c3760451026a39a51ab2c620eea531de56d9118a87c8728bad3045703a035be
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000076_test__val_iceplains__0__000013/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "76",
4
+ "global_test_index": "76",
5
+ "dataset_index": "277",
6
+ "relative_path": "test/val_iceplains/0/000013.mp4",
7
+ "clip_id": "test/val_iceplains/0/000013",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000076,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_iceplains/0/000013_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.0012489876244217157,
51
+ "psnr": 29.882312774658203,
52
+ "lpips": 0.014558336697518826
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.001941458904184401,
56
+ "psnr": 28.567697525024414,
57
+ "lpips": 0.018520863726735115
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.002243684371933341,
61
+ "psnr": 27.8562068939209,
62
+ "lpips": 0.023359812796115875
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.0007919078343547881,
66
+ "psnr": 31.069229125976562,
67
+ "lpips": 0.011025523766875267
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.012342230416834354,
71
+ "psnr": 21.896635055541992,
72
+ "lpips": 0.12641845643520355
73
+ },
74
+ "overall": {
75
+ "mse": 0.0037136536557227373,
76
+ "psnr": 27.854413986206055,
77
+ "lpips": 0.03877659887075424
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000077_test__val_desert__1__000010/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0cb68aacef5dfa7310d653eaec4f5b29e422b4bfe470b0e163aaa416dabd4dd
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000077_test__val_desert__1__000010/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "77",
4
+ "global_test_index": "77",
5
+ "dataset_index": "26",
6
+ "relative_path": "test/val_desert/1/000010.mp4",
7
+ "clip_id": "test/val_desert/1/000010",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000077,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_desert/1/000010_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.012197958305478096,
51
+ "psnr": 19.62904930114746,
52
+ "lpips": 0.28380635380744934
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.020117277279496193,
56
+ "psnr": 17.162813186645508,
57
+ "lpips": 0.39623939990997314
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.01860169880092144,
61
+ "psnr": 17.54905128479004,
62
+ "lpips": 0.3376738429069519
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.01646866649389267,
66
+ "psnr": 18.835878372192383,
67
+ "lpips": 0.3862321376800537
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.009749620221555233,
71
+ "psnr": 20.240331649780273,
72
+ "lpips": 0.3306339383125305
73
+ },
74
+ "overall": {
75
+ "mse": 0.015427043661475182,
76
+ "psnr": 18.683425903320312,
77
+ "lpips": 0.34691715240478516
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000078_test__val_desert__0__000029/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:abc3ea52727cd8400ea031dd3b2af7fc8d0b056cd14a9c81f557ddc1524fb2bb
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000078_test__val_desert__0__000029/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "78",
4
+ "global_test_index": "78",
5
+ "dataset_index": "77",
6
+ "relative_path": "test/val_desert/0/000029.mp4",
7
+ "clip_id": "test/val_desert/0/000029",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000078,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_desert/0/000029_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.006035820115357637,
51
+ "psnr": 22.439062118530273,
52
+ "lpips": 0.21610312163829803
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.008181639015674591,
56
+ "psnr": 21.663942337036133,
57
+ "lpips": 0.2001114785671234
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.006328573450446129,
61
+ "psnr": 22.289691925048828,
62
+ "lpips": 0.21686047315597534
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.006920985411852598,
66
+ "psnr": 21.679929733276367,
67
+ "lpips": 0.24497073888778687
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.004025949630886316,
71
+ "psnr": 24.53351593017578,
72
+ "lpips": 0.08744464069604874
73
+ },
74
+ "overall": {
75
+ "mse": 0.006298593711107969,
76
+ "psnr": 22.521230697631836,
77
+ "lpips": 0.19309809803962708
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000079_test__val_iceplains__1__000019/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e644fec1bb259a05c6954aa0e643b88b9ed352ccf86d39dbfe022ddc96ec7e99
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000079_test__val_iceplains__1__000019/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "79",
4
+ "global_test_index": "79",
5
+ "dataset_index": "294",
6
+ "relative_path": "test/val_iceplains/1/000019.mp4",
7
+ "clip_id": "test/val_iceplains/1/000019",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000079,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_iceplains/1/000019_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.07124108076095581,
51
+ "psnr": 13.09792709350586,
52
+ "lpips": 0.16915149986743927
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.07218680530786514,
56
+ "psnr": 13.233342170715332,
57
+ "lpips": 0.21796123683452606
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.005865737795829773,
61
+ "psnr": 23.287294387817383,
62
+ "lpips": 0.04332983121275902
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.008965272456407547,
66
+ "psnr": 20.71026611328125,
67
+ "lpips": 0.04578326269984245
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.010463632643222809,
71
+ "psnr": 19.8395938873291,
72
+ "lpips": 0.04343164339661598
73
+ },
74
+ "overall": {
75
+ "mse": 0.033744506537914276,
76
+ "psnr": 18.0336856842041,
77
+ "lpips": 0.1039314940571785
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000080_test__val_desert__1__000014/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c051b2609d71093d2a5fa0012421d24ffe17a52607913142d7a8c206e278d4ce
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000080_test__val_desert__1__000014/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "80",
4
+ "global_test_index": "80",
5
+ "dataset_index": "34",
6
+ "relative_path": "test/val_desert/1/000014.mp4",
7
+ "clip_id": "test/val_desert/1/000014",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000080,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_desert/1/000014_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.0016106539405882359,
51
+ "psnr": 30.389219284057617,
52
+ "lpips": 0.014143029227852821
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.0008103664149530232,
56
+ "psnr": 30.92504119873047,
57
+ "lpips": 0.01078521367162466
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.0016456968151032925,
61
+ "psnr": 30.239500045776367,
62
+ "lpips": 0.01528274454176426
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.0016629217425361276,
66
+ "psnr": 29.953857421875,
67
+ "lpips": 0.017112256959080696
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.001559185329824686,
71
+ "psnr": 28.758094787597656,
72
+ "lpips": 0.02505512163043022
73
+ },
74
+ "overall": {
75
+ "mse": 0.0014577648835256696,
76
+ "psnr": 30.05314064025879,
77
+ "lpips": 0.016475673764944077
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000081_test__val_savanna__1__000029/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54c6ce387202c82fb19955fa0af563a9e73095b07ab4c55e40f9a0ae88bb2348
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000081_test__val_savanna__1__000029/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "81",
4
+ "global_test_index": "81",
5
+ "dataset_index": "188",
6
+ "relative_path": "test/val_savanna/1/000029.mp4",
7
+ "clip_id": "test/val_savanna/1/000029",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000081,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_savanna/1/000029_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.006841068156063557,
51
+ "psnr": 23.338159561157227,
52
+ "lpips": 0.05330326408147812
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.018118485808372498,
56
+ "psnr": 18.39935302734375,
57
+ "lpips": 0.15198276937007904
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.04170526936650276,
61
+ "psnr": 13.90273666381836,
62
+ "lpips": 0.3556647002696991
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.025220200419425964,
66
+ "psnr": 16.141603469848633,
67
+ "lpips": 0.34059828519821167
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.029699452221393585,
71
+ "psnr": 15.861163139343262,
72
+ "lpips": 0.4500250220298767
73
+ },
74
+ "overall": {
75
+ "mse": 0.024316895753145218,
76
+ "psnr": 17.528600692749023,
77
+ "lpips": 0.2703147828578949
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000082_test__val_savanna__0__000017/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f49e87763e04233137637045c75099bc521e30828577d5c16c3bffedad115728
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000082_test__val_savanna__0__000017/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "82",
4
+ "global_test_index": "82",
5
+ "dataset_index": "170",
6
+ "relative_path": "test/val_savanna/0/000017.mp4",
7
+ "clip_id": "test/val_savanna/0/000017",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000082,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_savanna/0/000017_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.010828632861375809,
51
+ "psnr": 19.821306228637695,
52
+ "lpips": 0.2727340757846832
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.027536902576684952,
56
+ "psnr": 15.764906883239746,
57
+ "lpips": 0.31653282046318054
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.053111132234334946,
61
+ "psnr": 13.122760772705078,
62
+ "lpips": 0.41764411330223083
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.07883300632238388,
66
+ "psnr": 11.294658660888672,
67
+ "lpips": 0.3880220055580139
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.08292018622159958,
71
+ "psnr": 11.375901222229004,
72
+ "lpips": 0.4811946153640747
73
+ },
74
+ "overall": {
75
+ "mse": 0.05064597353339195,
76
+ "psnr": 14.275906562805176,
77
+ "lpips": 0.3752255439758301
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000083_test__val_plains__2__000017/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ea7d2acfbced61dc4adf5ba25af7a01e282a6ec1c580065f507d5a80b314799
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000083_test__val_plains__2__000017/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "83",
4
+ "global_test_index": "83",
5
+ "dataset_index": "415",
6
+ "relative_path": "test/val_plains/2/000017.mp4",
7
+ "clip_id": "test/val_plains/2/000017",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000083,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_plains/2/000017_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.006784230936318636,
51
+ "psnr": 22.10260009765625,
52
+ "lpips": 0.11409496515989304
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.022898998111486435,
56
+ "psnr": 17.466732025146484,
57
+ "lpips": 0.3183107078075409
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.020590664818882942,
61
+ "psnr": 17.65020179748535,
62
+ "lpips": 0.3238554000854492
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.011044168844819069,
66
+ "psnr": 19.74256134033203,
67
+ "lpips": 0.2604333162307739
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.011853259988129139,
71
+ "psnr": 19.429931640625,
72
+ "lpips": 0.33292847871780396
73
+ },
74
+ "overall": {
75
+ "mse": 0.014634265564382076,
76
+ "psnr": 19.278406143188477,
77
+ "lpips": 0.2699245512485504
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000084_test__val_savanna__0__000006/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fba6131250b389bcbcaaa7f1bbb42e0176a40c2584cd6821b33a70d3a69a5116
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000084_test__val_savanna__0__000006/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "84",
4
+ "global_test_index": "84",
5
+ "dataset_index": "145",
6
+ "relative_path": "test/val_savanna/0/000006.mp4",
7
+ "clip_id": "test/val_savanna/0/000006",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000084,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_savanna/0/000006_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.015611572191119194,
51
+ "psnr": 18.780807495117188,
52
+ "lpips": 0.12070924788713455
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.018946601077914238,
56
+ "psnr": 17.455018997192383,
57
+ "lpips": 0.21950136125087738
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.014058049768209457,
61
+ "psnr": 18.930194854736328,
62
+ "lpips": 0.3905223608016968
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.0074511026032269,
66
+ "psnr": 21.52984619140625,
67
+ "lpips": 0.4059504270553589
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.0058480678126215935,
71
+ "psnr": 22.511276245117188,
72
+ "lpips": 0.3924799859523773
73
+ },
74
+ "overall": {
75
+ "mse": 0.01238307822495699,
76
+ "psnr": 19.841426849365234,
77
+ "lpips": 0.3058326840400696
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000085_test__val_savanna__1__000002/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:703432f3ebc7590f8cee8f3598d1286c267deb8ee7cc168f0ff412500bebfa42
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000085_test__val_savanna__1__000002/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "85",
4
+ "global_test_index": "85",
5
+ "dataset_index": "133",
6
+ "relative_path": "test/val_savanna/1/000002.mp4",
7
+ "clip_id": "test/val_savanna/1/000002",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000085,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_savanna/1/000002_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.009289972484111786,
51
+ "psnr": 20.701269149780273,
52
+ "lpips": 0.2666792571544647
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.011189105920493603,
56
+ "psnr": 19.91399383544922,
57
+ "lpips": 0.3061886131763458
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.014358417131006718,
61
+ "psnr": 18.747610092163086,
62
+ "lpips": 0.3554259240627289
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.01757115311920643,
66
+ "psnr": 18.271543502807617,
67
+ "lpips": 0.4052739441394806
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.02507631853222847,
71
+ "psnr": 17.426790237426758,
72
+ "lpips": 0.41056838631629944
73
+ },
74
+ "overall": {
75
+ "mse": 0.015496992506086826,
76
+ "psnr": 19.012239456176758,
77
+ "lpips": 0.34882721304893494
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000086_test__val_iceplains__2__000039/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee2054e038e25639bee95639ee41ff141bffbf8eb5294e94135c7e12930aedf3
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000086_test__val_iceplains__2__000039/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "86",
4
+ "global_test_index": "86",
5
+ "dataset_index": "347",
6
+ "relative_path": "test/val_iceplains/2/000039.mp4",
7
+ "clip_id": "test/val_iceplains/2/000039",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000086,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_iceplains/2/000039_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.017345823347568512,
51
+ "psnr": 17.94886589050293,
52
+ "lpips": 0.1950908601284027
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.014885928481817245,
56
+ "psnr": 19.007240295410156,
57
+ "lpips": 0.2708568871021271
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.024832768365740776,
61
+ "psnr": 17.850805282592773,
62
+ "lpips": 0.34176427125930786
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.08388739824295044,
66
+ "psnr": 11.177172660827637,
67
+ "lpips": 0.44589027762413025
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.08137616515159607,
71
+ "psnr": 11.372044563293457,
72
+ "lpips": 0.4673200845718384
73
+ },
74
+ "overall": {
75
+ "mse": 0.04446560889482498,
76
+ "psnr": 15.471226692199707,
77
+ "lpips": 0.3441844582557678
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000087_test__val_savanna__0__000038/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15a3e538ee0bfa3c7bdbb2c436ed9b540c0b26f83ef27693ae9ed356ee91a362
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000087_test__val_savanna__0__000038/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "87",
4
+ "global_test_index": "87",
5
+ "dataset_index": "209",
6
+ "relative_path": "test/val_savanna/0/000038.mp4",
7
+ "clip_id": "test/val_savanna/0/000038",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000087,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_savanna/0/000038_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.012337503023445606,
51
+ "psnr": 19.54021644592285,
52
+ "lpips": 0.25639790296554565
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.020619649440050125,
56
+ "psnr": 17.437349319458008,
57
+ "lpips": 0.36619749665260315
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.018356572836637497,
61
+ "psnr": 17.68050193786621,
62
+ "lpips": 0.3532315194606781
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.006292724050581455,
66
+ "psnr": 22.170820236206055,
67
+ "lpips": 0.3768339455127716
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.009856903925538063,
71
+ "psnr": 20.439516067504883,
72
+ "lpips": 0.4309653341770172
73
+ },
74
+ "overall": {
75
+ "mse": 0.013492671772837639,
76
+ "psnr": 19.45368194580078,
77
+ "lpips": 0.35672521591186523
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000088_test__val_plains__1__000013/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d340afcc04d0e2bb093f67a0e3ff52ea2894f02200d2bff2dd897c78f69dcc9a
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000088_test__val_plains__1__000013/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "88",
4
+ "global_test_index": "88",
5
+ "dataset_index": "404",
6
+ "relative_path": "test/val_plains/1/000013.mp4",
7
+ "clip_id": "test/val_plains/1/000013",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000088,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_plains/1/000013_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.0014102114364504814,
51
+ "psnr": 28.685462951660156,
52
+ "lpips": 0.12700296938419342
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.0037337630055844784,
56
+ "psnr": 25.541946411132812,
57
+ "lpips": 0.19388142228126526
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.009554834105074406,
61
+ "psnr": 20.841018676757812,
62
+ "lpips": 0.2509021461009979
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.004533831495791674,
66
+ "psnr": 24.2772159576416,
67
+ "lpips": 0.14964593946933746
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.011396617628633976,
71
+ "psnr": 20.10059356689453,
72
+ "lpips": 0.25289323925971985
73
+ },
74
+ "overall": {
75
+ "mse": 0.006125851534307003,
76
+ "psnr": 23.889245986938477,
77
+ "lpips": 0.1948651224374771
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000089_test__val_savanna__2__000038/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98aac07b6e1fb7aae27756d876926781932352f884b44a6df7cd7f778e6e40a3
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000089_test__val_savanna__2__000038/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "89",
4
+ "global_test_index": "89",
5
+ "dataset_index": "211",
6
+ "relative_path": "test/val_savanna/2/000038.mp4",
7
+ "clip_id": "test/val_savanna/2/000038",
8
+ "source_offset": "100",
9
+ "context_start": "100",
10
+ "context_end": "700",
11
+ "target_start": "700",
12
+ "target_end": "1200",
13
+ "n_frames_valid": "1100",
14
+ "context_frames": "600",
15
+ "target_frames": "500",
16
+ "frame_skip": "1",
17
+ "raw_frame_count": "1501",
18
+ "action_count": "1501",
19
+ "pose_count": "1501",
20
+ "latent_count": "1501"
21
+ },
22
+ "sample_seed": 43000089,
23
+ "dataset": "minecraft",
24
+ "frames": 500,
25
+ "native_resolution": [
26
+ 352,
27
+ 640
28
+ ],
29
+ "metric_resolution": [
30
+ 360,
31
+ 640
32
+ ],
33
+ "resize": "bicubic align_corners=False antialias=True",
34
+ "reference_source": "/proj/cvl/users/x_fahkh2/WorldMem_Repro/datasets/minecraft/vae_features/test/val_savanna/2/000038_vae_feature.npy",
35
+ "vae_checkpoint": "/proj/cvl/users/x_fahkh2/worldmem-baseline-evals/checkpoints/worldmem/vae_only.ckpt",
36
+ "torch": "2.8.0+cu128",
37
+ "torchmetrics": "0.11.4",
38
+ "lpips": "alex; normalize=True; mean of individual frame values",
39
+ "psnr_epsilon": 1.1920928955078125e-07,
40
+ "fid": "OpenCV decoding of libx264 quality8 yuv420p uint8 videos",
41
+ "video": {
42
+ "codec": "libx264",
43
+ "quality": 8,
44
+ "pixelformat": "yuv420p",
45
+ "macro_block_size": 1,
46
+ "fps": 15
47
+ },
48
+ "ranges": {
49
+ "frames_000_100": {
50
+ "mse": 0.013012063689529896,
51
+ "psnr": 19.066543579101562,
52
+ "lpips": 0.3802356719970703
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.006951131857931614,
56
+ "psnr": 21.698877334594727,
57
+ "lpips": 0.41791361570358276
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.006270250305533409,
61
+ "psnr": 22.11797523498535,
62
+ "lpips": 0.37911689281463623
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.010513953864574432,
66
+ "psnr": 20.22835922241211,
67
+ "lpips": 0.4178915023803711
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.009852658957242966,
71
+ "psnr": 20.3792667388916,
72
+ "lpips": 0.458982914686203
73
+ },
74
+ "overall": {
75
+ "mse": 0.009320011362433434,
76
+ "psnr": 20.698205947875977,
77
+ "lpips": 0.4108281433582306
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed43_job84976/cases/g000090_test__val_plains__2__000044/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4dda458c94d1191ecda05e2ba70bb4806e957d58c25a13c6c3ff02450b199987
3
+ size 6742