Amshaker commited on
Commit
9ef3ef8
·
verified ·
1 Parent(s): 663b185

Upload folder using huggingface_hub (part 2)

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_seed42_job84976/cases/g000124_test__val_desert__2__000032/metrics.json +80 -0
  2. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000125_test__val_iceplains__0__000044/frame_metrics.npz +3 -0
  3. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000125_test__val_iceplains__0__000044/metrics.json +80 -0
  4. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000126_test__val_desert__1__000020/frame_metrics.npz +3 -0
  5. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000126_test__val_desert__1__000020/metrics.json +80 -0
  6. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000127_test__val_savanna__0__000036/frame_metrics.npz +3 -0
  7. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000127_test__val_savanna__0__000036/metrics.json +80 -0
  8. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000128_test__val_iceplains__2__000034/frame_metrics.npz +3 -0
  9. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000128_test__val_iceplains__2__000034/metrics.json +80 -0
  10. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000129_test__val_savanna__0__000001/frame_metrics.npz +3 -0
  11. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000129_test__val_savanna__0__000001/metrics.json +80 -0
  12. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000130_test__val_plains__3__000021/frame_metrics.npz +3 -0
  13. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000130_test__val_plains__3__000021/metrics.json +80 -0
  14. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000131_test__val_desert__2__000025/frame_metrics.npz +3 -0
  15. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000131_test__val_desert__2__000025/metrics.json +80 -0
  16. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000132_test__val_savanna__3__000009/frame_metrics.npz +3 -0
  17. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000132_test__val_savanna__3__000009/metrics.json +80 -0
  18. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000133_test__val_plains__0__000004/frame_metrics.npz +3 -0
  19. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000133_test__val_plains__0__000004/metrics.json +80 -0
  20. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000134_test__val_savanna__2__000005/frame_metrics.npz +3 -0
  21. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000134_test__val_savanna__2__000005/metrics.json +80 -0
  22. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000135_test__val_iceplains__1__000040/frame_metrics.npz +3 -0
  23. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000135_test__val_iceplains__1__000040/metrics.json +80 -0
  24. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000136_test__val_plains__2__000005/frame_metrics.npz +3 -0
  25. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000136_test__val_plains__2__000005/metrics.json +80 -0
  26. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000137_test__val_desert__0__000012/frame_metrics.npz +3 -0
  27. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000137_test__val_desert__0__000012/metrics.json +80 -0
  28. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000138_test__val_savanna__2__000034/frame_metrics.npz +3 -0
  29. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000138_test__val_savanna__2__000034/metrics.json +80 -0
  30. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000139_test__val_savanna__0__000022/frame_metrics.npz +3 -0
  31. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000139_test__val_savanna__0__000022/metrics.json +80 -0
  32. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000140_test__val_iceplains__2__000003/frame_metrics.npz +3 -0
  33. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000140_test__val_iceplains__2__000003/metrics.json +80 -0
  34. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000141_test__val_iceplains__2__000016/frame_metrics.npz +3 -0
  35. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000141_test__val_iceplains__2__000016/metrics.json +80 -0
  36. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000142_test__val_plains__2__000020/frame_metrics.npz +3 -0
  37. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000142_test__val_plains__2__000020/metrics.json +80 -0
  38. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000143_test__val_desert__0__000008/frame_metrics.npz +3 -0
  39. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000143_test__val_desert__0__000008/metrics.json +80 -0
  40. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000144_test__val_savanna__0__000000/frame_metrics.npz +3 -0
  41. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000144_test__val_savanna__0__000000/metrics.json +80 -0
  42. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000145_test__val_iceplains__1__000020/frame_metrics.npz +3 -0
  43. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000145_test__val_iceplains__1__000020/metrics.json +80 -0
  44. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000146_test__val_iceplains__3__000013/frame_metrics.npz +3 -0
  45. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000146_test__val_iceplains__3__000013/metrics.json +80 -0
  46. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000147_test__val_iceplains__0__000009/frame_metrics.npz +3 -0
  47. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000147_test__val_iceplains__0__000009/metrics.json +80 -0
  48. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000148_test__val_iceplains__0__000028/frame_metrics.npz +3 -0
  49. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000148_test__val_iceplains__0__000028/metrics.json +80 -0
  50. results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000149_test__val_iceplains__0__000014/frame_metrics.npz +3 -0
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000124_test__val_desert__2__000032/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "124",
4
+ "global_test_index": "124",
5
+ "dataset_index": "85",
6
+ "relative_path": "test/val_desert/2/000032.mp4",
7
+ "clip_id": "test/val_desert/2/000032",
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": 42000124,
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/2/000032_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": 0.0038343712221831083,
51
+ "psnr": 24.4262638092041,
52
+ "lpips": 0.1106785237789154
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.008797272108495235,
56
+ "psnr": 20.727378845214844,
57
+ "lpips": 0.18920160830020905
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.008551839739084244,
61
+ "psnr": 20.859403610229492,
62
+ "lpips": 0.15778397023677826
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.011435375548899174,
66
+ "psnr": 19.606718063354492,
67
+ "lpips": 0.21109773218631744
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.0058588432148098946,
71
+ "psnr": 23.32453155517578,
72
+ "lpips": 0.13388223946094513
73
+ },
74
+ "overall": {
75
+ "mse": 0.0076955403201282024,
76
+ "psnr": 21.788862228393555,
77
+ "lpips": 0.16052880883216858
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000125_test__val_iceplains__0__000044/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8a309b97fc4b1451a00268fa5b5dbffba92ecfbfecc82630fc7a4fbceaec712
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000125_test__val_iceplains__0__000044/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "125",
4
+ "global_test_index": "125",
5
+ "dataset_index": "358",
6
+ "relative_path": "test/val_iceplains/0/000044.mp4",
7
+ "clip_id": "test/val_iceplains/0/000044",
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": 42000125,
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/0/000044_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": 0.0009331777109764516,
51
+ "psnr": 30.3817195892334,
52
+ "lpips": 0.011008551344275475
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.000886870373506099,
56
+ "psnr": 30.557165145874023,
57
+ "lpips": 0.011660434305667877
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.0008535321103408933,
61
+ "psnr": 30.711021423339844,
62
+ "lpips": 0.011796060018241405
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.0009378630202263594,
66
+ "psnr": 30.307851791381836,
67
+ "lpips": 0.014937634579837322
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.0008990838541649282,
71
+ "psnr": 30.543855667114258,
72
+ "lpips": 0.015569162555038929
73
+ },
74
+ "overall": {
75
+ "mse": 0.0009021054720506072,
76
+ "psnr": 30.500322341918945,
77
+ "lpips": 0.012994368560612202
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000126_test__val_desert__1__000020/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6826f6e85f9dccedbd971b42f473856a77a9a67ec1d0a1bf7319eb6f78a00271
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000126_test__val_desert__1__000020/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "126",
4
+ "global_test_index": "126",
5
+ "dataset_index": "50",
6
+ "relative_path": "test/val_desert/1/000020.mp4",
7
+ "clip_id": "test/val_desert/1/000020",
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": 42000126,
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/1/000020_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": 0.001484862295910716,
51
+ "psnr": 29.569835662841797,
52
+ "lpips": 0.018319694325327873
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.0016191304894164205,
56
+ "psnr": 30.244709014892578,
57
+ "lpips": 0.015003266744315624
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.0008267026278190315,
61
+ "psnr": 30.85266876220703,
62
+ "lpips": 0.011609358713030815
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.0020510931499302387,
66
+ "psnr": 28.709508895874023,
67
+ "lpips": 0.027202578261494637
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.006186373997479677,
71
+ "psnr": 24.444780349731445,
72
+ "lpips": 0.04963857680559158
73
+ },
74
+ "overall": {
75
+ "mse": 0.0024336325004696846,
76
+ "psnr": 28.764301300048828,
77
+ "lpips": 0.02435469441115856
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000127_test__val_savanna__0__000036/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fdee9e328d93265037da10e8d3e3982086817d9eb2bc17d5cfa8ccfba05c791
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000127_test__val_savanna__0__000036/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "127",
4
+ "global_test_index": "127",
5
+ "dataset_index": "207",
6
+ "relative_path": "test/val_savanna/0/000036.mp4",
7
+ "clip_id": "test/val_savanna/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": 42000127,
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/0/000036_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": 0.006707283202558756,
51
+ "psnr": 22.234464645385742,
52
+ "lpips": 0.35190626978874207
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.0072944206185638905,
56
+ "psnr": 21.453563690185547,
57
+ "lpips": 0.39522361755371094
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.006893693003803492,
61
+ "psnr": 21.749900817871094,
62
+ "lpips": 0.3911628723144531
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.016896530985832214,
66
+ "psnr": 19.2712459564209,
67
+ "lpips": 0.4398539364337921
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.024989934638142586,
71
+ "psnr": 16.752212524414062,
72
+ "lpips": 0.5247270464897156
73
+ },
74
+ "overall": {
75
+ "mse": 0.01255637127906084,
76
+ "psnr": 20.292278289794922,
77
+ "lpips": 0.42057475447654724
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000128_test__val_iceplains__2__000034/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8569d946e4b4e565e3ed244e56acb556e0cef9f006ce0d7e2c2892eec7b32977
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000128_test__val_iceplains__2__000034/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "128",
4
+ "global_test_index": "128",
5
+ "dataset_index": "330",
6
+ "relative_path": "test/val_iceplains/2/000034.mp4",
7
+ "clip_id": "test/val_iceplains/2/000034",
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": 42000128,
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/2/000034_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": 0.0009549509268254042,
51
+ "psnr": 30.322065353393555,
52
+ "lpips": 0.01559457927942276
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.0011458741500973701,
56
+ "psnr": 30.461318969726562,
57
+ "lpips": 0.014609893783926964
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.016090119257569313,
61
+ "psnr": 24.57937240600586,
62
+ "lpips": 0.0829785168170929
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.054150525480508804,
66
+ "psnr": 13.15584945678711,
67
+ "lpips": 0.3769654333591461
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.09743870049715042,
71
+ "psnr": 10.951923370361328,
72
+ "lpips": 0.4895250201225281
73
+ },
74
+ "overall": {
75
+ "mse": 0.03395603224635124,
76
+ "psnr": 21.894105911254883,
77
+ "lpips": 0.1959347128868103
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000129_test__val_savanna__0__000001/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69faf235a6fcb88dd51eb9ef4e69cba4f0014ffce769971cd130de0c2c191f5f
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000129_test__val_savanna__0__000001/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "129",
4
+ "global_test_index": "129",
5
+ "dataset_index": "132",
6
+ "relative_path": "test/val_savanna/0/000001.mp4",
7
+ "clip_id": "test/val_savanna/0/000001",
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": 42000129,
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/0/000001_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": 0.0014532122295349836,
51
+ "psnr": 28.748018264770508,
52
+ "lpips": 0.021940717473626137
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.0019861303735524416,
56
+ "psnr": 28.03094482421875,
57
+ "lpips": 0.026773037388920784
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.011648589745163918,
61
+ "psnr": 21.0721378326416,
62
+ "lpips": 0.21085555851459503
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.014550996012985706,
66
+ "psnr": 18.718036651611328,
67
+ "lpips": 0.2853127717971802
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.004384123254567385,
71
+ "psnr": 24.046506881713867,
72
+ "lpips": 0.06934908777475357
73
+ },
74
+ "overall": {
75
+ "mse": 0.0068046110682189465,
76
+ "psnr": 24.12312889099121,
77
+ "lpips": 0.12284623831510544
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000130_test__val_plains__3__000021/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71a676e20e92ec2a9d98f7723ed85af85745718040ae6dee84c88cf42ad9e069
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000130_test__val_plains__3__000021/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "130",
4
+ "global_test_index": "130",
5
+ "dataset_index": "423",
6
+ "relative_path": "test/val_plains/3/000021.mp4",
7
+ "clip_id": "test/val_plains/3/000021",
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": 42000130,
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/3/000021_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": 0.00328049180097878,
51
+ "psnr": 25.106420516967773,
52
+ "lpips": 0.2332487553358078
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.004358980804681778,
56
+ "psnr": 23.907419204711914,
57
+ "lpips": 0.2951502799987793
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.004142299760133028,
61
+ "psnr": 23.862367630004883,
62
+ "lpips": 0.31737029552459717
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.014077514410018921,
66
+ "psnr": 21.26189422607422,
67
+ "lpips": 0.36391833424568176
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.047864340245723724,
71
+ "psnr": 14.42151165008545,
72
+ "lpips": 0.3694884777069092
73
+ },
74
+ "overall": {
75
+ "mse": 0.014744725078344345,
76
+ "psnr": 21.71192169189453,
77
+ "lpips": 0.31583523750305176
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000131_test__val_desert__2__000025/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:167a06b9b7fb92e02450c0d5dbe39fd4d91438a0aa65c461a3bea16b2f30ae7b
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000131_test__val_desert__2__000025/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "131",
4
+ "global_test_index": "131",
5
+ "dataset_index": "66",
6
+ "relative_path": "test/val_desert/2/000025.mp4",
7
+ "clip_id": "test/val_desert/2/000025",
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": 42000131,
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/2/000025_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": 0.009860005229711533,
51
+ "psnr": 20.345754623413086,
52
+ "lpips": 0.24561290442943573
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.017666202038526535,
56
+ "psnr": 17.62787437438965,
57
+ "lpips": 0.2884877026081085
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.01360994577407837,
61
+ "psnr": 18.91973876953125,
62
+ "lpips": 0.32690712809562683
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.01751602254807949,
66
+ "psnr": 17.899829864501953,
67
+ "lpips": 0.3800671696662903
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.015232359059154987,
71
+ "psnr": 18.476234436035156,
72
+ "lpips": 0.3669703006744385
73
+ },
74
+ "overall": {
75
+ "mse": 0.014776906929910183,
76
+ "psnr": 18.653884887695312,
77
+ "lpips": 0.3216090500354767
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000132_test__val_savanna__3__000009/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77521499d34f2ccb8a59a3528cae6186e007cd490d74ca024c5b59de51bc2ddd
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000132_test__val_savanna__3__000009/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "132",
4
+ "global_test_index": "132",
5
+ "dataset_index": "153",
6
+ "relative_path": "test/val_savanna/3/000009.mp4",
7
+ "clip_id": "test/val_savanna/3/000009",
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": 42000132,
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/3/000009_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": 0.022195372730493546,
51
+ "psnr": 18.302406311035156,
52
+ "lpips": 0.22228436172008514
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.02921193093061447,
56
+ "psnr": 16.188037872314453,
57
+ "lpips": 0.22970882058143616
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.016484905034303665,
61
+ "psnr": 18.224079132080078,
62
+ "lpips": 0.13724012672901154
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.0289585143327713,
66
+ "psnr": 16.264480590820312,
67
+ "lpips": 0.21746228635311127
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.0348900705575943,
71
+ "psnr": 14.894621849060059,
72
+ "lpips": 0.24207791686058044
73
+ },
74
+ "overall": {
75
+ "mse": 0.026348158717155457,
76
+ "psnr": 16.77472496032715,
77
+ "lpips": 0.20975470542907715
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000133_test__val_plains__0__000004/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed5273a841f9a78a1e776a1cd11a015294d9426b22d2c8700173dc58e3138e53
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000133_test__val_plains__0__000004/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "133",
4
+ "global_test_index": "133",
5
+ "dataset_index": "382",
6
+ "relative_path": "test/val_plains/0/000004.mp4",
7
+ "clip_id": "test/val_plains/0/000004",
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": 42000133,
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/0/000004_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": 0.0058406745083630085,
51
+ "psnr": 22.48453140258789,
52
+ "lpips": 0.2485457956790924
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.0039766086265444756,
56
+ "psnr": 24.035066604614258,
57
+ "lpips": 0.3241359293460846
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.004747073166072369,
61
+ "psnr": 23.44042205810547,
62
+ "lpips": 0.3394511938095093
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.006929060444235802,
66
+ "psnr": 21.747617721557617,
67
+ "lpips": 0.4062713384628296
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.0046330345794558525,
71
+ "psnr": 23.454448699951172,
72
+ "lpips": 0.365752637386322
73
+ },
74
+ "overall": {
75
+ "mse": 0.0052252900786697865,
76
+ "psnr": 23.03241729736328,
77
+ "lpips": 0.33683139085769653
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000134_test__val_savanna__2__000005/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9ddb27134e23718ba5dbf1e8c27b279dc56e29a7c582a097add0cad9f279234
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000134_test__val_savanna__2__000005/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "134",
4
+ "global_test_index": "134",
5
+ "dataset_index": "143",
6
+ "relative_path": "test/val_savanna/2/000005.mp4",
7
+ "clip_id": "test/val_savanna/2/000005",
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": 42000134,
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/2/000005_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": 0.014485334046185017,
51
+ "psnr": 18.5859375,
52
+ "lpips": 0.3390156626701355
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.028073078021407127,
56
+ "psnr": 16.182510375976562,
57
+ "lpips": 0.41970911622047424
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.012548437342047691,
61
+ "psnr": 19.95612144470215,
62
+ "lpips": 0.4414370656013489
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.026289591565728188,
66
+ "psnr": 16.48810577392578,
67
+ "lpips": 0.4970936179161072
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.06847763806581497,
71
+ "psnr": 12.170432090759277,
72
+ "lpips": 0.547330915927887
73
+ },
74
+ "overall": {
75
+ "mse": 0.029974816367030144,
76
+ "psnr": 16.676620483398438,
77
+ "lpips": 0.4489172697067261
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000135_test__val_iceplains__1__000040/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2680df7858b3096a2890248ab32c69f8025ea90fa036d14334fabc1b0ec3725
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000135_test__val_iceplains__1__000040/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "135",
4
+ "global_test_index": "135",
5
+ "dataset_index": "348",
6
+ "relative_path": "test/val_iceplains/1/000040.mp4",
7
+ "clip_id": "test/val_iceplains/1/000040",
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": 42000135,
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/1/000040_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": 0.021408552303910255,
51
+ "psnr": 17.150901794433594,
52
+ "lpips": 0.20184822380542755
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.019069790840148926,
56
+ "psnr": 17.861360549926758,
57
+ "lpips": 0.18908923864364624
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.02559494972229004,
61
+ "psnr": 16.50379180908203,
62
+ "lpips": 0.2058320939540863
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.022684333845973015,
66
+ "psnr": 17.9940242767334,
67
+ "lpips": 0.26573479175567627
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.020618490874767303,
71
+ "psnr": 17.067407608032227,
72
+ "lpips": 0.16639107465744019
73
+ },
74
+ "overall": {
75
+ "mse": 0.021875223144888878,
76
+ "psnr": 17.31549835205078,
77
+ "lpips": 0.20577910542488098
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000136_test__val_plains__2__000005/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8083260fcb35c577f1aa1472b6ff3e39f8f6f1814b73fb9bd10f65a2e14d01eb
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000136_test__val_plains__2__000005/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "136",
4
+ "global_test_index": "136",
5
+ "dataset_index": "384",
6
+ "relative_path": "test/val_plains/2/000005.mp4",
7
+ "clip_id": "test/val_plains/2/000005",
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": 42000136,
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/2/000005_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": 0.004267084412276745,
51
+ "psnr": 23.781980514526367,
52
+ "lpips": 0.3505207896232605
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.003726803930476308,
56
+ "psnr": 24.324541091918945,
57
+ "lpips": 0.36018940806388855
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.06266765296459198,
61
+ "psnr": 14.357873916625977,
62
+ "lpips": 0.45021724700927734
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.06701135635375977,
66
+ "psnr": 12.021452903747559,
67
+ "lpips": 0.38467979431152344
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.0655532106757164,
71
+ "psnr": 11.978496551513672,
72
+ "lpips": 0.13128697872161865
73
+ },
74
+ "overall": {
75
+ "mse": 0.04064522683620453,
76
+ "psnr": 17.29286766052246,
77
+ "lpips": 0.33537882566452026
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000137_test__val_desert__0__000012/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3bff6aca606d89cd8d1dced7f87dc91cb48682d93457c57bc43f07e1ccdd9752
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000137_test__val_desert__0__000012/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "137",
4
+ "global_test_index": "137",
5
+ "dataset_index": "30",
6
+ "relative_path": "test/val_desert/0/000012.mp4",
7
+ "clip_id": "test/val_desert/0/000012",
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": 42000137,
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/0/000012_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": 0.0007693073130212724,
51
+ "psnr": 31.15523910522461,
52
+ "lpips": 0.009894480928778648
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.0007668848847970366,
56
+ "psnr": 31.1663875579834,
57
+ "lpips": 0.010341694578528404
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.0012762021506205201,
61
+ "psnr": 29.499380111694336,
62
+ "lpips": 0.022373294457793236
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.0011693239212036133,
66
+ "psnr": 30.59243392944336,
67
+ "lpips": 0.013964185491204262
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.0008222288452088833,
71
+ "psnr": 30.890939712524414,
72
+ "lpips": 0.011983408592641354
73
+ },
74
+ "overall": {
75
+ "mse": 0.0009607894462533295,
76
+ "psnr": 30.66087532043457,
77
+ "lpips": 0.013711412437260151
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000138_test__val_savanna__2__000034/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b95615973f714b899d0e3ab551925051eee910e5ea9cbd73b06538e6015ff970
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000138_test__val_savanna__2__000034/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "138",
4
+ "global_test_index": "138",
5
+ "dataset_index": "203",
6
+ "relative_path": "test/val_savanna/2/000034.mp4",
7
+ "clip_id": "test/val_savanna/2/000034",
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": 42000138,
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/2/000034_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": 0.0019806530326604843,
51
+ "psnr": 27.982154846191406,
52
+ "lpips": 0.019615111872553825
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.0014540873235091567,
56
+ "psnr": 29.562135696411133,
57
+ "lpips": 0.013446008786559105
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.0020869376603513956,
61
+ "psnr": 28.869915008544922,
62
+ "lpips": 0.018795687705278397
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.003144179005175829,
66
+ "psnr": 27.136390686035156,
67
+ "lpips": 0.02872839942574501
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.010119260288774967,
71
+ "psnr": 23.284276962280273,
72
+ "lpips": 0.05465582758188248
73
+ },
74
+ "overall": {
75
+ "mse": 0.0037570232525467873,
76
+ "psnr": 27.366973876953125,
77
+ "lpips": 0.027048209682106972
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000139_test__val_savanna__0__000022/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c52531befb8c9ccd99079b3149f2c120e57340e813da97b82860f01332e2095
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000139_test__val_savanna__0__000022/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "139",
4
+ "global_test_index": "139",
5
+ "dataset_index": "179",
6
+ "relative_path": "test/val_savanna/0/000022.mp4",
7
+ "clip_id": "test/val_savanna/0/000022",
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": 42000139,
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/0/000022_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": 0.0013348344946280122,
51
+ "psnr": 28.83977699279785,
52
+ "lpips": 0.13142722845077515
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.002187905367463827,
56
+ "psnr": 26.7196044921875,
57
+ "lpips": 0.14662164449691772
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.005070777144283056,
61
+ "psnr": 23.106117248535156,
62
+ "lpips": 0.12211448699235916
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.007739753928035498,
66
+ "psnr": 21.19198989868164,
67
+ "lpips": 0.11173173040151596
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.008362587541341782,
71
+ "psnr": 20.861509323120117,
72
+ "lpips": 0.12095114588737488
73
+ },
74
+ "overall": {
75
+ "mse": 0.00493917241692543,
76
+ "psnr": 24.143798828125,
77
+ "lpips": 0.1265692412853241
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000140_test__val_iceplains__2__000003/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6783169bf6064121234c12670e0b0870421d905b33d2674a3a6c34c7c6e1be2d
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000140_test__val_iceplains__2__000003/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "140",
4
+ "global_test_index": "140",
5
+ "dataset_index": "253",
6
+ "relative_path": "test/val_iceplains/2/000003.mp4",
7
+ "clip_id": "test/val_iceplains/2/000003",
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": 42000140,
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/2/000003_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": 0.008791802451014519,
51
+ "psnr": 21.091079711914062,
52
+ "lpips": 0.16836988925933838
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.024468516930937767,
56
+ "psnr": 16.746850967407227,
57
+ "lpips": 0.2349162995815277
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.07874393463134766,
61
+ "psnr": 11.419788360595703,
62
+ "lpips": 0.48818641901016235
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.04718368873000145,
66
+ "psnr": 14.619848251342773,
67
+ "lpips": 0.3406621813774109
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.04312962666153908,
71
+ "psnr": 14.540613174438477,
72
+ "lpips": 0.37893441319465637
73
+ },
74
+ "overall": {
75
+ "mse": 0.040463514626026154,
76
+ "psnr": 15.683636665344238,
77
+ "lpips": 0.3222138583660126
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000141_test__val_iceplains__2__000016/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:450fe56fc077c2a14687b5e6c6cfe2c3f2835c92d5cde394a1c57958c910d804
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000141_test__val_iceplains__2__000016/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "141",
4
+ "global_test_index": "141",
5
+ "dataset_index": "289",
6
+ "relative_path": "test/val_iceplains/2/000016.mp4",
7
+ "clip_id": "test/val_iceplains/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": 42000141,
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/2/000016_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": 0.0007456709281541407,
51
+ "psnr": 31.282577514648438,
52
+ "lpips": 0.009861068800091743
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.004965067841112614,
56
+ "psnr": 27.782520294189453,
57
+ "lpips": 0.040066178888082504
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.003117287065833807,
61
+ "psnr": 26.664409637451172,
62
+ "lpips": 0.03511936590075493
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.002085950458422303,
66
+ "psnr": 27.268447875976562,
67
+ "lpips": 0.02791164070367813
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.00514613464474678,
71
+ "psnr": 24.119783401489258,
72
+ "lpips": 0.060876477509737015
73
+ },
74
+ "overall": {
75
+ "mse": 0.0032120223622769117,
76
+ "psnr": 27.42354965209961,
77
+ "lpips": 0.03476694971323013
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000142_test__val_plains__2__000020/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:90517f0283146d055bf1d91e5d082954ae0c9426acbadb6f8ea33a4beb26f496
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000142_test__val_plains__2__000020/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "142",
4
+ "global_test_index": "142",
5
+ "dataset_index": "422",
6
+ "relative_path": "test/val_plains/2/000020.mp4",
7
+ "clip_id": "test/val_plains/2/000020",
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": 42000142,
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/2/000020_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": 0.004075692035257816,
51
+ "psnr": 24.302370071411133,
52
+ "lpips": 0.05757622420787811
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.008997473865747452,
56
+ "psnr": 20.75374984741211,
57
+ "lpips": 0.22283805906772614
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.007085732650011778,
61
+ "psnr": 22.535398483276367,
62
+ "lpips": 0.15993793308734894
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.0048640938475728035,
66
+ "psnr": 23.634279251098633,
67
+ "lpips": 0.09337618947029114
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.004075458738952875,
71
+ "psnr": 25.187114715576172,
72
+ "lpips": 0.05961039662361145
73
+ },
74
+ "overall": {
75
+ "mse": 0.005819690879434347,
76
+ "psnr": 23.282583236694336,
77
+ "lpips": 0.11866774410009384
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000143_test__val_desert__0__000008/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f529641f2d466855a6942680d721821c54491640031a8ee659fcfe756c484665
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000143_test__val_desert__0__000008/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "143",
4
+ "global_test_index": "143",
5
+ "dataset_index": "20",
6
+ "relative_path": "test/val_desert/0/000008.mp4",
7
+ "clip_id": "test/val_desert/0/000008",
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": 42000143,
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/0/000008_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": 0.0020590005442500114,
51
+ "psnr": 29.03612518310547,
52
+ "lpips": 0.030739037320017815
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.00984747800976038,
56
+ "psnr": 20.688880920410156,
57
+ "lpips": 0.17833548784255981
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.009776131249964237,
61
+ "psnr": 20.306804656982422,
62
+ "lpips": 0.2711389660835266
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.011499784886837006,
66
+ "psnr": 19.532052993774414,
67
+ "lpips": 0.17814262211322784
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.0026700920425355434,
71
+ "psnr": 28.21206283569336,
72
+ "lpips": 0.02954581193625927
73
+ },
74
+ "overall": {
75
+ "mse": 0.00717049790546298,
76
+ "psnr": 23.555185317993164,
77
+ "lpips": 0.13758039474487305
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000144_test__val_savanna__0__000000/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5c5d13766bf22f2f2970711e67ad72ee951b42d2e7012037e5ebc93b1d795fa
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000144_test__val_savanna__0__000000/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "144",
4
+ "global_test_index": "144",
5
+ "dataset_index": "130",
6
+ "relative_path": "test/val_savanna/0/000000.mp4",
7
+ "clip_id": "test/val_savanna/0/000000",
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": 42000144,
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/0/000000_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": 0.011638644151389599,
51
+ "psnr": 19.769290924072266,
52
+ "lpips": 0.16591542959213257
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.019459480419754982,
56
+ "psnr": 17.360668182373047,
57
+ "lpips": 0.1535501331090927
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.018692554906010628,
61
+ "psnr": 18.75762176513672,
62
+ "lpips": 0.13838034868240356
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.031162414699792862,
66
+ "psnr": 16.123083114624023,
67
+ "lpips": 0.17314361035823822
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.045554712414741516,
71
+ "psnr": 13.491724014282227,
72
+ "lpips": 0.23765979707241058
73
+ },
74
+ "overall": {
75
+ "mse": 0.025301562622189522,
76
+ "psnr": 17.10047721862793,
77
+ "lpips": 0.17372985184192657
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000145_test__val_iceplains__1__000020/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6820643762011af5b6a06a11eea8b1365a5c25b611cc99e84cb14c24928a1b61
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000145_test__val_iceplains__1__000020/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "145",
4
+ "global_test_index": "145",
5
+ "dataset_index": "296",
6
+ "relative_path": "test/val_iceplains/1/000020.mp4",
7
+ "clip_id": "test/val_iceplains/1/000020",
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": 42000145,
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/1/000020_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": 0.016957759857177734,
51
+ "psnr": 17.81981658935547,
52
+ "lpips": 0.18485485017299652
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.019970489665865898,
56
+ "psnr": 17.228933334350586,
57
+ "lpips": 0.1870379000902176
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.03969457000494003,
61
+ "psnr": 14.1196928024292,
62
+ "lpips": 0.29053473472595215
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.0758398100733757,
66
+ "psnr": 11.391704559326172,
67
+ "lpips": 0.43325960636138916
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.07288557291030884,
71
+ "psnr": 11.589159965515137,
72
+ "lpips": 0.3946675956249237
73
+ },
74
+ "overall": {
75
+ "mse": 0.04506963863968849,
76
+ "psnr": 14.429861068725586,
77
+ "lpips": 0.2980709671974182
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000146_test__val_iceplains__3__000013/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8de28dc9d9c824d249456abdc0dec47326b992e095aef05b5552f9ce5a55b0e6
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000146_test__val_iceplains__3__000013/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "146",
4
+ "global_test_index": "146",
5
+ "dataset_index": "278",
6
+ "relative_path": "test/val_iceplains/3/000013.mp4",
7
+ "clip_id": "test/val_iceplains/3/000013",
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": 42000146,
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/3/000013_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": 0.014167653396725655,
51
+ "psnr": 20.767545700073242,
52
+ "lpips": 0.08324377238750458
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.04269548878073692,
56
+ "psnr": 14.481380462646484,
57
+ "lpips": 0.3006751537322998
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.09141496568918228,
61
+ "psnr": 10.570194244384766,
62
+ "lpips": 0.4685143530368805
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.03654433414340019,
66
+ "psnr": 22.187768936157227,
67
+ "lpips": 0.2141980677843094
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.004716005176305771,
71
+ "psnr": 25.456411361694336,
72
+ "lpips": 0.0629841685295105
73
+ },
74
+ "overall": {
75
+ "mse": 0.037907689809799194,
76
+ "psnr": 18.692659378051758,
77
+ "lpips": 0.2259231060743332
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000147_test__val_iceplains__0__000009/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a18ade880dcca3c327c548507b42238f9374de186f684193d09cb3392f8daf0b
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000147_test__val_iceplains__0__000009/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "147",
4
+ "global_test_index": "147",
5
+ "dataset_index": "269",
6
+ "relative_path": "test/val_iceplains/0/000009.mp4",
7
+ "clip_id": "test/val_iceplains/0/000009",
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": 42000147,
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/0/000009_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": 0.007628425490111113,
51
+ "psnr": 21.40215492248535,
52
+ "lpips": 0.09683271497488022
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.01973353885114193,
56
+ "psnr": 17.855030059814453,
57
+ "lpips": 0.19308441877365112
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.014538093470036983,
61
+ "psnr": 19.753524780273438,
62
+ "lpips": 0.10630655288696289
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.008386390283703804,
66
+ "psnr": 22.098949432373047,
67
+ "lpips": 0.059912487864494324
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.006873622070997953,
71
+ "psnr": 22.614782333374023,
72
+ "lpips": 0.06482533365488052
73
+ },
74
+ "overall": {
75
+ "mse": 0.011432014405727386,
76
+ "psnr": 20.744888305664062,
77
+ "lpips": 0.10419230908155441
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000148_test__val_iceplains__0__000028/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b4e6d1b176a838199f3c95424da1e63bc848416fc7235a6f2bec963243de652
3
+ size 6742
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000148_test__val_iceplains__0__000028/metrics.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "case": {
3
+ "manifest_index": "148",
4
+ "global_test_index": "148",
5
+ "dataset_index": "314",
6
+ "relative_path": "test/val_iceplains/0/000028.mp4",
7
+ "clip_id": "test/val_iceplains/0/000028",
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": 42000148,
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/0/000028_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": 0.0007975163171067834,
51
+ "psnr": 31.004505157470703,
52
+ "lpips": 0.010168538428843021
53
+ },
54
+ "frames_100_200": {
55
+ "mse": 0.0007841441547498107,
56
+ "psnr": 31.065732955932617,
57
+ "lpips": 0.009760734625160694
58
+ },
59
+ "frames_200_300": {
60
+ "mse": 0.0022047495003789663,
61
+ "psnr": 29.552980422973633,
62
+ "lpips": 0.021267982199788094
63
+ },
64
+ "frames_300_400": {
65
+ "mse": 0.037726834416389465,
66
+ "psnr": 14.477999687194824,
67
+ "lpips": 0.2727294862270355
68
+ },
69
+ "frames_400_500": {
70
+ "mse": 0.056888509541749954,
71
+ "psnr": 12.888333320617676,
72
+ "lpips": 0.3455289602279663
73
+ },
74
+ "overall": {
75
+ "mse": 0.01968034729361534,
76
+ "psnr": 23.797908782958984,
77
+ "lpips": 0.131891131401062
78
+ }
79
+ }
80
+ }
results/worldmem-baseline-evals/decmem/decmem_minecraft_300cases_seed42_job84976/cases/g000149_test__val_iceplains__0__000014/frame_metrics.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:48c5742d19eb37a96a6590ed2aacd1777f70d5c589003b3edca7ba362e7716ca
3
+ size 6742