Hugging Face's logo

gatilin
/
mkb_weather_encoder_decoder

PyTorch
Safetensors
weather
multimodal
polaris
component-extraction
Model card Files
xet
 Community
mkb_weather_encoder_decoder
File size: 1,934 Bytes
95c892c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
{
  "strict_load": true,
  "source_tensor_bytes_equal": true,
  "checks": [
    {
      "name": "strict_weather_components",
      "passed": true,
      "tensors": 299
    },
    {
      "name": "decoder_encoder_reference",
      "passed": true
    },
    {
      "name": "decoder_cold_cache",
      "passed": true
    },
    {
      "name": "weather_encode_native",
      "passed": true,
      "shape": [
        1,
        400,
        4096
      ]
    },
    {
      "name": "weather_identity_projector",
      "passed": true
    },
    {
      "name": "encoder_decoder_cache",
      "passed": true
    },
    {
      "name": "weather_decode_native",
      "passed": true,
      "shape": [
        1,
        70,
        120,
        120
      ],
      "dtype": "torch.float32"
    },
    {
      "name": "native_float32_residual",
      "passed": true
    },
    {
      "name": "weather_source_bytes_exact",
      "passed": true
    }
  ],
  "smoke": {
    "passed": true,
    "device": "cpu",
    "dtype": "float32",
    "geometry_override": [
      120,
      120
    ],
    "original_geometry": [
      721,
      1440
    ],
    "fields": "synthetic standardized inputs; zero static constants",
    "decoder_hidden_states": "synthetic zeros, not MKB language outputs"
  },
  "elapsed_seconds": 13.903923988342285,
  "limitations": [
    "No 721x1440 full-resolution forecast, ERA5 evaluation, physical-value de-normalization, or multi-day rollout.",
    "Real forecasting additionally requires original ERA5 variable ordering, normalization/static-field data and the MKB language backbone.",
    "No medical SAM3 or biological sequence components included."
  ],
  "runtime_versions": {
    "torch": "2.14.1",
    "torchvision": "0.29.1",
    "transformers": "5.20.1",
    "diffusers": "0.41.0",
    "safetensors": "0.8.0",
    "accelerate": "1.15.0",
    "numpy": "2.5.4",
    "einops": "0.8.2",
    "pandas": "3.0.6"
  }
}