File size: 10,861 Bytes
9bc9286 8320d87 9bc9286 8320d87 9bc9286 8320d87 9bc9286 6448a8a 9bc9286 e1220cb 9bc9286 e157335 9bc9286 |
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 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 |
data:
format: zarr
resolution: n320
frequency: 6h
timestep: 6h
forcing:
- cos_latitude
- cos_longitude
- sin_latitude
- sin_longitude
- cos_julian_day
- cos_local_time
- sin_julian_day
- sin_local_time
- insolation
- lsm
- sdor
- slor
- z
diagnostic:
- tp
- cp
- sf
- tcc
- hcc
- lcc
- mcc
- ro
- ssrd
- strd
- 100u
- 100v
remapped: null
normalizer:
default: mean-std
remap:
cp: tp
sf: tp
std:
- tp
- cp
- sf
- ro
- tcw
- ssrd
- q_50
- q_100
- q_150
- q_200
- q_250
- q_300
- q_400
- q_500
- q_600
- q_700
- q_850
- q_925
- q_1000
min-max: null
max:
- sdor
- slor
- z
none:
- cos_latitude
- cos_longitude
- sin_latitude
- sin_longitude
- cos_julian_day
- cos_local_time
- sin_julian_day
- sin_local_time
- insolation
- lsm
- tcc
- mcc
- hcc
- lcc
- swvl1
- swvl2
imputer:
default: none
remapper:
default: none
processors:
normalizer:
_target_: anemoi.models.preprocessing.normalizer.InputNormalizer
_convert_: all
config:
default: mean-std
remap:
cp: tp
sf: tp
std:
- tp
- cp
- sf
- ro
- tcw
- ssrd
- q_50
- q_100
- q_150
- q_200
- q_250
- q_300
- q_400
- q_500
- q_600
- q_700
- q_850
- q_925
- q_1000
min-max: null
max:
- sdor
- slor
- z
none:
- cos_latitude
- cos_longitude
- sin_latitude
- sin_longitude
- cos_julian_day
- cos_local_time
- sin_julian_day
- sin_local_time
- insolation
- lsm
- tcc
- mcc
- hcc
- lcc
- swvl1
- swvl2
num_features: 115
dataloader:
prefetch_factor: 2
pin_memory: True
read_group_size: 4
num_workers:
training: 4
validation: 4
test: 8
predict: 8
batch_size:
training: 1
validation: 1
test: 4
predict: 4
limit_batches:
training: null
validation: 10
test: 20
predict: 20
dataset: ${hardware.paths.data}/${hardware.files.dataset}
land_dataset: ${hardware.paths.data}/${hardware.files.dataset_land}
land_variables: [100u, 100v, swvl1, swvl2, stl1, stl2, tcc, lcc, mcc, hcc, sf, ro, strd, ssrd]
training:
dataset:
- dataset: ${dataloader.dataset}
start: null
end: 2022
frequency: ${data.frequency}
drop: []
- dataset: ${dataloader.land_dataset}
start: null
end: 2022
frequency: ${data.frequency}
select: ${dataloader.land_variables}
start: null
end: 2022
drop: []
validation:
dataset:
- dataset: ${dataloader.dataset}
start: 2022
end: 2022
frequency: ${data.frequency}
drop: []
- dataset: ${dataloader.land_dataset}
start: 2022
end: 2022
frequency: ${data.frequency}
select: ${dataloader.land_variables}
start: 2022
end: 2022
drop: []
validation_rollout: 1
diagnostics:
plot:
asynchronous: False
datashader: True
frequency:
batch: 750
epoch: 10
parameters: [tp]
sample_idx: 0
callbacks:
- _target_: anemoi.training.diagnostics.callbacks.plot.PlotLoss
parameter_groups:
moisture: [tp, cp, tcw]
sfc_wind: [10u, 10v]
- _target_: anemoi.training.diagnostics.callbacks.plot.PlotSample
sample_idx: 0
per_sample: 6
parameters: [tp]
accumulation_levels_plot: [0, 0.05, 0.1, 0.25, 0.5, 1, 1.5, 2, 3, 4, 5, 6, 7, 100]
cmap_accumulation:
- "#ffffff"
- "#04e9e7"
- "#019ff4"
- "#0300f4"
- "#02fd02"
- "#01c501"
- "#008e00"
- "#fdf802"
- "#e5bc00"
- "#fd9500"
- "#fd0000"
- "#d40000"
- "#bc0000"
- "#f800fd"
precip_and_related_fields: [tp, cp]
enabled: True
scatter: False
mode: asyncio
callbacks: {}
benchmark_profiler:
memory:
enabled: True
steps: 5
warmup: 2
extra_plots: False
trace_rank0_only: False
time:
enabled: True
verbose: False
speed:
enabled: True
system:
enabled: True
model_summary:
enabled: True
snapshot:
enabled: True
steps: 4
warmup: 0
debug:
anomaly_detection: False
profiler: False
enable_checkpointing: True
checkpoint:
every_n_minutes:
save_frequency: 30
num_models_saved: 3
every_n_epochs:
save_frequency: 1
num_models_saved: 3
every_n_train_steps:
save_frequency: null
num_models_saved: 0
log:
wandb:
enabled: False
tensorboard:
enabled: False
mlflow:
enabled: False
interval: 100
enable_progress_bar: True
print_memory_summary: False
hardware:
paths:
data: ${oc.decode:${oc.env:DATASETS_PATH}}
output: ${oc.decode:${oc.env:OUTPUT_DIR}}
logs:
base: ${hardware.paths.output}/logs
wandb: ${hardware.paths.output}/logs/wandb
mlflow: ${hardware.paths.output}/logs/mlflow
tensorboard: ${hardware.paths.output}/logs/tensorboard
checkpoints: ${hardware.paths.output}/checkpoint
plots: ${hardware.paths.output}/plots
profiler: ${hardware.paths.output}/profiler
graph: ${hardware.paths.output}/graphs
files:
dataset: aifs-ea-an-oper-0001-mars-n320-1979-2022-6h-v6.zarr
dataset_land: aifs-ea-an-oper-0001-mars-n320-1979-2023-6h-v1-land.zarr
graph: graph_enc_proc_dec_n320.pt
checkpoint:
every_n_epochs: aifs-by_epoch-epoch_{epoch:03d}-val_wmse_{val_wmse:.3e}
every_n_train_steps: aifs-by_step-epoch_{epoch:03d}-step_{step:06d}
every_n_minutes: aifs-by_time-epoch_{epoch:03d}-step_{step:06d}
warm_start: null
accelerator: auto
num_gpus_per_node: 4
num_nodes: 16
num_gpus_per_model: 4
graph:
overwrite: True
data: data
hidden: hidden
nodes:
data:
node_builder:
_target_: anemoi.graphs.nodes.ZarrDatasetNodes
dataset: ${dataloader.dataset}
attributes:
area_weight:
_target_: anemoi.graphs.nodes.attributes.AreaWeights
norm: unit-max
hidden:
node_builder:
_target_: anemoi.graphs.nodes.ReducedGaussianGridNodes
grid: o96
edges:
- source_name: data
target_name: hidden
edge_builder:
_target_: anemoi.graphs.edges.CutOffEdges
cutoff_factor: 0.6
attributes:
edge_length:
_target_: anemoi.graphs.edges.attributes.EdgeLength
norm: unit-std
edge_dirs:
_target_: anemoi.graphs.edges.attributes.EdgeDirection
norm: unit-std
- source_name: hidden
target_name: data
edge_builder:
_target_: anemoi.graphs.edges.KNNEdges
num_nearest_neighbours: 3
attributes:
edge_length:
_target_: anemoi.graphs.edges.attributes.EdgeLength
norm: unit-std
edge_dirs:
_target_: anemoi.graphs.edges.attributes.EdgeDirection
norm: unit-std
model:
activation: GELU
num_channels: 1024
model:
_target_: anemoi.models.models.encoder_processor_decoder.AnemoiModelEncProcDec
processor:
_target_: anemoi.models.layers.processor.TransformerProcessor
_convert_: all
activation: GELU
num_layers: 16
num_chunks: 2
mlp_hidden_ratio: 4
num_heads: 16
window_size: 1120
dropout_p: 0
encoder:
_target_: anemoi.models.layers.mapper.GraphTransformerForwardMapper
_convert_: all
trainable_size: 8
sub_graph_edge_attributes: [edge_length, edge_dirs]
activation: GELU
num_chunks: 1
mlp_hidden_ratio: 4
num_heads: 16
decoder:
_target_: anemoi.models.layers.mapper.GraphTransformerBackwardMapper
_convert_: all
trainable_size: 8
sub_graph_edge_attributes: [edge_length, edge_dirs]
activation: GELU
num_chunks: 1
mlp_hidden_ratio: 4
num_heads: 16
trainable_parameters:
data: 8
hidden: 8
data2hidden: 8
hidden2data: 8
attributes:
edges: [edge_length, edge_dirs]
nodes: []
node_loss_weight: area_weight
bounding:
- _target_: anemoi.models.layers.bounding.ReluBounding
variables:
- tp
- ro
- tcw
- ssrd
- q_50
- q_100
- q_150
- q_200
- q_250
- q_300
- q_400
- q_500
- q_600
- q_700
- q_850
- q_925
- q_1000
- _target_: anemoi.models.layers.bounding.HardtanhBounding
variables: [tcc, swvl1, swvl2]
min_val: 0
max_val: 1
- _target_: anemoi.models.layers.bounding.FractionBounding
variables: [cp, sf]
min_val: 0
max_val: 1
total_var: tp
- _target_: anemoi.models.layers.bounding.FractionBounding
variables: [lcc, mcc, hcc]
min_val: 0
max_val: 1
total_var: tcc
training:
run_id: null
fork_run_id: null
load_weights_only: null
deterministic: False
precision: 16-mixed
multistep_input: 2
accum_grad_batches: 1
num_sanity_val_steps: 6
gradient_clip:
val: 32
algorithm: value
swa:
enabled: False
lr: 0.0001
zero_optimizer: False
training_loss:
_target_: anemoi.training.losses.mse.WeightedMSELoss
scalars: [variable, loss_weights_mask]
ignore_nans: False
loss_gradient_scaling: False
validation_metrics:
- _target_: anemoi.training.losses.mse.WeightedMSELoss
scalars: []
ignore_nans: True
rollout:
start: 1
epoch_increment: 0
max: 1
max_epochs: null
max_steps: 260000
lr:
rate: 0.00003125
iterations: 260000
min: 3.0e-7
variable_loss_scaling:
default: 1
pl:
q: 0.6
t: 6
u: 0.8
v: 0.5
w: 0.001
z: 12
sfc:
sp: 10
10u: 0.5
10v: 0.5
100u: 0.1
100v: 0.1
2d: 0.5
tp: 0.025
cp: 0.0025
ro: 0.005
sf: 0.025
tcc: 0.1
mcc: 0.1
lcc: 0.1
hcc: 0.1
swvl2: 200
swvl1: 100
stl2: 10
stl1: 1
ssrd: 0.05
strd: 0.1
metrics: [z_500, t_850, u_850, v_850]
pressure_level_scaler:
_target_: anemoi.training.data.scaling.ReluPressureLevelScaler
minimum: 0.2
slope: 0.001 |