{
  "dataset_name": "LTMI_Multimode",
  "normalizer_dataset_name": "LTMI_Multimode",
  "field_names": [
    "a",
    "u"
  ],
  "field_channels": {
    "a": [
      "sigma_t_map_1",
      "sigma_t_map_2"
    ],
    "u": [
      "u_channel0",
      "u2_channel0"
    ]
  },
  "field_slices": {
    "a": [
      0,
      2
    ],
    "u": [
      2,
      4
    ]
  },
  "shape": [
    64,
    64
  ],
  "num_samples": 100000,
  "source_count": 1173621,
  "selection": "first 100000 samples from full LTMI_Multimode HF conversion",
  "rgb_response_policy": "kept channel 0 only",
  "stats": {
    "mean": [
      2.516578340749906,
      2.518172909586568,
      0.3230067392023618,
      0.46580631291939256
    ],
    "std": [
      1.7554190007156716,
      1.7551658526673308,
      0.1308156101844342,
      0.45600807851895414
    ],
    "min": [
      0.009999999776482582,
      0.009999999776482582,
      0.0,
      0.0
    ],
    "max": [
      5.0,
      5.0,
      0.6829288005828857,
      2.611924886703491
    ]
  },
  "normalization": "(x - mean) * (0.5 / std)",
  "source": "data/PBench/ltmi_training.zip",
  "__version__": "2.0"
}