File size: 2,133 Bytes
986404c 5b3329b 986404c 5b3329b 986404c 5b3329b 986404c 5b3329b 986404c 5b3329b 986404c 5b3329b 986404c | 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 89 90 91 92 93 94 95 96 97 | {
"model_name": "CorrDiff",
"model_type": "corrdiff",
"architectures": [
"CorrDiff",
"RegressionUNet",
"ResidualDenoiser"
],
"framework": "PyTorch",
"domain": "atmosphere",
"task": "probabilistic-weather-downscaling",
"implementation": {
"entry_point": "model/corrdiff.py",
"scope": "conditional-regression mean prediction followed by residual EDM ensemble generation for kilometer-scale atmospheric downscaling",
"train_script": "scripts/train.py",
"inference_script": "scripts/inference.py",
"evaluation_script": "scripts/result.py",
"synthetic_data_script": "scripts/fake_data.py"
},
"architecture": {
"family": "conditional regression plus residual corrective diffusion",
"in_channels": 12,
"out_channels": 4,
"base_channels": 8,
"feature_size": 28,
"output_size": 448,
"sigma_data": 0.5,
"regression_component": "RegressionUNet",
"diffusion_component": "ResidualDenoiser",
"diffusion_preconditioning": "EDM",
"noise_schedule": "Karras",
"sampling": {
"ensemble_size": 2,
"steps": 3,
"sigma_min": 0.002,
"sigma_max": 5.0,
"rho": 7.0,
"solver": "heun"
}
},
"data": {
"datasets": [
"ERA5",
"CWA-WRF"
],
"protocol": "corrdiff_npz_v1",
"format": "NPZ",
"default_file": "data/corrdiff.npz",
"input_key": "input",
"input_shape": [
"N",
12,
36,
36
],
"target_key": "target",
"target_shape": [
"N",
4,
448,
448
],
"input_variables": [
"tcwv",
"t2m",
"u10m",
"v10m",
"t500",
"z500",
"u500",
"v500",
"t850",
"z850",
"u850",
"v850"
],
"target_variables": [
"t2m",
"u10m",
"v10m",
"maximum_radar_reflectivity"
],
"required_metadata": [
"protocol",
"data_source"
]
},
"configuration_sources": [
"conf/config.yaml",
"model/corrdiff.py",
"scripts/fake_data.py",
"scripts/train.py",
"scripts/inference.py",
"scripts/result.py"
]
}
|